# CommerceClarity > AI Agents for retail catalog operations. CommerceClarity runs AI agents that execute product catalog operations for enterprise retailers and brands: enrichment, sourcing, localization, channel publishing, compliance. Every page of CommerceClarity below, in full. A single page is also available on its own: append `.md` to its path. The index is at https://commerceclarity.com/llms.txt. --- # Your catalog, run by AI agents. > CommerceClarity is the AI agent platform for retail catalog operations, founded in Milan in 2024 and working from Milan and Barcelona. Its agents ingest, structure, enrich and validate product data against each retailer’s own rules, across eight areas from catalog foundation to agentic commerce. It is sold as an outcome rather than a licence: dedicated Retail Engineers bring the agents into production category by category and channel by channel. CommerceClarity lets you automate, scale, and take full control of your catalog operations - from data enrichment and localization to channel adaptation, compliance and AI discovery. [Book a demo](/book-a-demo) ![The CommerceClarity grid: a catalog of products with their attributes as columns, values the agent flagged in amber, and a status bar counting issues and missing values.](https://cdn.sanity.io/images/xd4hrbt2/production/fc4e47cd302ace22f34282f338c98e537d9aed7a-2880x1920.webp?auto=format) **Wall caption:** Trusted by 50+ enterprise retailers and brands ## Wall tiles - Prada - Nestlé · [/customers/purina](/customers/purina) - Cisalfa Sport · [/customers/cisalfa-group](/customers/cisalfa-group) - Arcaplanet - Intersport - Dr. Max - STIGA · [/customers/stiga](/customers/stiga) - Prénatal - Atida eFarma - Gruppo Hippocrates - Yeppon · [/customers/yeppon](/customers/yeppon) - Scalapay - Next Hardware & Software · [/customers/next-hardware-software](/customers/next-hardware-software) - Caddy’s · [/customers/caddys](/customers/caddys) - BeautyStar - Farma.it - Top Farmacia - ClimaConvenienza **Manifesto:** Retail has software to store product data. We built the agents that actually do the work, freeing people to focus on what matters. ## What changes when agents run the catalog. *Customers* See what happens when catalog work stops being manual. - STIGA · How STIGA launched on seven marketplaces in four months · 500+ · PDPs live in five languages · [/customers/stiga](/customers/stiga) - Cisalfa Sport · How Cisalfa cut the time to publish a product by eighty per cent · 80% · less time to publish a new product · [/customers/cisalfa-group](/customers/cisalfa-group) - Nestlé Purina · How Purina got its catalog ready to be recommended by AI agents · 2 · categories live in AI answers, cat food first · [/customers/purina](/customers/purina) ## Four layers make it work. *The platform* CommerceClarity reads the systems you already use, puts specialised agents to work on your catalog, and applies your rules before anything gets published. ![The CommerceClarity grid: products and their attributes as columns, with flagged values in amber.](https://cdn.sanity.io/images/xd4hrbt2/production/fc4e47cd302ace22f34282f338c98e537d9aed7a-2880x1920.webp?auto=format) - Context · Everything the agents know · Product data, market signals and company rules, connected from the systems you already use. - Agents · Specialised agents do the work · Enrich, translate, classify, audit, optimise and publish. Each agent is built for a specific retail workflow. - Governance · You decide what goes live · Your rules check every output. Humans review what’s uncertain. Nothing publishes unchecked. - Grid View · Your catalog in one place · Work across the whole catalog in one place, with every unresolved value surfaced for review. ## One catalog. Eight areas. Deploy agents where you need them, one workflow at a time. Every agent works from the same product record, so every improvement compounds across the business. ## What the agents have already done. *The record* Already running on enterprise catalogs. - 30M+ · attributes written and checked · Written on live catalogs, each on the customer’s own attribute model. - up to 95% · lower catalog ops cost · Your team stops typing product data and starts reviewing it. - 8x · faster than the manual process · Your team built each product by hand. One run now does it in a day. ## Your catalog data stays yours *Trust* Agents work on your data without taking ownership of it. Every action is governed, auditable and controlled by your team. - ISO/IEC 27001:2022 - GDPR ## Trust commitments - No training on your data · Your product data is never used to train shared or third-party models. - Encrypted, and least privilege · TLS 1.2+ in transit, AES-256 at rest, least-privilege internal access. - Every run on the record · A full audit trail of what published, what was held, and which check held it. - Yours to export · Your product data are exportable at any time. What happens to them when the contract ends is agreed when you sign. [Read the Trust Center](/trust-center) ## See what this looks like on your catalog. We map your catalog operations and costs, and show you where we can unlock growth. [Book a demo](/book-a-demo) ## Solution areas - [Catalog Foundation](/solutions/catalog-foundation) - [Channels & Markets](/solutions/channels-markets) - [Search & Discovery](/solutions/search-discovery) - [SEO & GEO/AEO](/solutions/seo-geo) - [Compliance & Quality](/solutions/compliance-quality) - [Merchandising](/solutions/merchandising) - [Physical Retail](/solutions/physical-retail) - [Agentic Commerce](/solutions/agentic-commerce) ## Entities - [Catalog Foundation](/solutions/catalog-foundation) · mentions - [Channels & Markets](/solutions/channels-markets) · mentions - [Search & Discovery](/solutions/search-discovery) · mentions - [SEO & GEO/AEO](/solutions/seo-geo) · mentions - [Compliance & Quality](/solutions/compliance-quality) · mentions - [Merchandising](/solutions/merchandising) · mentions - [Physical Retail](/solutions/physical-retail) · mentions - [Agentic Commerce](/solutions/agentic-commerce) · mentions - How STIGA launched on seven marketplaces in four months · mentions - How Cisalfa Group got its catalog ready for the new season without growing the team · mentions - [Prada](https://www.prada.com) · Organization, mentions - [Arcaplanet](https://www.arcaplanet.it) · Organization, mentions ## Sources - [CommerceClarity completes €2.7M funding to power the agentic era of e-commerce](https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/) · Tech.eu - [CommerceClarity acquires Katalogo.ai](https://tech.eu/2026/06/04/commerceclarity-acquires-katalogoai/) · Tech.eu - [CommerceClarity raccoglie 2,7 milioni per potenziare l'e-commerce con l'IA](https://forbes.it/2025/11/13/commerceclarity-raccoglie-27-milioni-per-potenziare-le-commerce-con-lia) · Forbes Italia - [Why we invested in CommerceClarity](https://italianfoundersfund.substack.com/p/why-we-invested-in-commerceclarity) · Italian Founders Fund --- # The agents do the work. Your team decides what ships. *Platform* > CommerceClarity Agent OS is the platform the company’s catalog agents run on, and it has four surfaces: Context, what an agent knows before it starts; Agents, the plan it builds from a target; Governance, the customer’s own rules turned into the standard every output is scored against; and Grid View, the catalog itself with every value the agents touched. Anything the checks cannot settle is held for a person to approve before it reaches a channel. Four surfaces: what they know, what they do, what they pass, and where your team picks it up. [Book a demo](/book-a-demo) ## Surfaces ### The agents start from what your company knows context *Context* The agent reads all of it before it writes a word: your rules, your systems, and the market outside. - Yours · Rules, glossary, attribute model, and the systems you already run. - And what you do not control · Search demand, what the engines answer, and what a category must state. [See Context](/platform/context) ### You give it the target. It builds the plan. agents *Agents* Bring the number you are measured on and the date you need it by. The agent works out the steps. - Fitted to one catalog · Same shape for everyone, your model inside. - Human in the loop · What the checks cannot settle is held for a person to decide. [See Agents](/platform/agents) ### Your rules become the standard every agent is held to governance *Governance* Encoded once, scored on every output, and a person decides what the score cannot settle. - Measured, not asserted · Every agent carries a quality target, and every output is scored on it before it moves. - Explainable one value at a time · The rule it failed, and the sentence it would have written. [See Governance](/platform/governance) ![The catalog grid with a flagged short description and the issue read where the value is: too generic and not aligned with the retailer tone.](https://cdn.sanity.io/images/xd4hrbt2/production/d28affc652f461d768e312804e68de5ce65dead5-2880x1920.webp?auto=format) ### The screen your team works in gridview *Grid View* Your team goes straight to the values the agent could not settle, marked in the cell. - The view is the export · Build it once, save it, and export it in the column order you set. - A thousand rows in one gesture · Filter, select, and run an agent on the selection. [See Grid View](/platform/gridview) ![The grid with the view switcher open: saved custom views, all products, errored products, and the rename, duplicate and delete actions on one of them.](https://cdn.sanity.io/images/xd4hrbt2/production/42732d07f4ed5cc491b58213f6b67668e37e6dd0-1254x970.webp?auto=format) ## Open it on your own catalog Bring one category and the attributes you never got to fill. Half an hour in the platform beats a deck. [Book a demo](/book-a-demo) ## Questions ### What is CommerceClarity Agent OS? The platform our catalog agents run on. It has four surfaces: the context an agent reads, the plan it builds, the rules it is scored on, and the grid where your team reviews the output. ### Do we operate the platform ourselves? No. We run it alongside your team, and a Retail Engineer brings agents live category by category. ### Does it replace our PIM, ERP or DAM? No. The agents read those systems and write the finished record back. You do not migrate anything, and you keep one catalog. ### What stops an agent publishing something wrong? Every value it writes is scored against your rules first. What clears the bar you set goes to the channel; anything the checks cannot settle is held with the reason, and a person decides. Every published value carries the rule it passed. ### How long before the first workflow is live? Weeks. The first workflow goes live on one category, and the next starts from the records it built. ## Entities - [CommerceClarity Agent OS](https://commerceclarity.com/platform) · Product, about ## Sources - [AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework) · NIST - [Regulation (EU) 2024/1689, the Artificial Intelligence Act](https://eur-lex.europa.eu/eli/reg/2024/1689/oj) · EUR-Lex - [Product structured data](https://developers.google.com/search/docs/appearance/structured-data/product) · Google Search Central - [GTIN, the GS1 global trade item number](https://www.gs1.org/standards/id-keys/gtin) · GS1 --- # Any model can write. Knowing your rules is the work. *Platform · Context* > Context is the CommerceClarity platform surface that holds what an agent knows before it writes a word. It carries three things: the rules inside the company, including which source may fill which attribute and the family, channel or market each rule holds for; the systems the data already lives in, mapped once to canonical names and written back to rather than replaced; and the two contexts nobody controls, market demand and what a regulated category has to state. The rules your team never wrote down, the systems your data already lives in, and what the market and the law require. The agent reads all of it before it writes. [Book a demo](/book-a-demo) ## Blocks ### Your rules, and where each one holds inside *Inside your company* What you always say and never say, which source may fill which attribute, how a unit is written, what a category has to state. Most of it was never written down. It lives with the people who have been there longest. - The rules themselves · Claims you refuse, terms you approve, and the source each attribute may read from. - Where each one holds · A rule can be true for one family in one market and nowhere else. The agent is told which. ### It reads your systems. It does not replace them. systems *Where your data lives* The supplier feed, the PIM, the ERP, the DAM, the barcode. Each field is mapped once to a canonical name, and the finished record goes back to the system your team already works in. You keep one catalog, where it lives today. - Feed, API, or a file drop · Whatever the source is today, on the schedule it already runs. - It writes back · The finished record returns to your system of record, ready for the channel. ### The market, and what the law requires outside *Outside your walls* Two things you do not control. What shoppers search for in your category and what the AI assistants tell them about it, and what a regulated product has to state before it can be sold. - Market data · The words buyers use, the demand behind each one, and what the engines say about the category. - What the category has to state · The labels, claims and declarations a product cannot go on sale without. ## Show us the parts nobody wrote down A category, the glossary if you have one, and the two people who know the rest. That is where context starts. [Book a demo](/book-a-demo) ## Questions ### What counts as context for a catalog agent? Three things: the rules inside your company and where each one holds, the systems your product data already lives in, and what happens outside them, search demand and what a regulated category has to state. ### Most of our rules were never written down. Is that a blocker? No, it is the normal starting point. A category, the glossary if there is one, and the two people who know the rest is enough to begin. Writing the rules down is part of the work, and it is what the first weeks produce. ### Do you migrate our data to run this? No. We read the supplier feed, the PIM, the ERP and the DAM where they are, map each field once to a canonical name, and send the finished record back to your system of record. You keep one catalog. ### Can a rule apply to only part of the catalog? Almost every one does. The scope is stored with the rule, so a rule that is true only for Apparel in Germany is applied only there. ### Where does the outside context come from? Search demand for the words buyers actually use, what the AI assistants answer about your category, and the requirements a regulated product has to state. None of it is yours to control, and all of it changes what the record has to say. ## Entities - [GS1](https://www.gs1.org) · Organization, mentions ## Sources - [GTIN, the GS1 global trade item number](https://www.gs1.org/standards/id-keys/gtin) · GS1 - [GDSN, the Global Data Synchronisation Network](https://www.gs1.org/standards/gdsn) · GS1 - [Regulation (EU) 2023/988 on general product safety](https://eur-lex.europa.eu/eli/reg/2023/988/oj) · EUR-Lex - [Product](https://schema.org/Product) · Schema.org --- # An agent for your catalog, not a tool to configure. *Platform · Agents* > Agents is the CommerceClarity platform surface where a catalog agent is given a target rather than a configuration. It reads the context and attribute model already agreed with the customer’s team, works out what it reads, what it fills and what it does with what it cannot source, then goes live on one category before taking the next. Agents come from a library of templates adapted to the customer, or are co-designed when no template covers the workflow. Every step runs against your attribute model and the bar your team agreed. The agent goes live on one category, then takes on the next, and a person approves anything that reaches a channel. [Book a demo](/book-a-demo) ## Blocks ### You give it the target. It builds the plan. plan *It starts from your KPIs* Bring the number you are measured on and the date you need it by. The agent works out what to read, what to fill, and what to hold for your team. - Your context, not a blank slate · It reads the product and company data you already have. · [See what it knows](/platform/context) - Your governance, from the first run · The bar and the rules it is scored against are yours. · [See how it is checked](/platform/governance) ### It grows without a new project grows *After it goes live* The agent goes live on one category with three steps, in weeks. Every month after that the Retail Engineer who put it live adds something else, and the bar it clears goes up. - Read it down its own history · Three steps in January, six now, and the date each one arrived. - Someone owns it after launch · A Retail Engineer stays on the account after go-live. ### A person decides everything the agent could not human *Human in the loop* The agent scores its own work against your bar. What clears it goes to the channel, and the rest waits in a queue with the reason it was held. - The bar is yours to move · Raise it and more comes to you. Lower it and less does. - 50, not 1,200 · The queue holds only what the agent could not settle. ## Put an agent on one of your categories Bring a category, the attributes you never got to fill, and the rules it has to respect. You will see the run and what it holds back for you. [Book a demo](/book-a-demo) ## Questions ### What is a catalog agent, in practice? One workflow a retail team does by hand today, now run by software: sourcing an attribute, matching a supplier record, translating a description. Each agent has its own output, its own bar, and its own numbers. ### Do we configure the agent ourselves? No. You bring the number you are measured on and the date you need it by, and the agent works out the steps against your attribute model. ### What if the agent we need does not exist yet? Most already exist as library templates built for other customers, and adapting one to your catalog is part of the service. A brand-new agent is designed with your team and scoped separately. ### Does it run on the whole catalog at once? No. It goes live on one category, proves the number moved, and then takes on the next. Every category it finishes makes the following one cheaper, because the records are already there. ### Where does a person come into it? At the end, where it matters. The agent marks what it could not settle, a reviewer resolves those values in Grid View, and the correction becomes what the agent carries next. ## Sources - [AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework) · NIST - [Regulation (EU) 2024/1689, the Artificial Intelligence Act](https://eur-lex.europa.eu/eli/reg/2024/1689/oj) · EUR-Lex - [Product data specification](https://support.google.com/merchants/answer/7052112) · Google Merchant Center - [GTIN, the GS1 global trade item number](https://www.gs1.org/standards/id-keys/gtin) · GS1 --- # Control, by design. *Platform · Governance* > Governance is the CommerceClarity platform surface that keeps the output controlled rather than unmanaged. The customer’s own rules become the standard every agent is held to, encoded once and applied across every SKU, market and channel. Every output is scored against those rules before it moves, with layered checks that catch regressions and drift. What the system is not sure about is flagged and waits for a person to approve, so human review lands on the exceptions rather than on the whole catalog. Product data is strategic, and in many categories it is regulated. Our job is to make sure the output is always controlled, never unmanaged. Your rules become the standard, every output is scored against it, and a human keeps the final say. [Book a demo](/book-a-demo) ![The catalog grid on the online store dataset: eighteen products, values the agent flagged in amber, missing values in pink, and the issue on a short description read where the value is.](https://cdn.sanity.io/images/xd4hrbt2/production/d28affc652f461d768e312804e68de5ce65dead5-2880x1920.webp?auto=format) ## Blocks ### Every agent is held to the same standard, and that standard is your rules. standard *North Star rules* Your rules - from brand voice to category and compliance requirements - become instructions every agent follows automatically. - Your rules, applied automatically. · Brand voice, content guidelines, category requirements and compliance rules - encoded once and applied every time. - One standard, everywhere. · Set once, applied across every SKU, market, and channel, scoped per market and category where needed. **Alt** What the agent read and what it checked on one value: the supplier feed and the brand data sheet both had something, web search did not, and of the four checks it ran three passed and one failed on technical materials. **Url** [https://cdn.sanity.io/images/xd4hrbt2/production/763b8520034a300ff2cea7f91f83e47fc6cfeab5-1600x976.png?auto=format](https://cdn.sanity.io/images/xd4hrbt2/production/763b8520034a300ff2cea7f91f83e47fc6cfeab5-1600x976.png?auto=format) ### Your team stays in control, without checking every product. measured *Automated checks, human judgment* Every output is scored against your rules before it moves. What the system is not sure about is flagged and waits for a person; what passes with confidence publishes. - Every output scored before it moves. · Format and schema checks catch the hard errors, an AI grader reads the rest against your rules, and the scores track quality across the whole catalog. - Review only what needs your attention. · Issues are flagged automatically, so your team spends its judgment on the exceptions instead of re-reading the catalog. **Alt** The review panel beside the grid on one attribute: the current short description, the rule it breaks, the suggested revision, apply or keep, and the related attributes the agent read to decide. **Url** [https://cdn.sanity.io/images/xd4hrbt2/production/4f126f984186173c87085870047cf04b4ad6bb64-1538x1140.webp?auto=format](https://cdn.sanity.io/images/xd4hrbt2/production/4f126f984186173c87085870047cf04b4ad6bb64-1538x1140.webp?auto=format) ## Measured, not assumed *What control looks like* - 100% · of outputs checked against your rules - Continuous · ∞ · quality tracking, not a one-off - 0 · unmanaged outputs reach your catalog ## Control every output. Build trust, stay compliant, and publish with confidence. *Why it matters* - Trust · Every output is checked against your standards before it ships, so your team can trust AI with more of the work. - Compliance · Regulated claims and category rules are applied automatically, with a full audit trail behind every decision. - A compounding advantage · Your team's decisions become part of the standard, so quality improves over time. ## Bring the checks you cannot get wrong Bring a category, its rules, and the products you would use to judge it. We build the set with your team. [Book a demo](/book-a-demo) ## Questions ### What counts as a rule? Anything you would send a product page back for. How you sound, the claims you refuse, what a category is required to state, the format a value has to take, the target a market is held to. You write it down once, and from then on every agent answers to it. ### How do you test quality? In layers, cheapest test first. Format and schema are settled outright, because a malformed value needs no judgment. What is left goes to a grader that reads it against your own rubric, and those scores roll up across the catalog so no category hides behind an average. Where the two disagree, a person decides, and that decision becomes the reference. Nothing moves without a score. ### Do you check every output? Yes. Every output runs through the quality checks against your rules, continuously, not sampled once at launch. ### How do we stay in control of the output? Every output is checked against your rules before it can publish. What the system is not sure about is flagged and waits for a person. What passes with confidence goes out, so your team is not re-reading a catalog to find the few values that need a decision. ## Entities - [EU Artificial Intelligence Act](https://eur-lex.europa.eu/eli/reg/2024/1689/oj) · Thing, mentions ## Sources - [AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework) · NIST - [Regulation (EU) 2024/1689, the Artificial Intelligence Act](https://eur-lex.europa.eu/eli/reg/2024/1689/oj) · EUR-Lex - [Regulation (EU) 2023/988 on general product safety](https://eur-lex.europa.eu/eli/reg/2023/988/oj) · EUR-Lex - [Creating helpful, reliable, people-first content](https://developers.google.com/search/docs/fundamentals/creating-helpful-content) · Google Search Central --- # Everything you do in the spreadsheet, on the catalog itself. *Platform · Grid View* > Grid View is the CommerceClarity platform surface where a person picks up what the agents left. Thousands of products sit in rows, every attribute is a column, and each value the agent could not settle is marked in its own cell with the rule it failed and what it read. Views are named, duplicated and updated, one per channel where needed, and the export is the view: the same columns in the same order. Filters hold a condition, and an agent can be run on the selection. Thousands of products, every attribute a column, and the values the agent could not settle marked in the cell. You fix them where they are, and export the view exactly as you built it. [Book a demo](/book-a-demo) ![The CommerceClarity grid: a catalog of products with their attributes as columns, values the agent flagged in amber, and a status bar counting issues and missing values.](https://cdn.sanity.io/images/xd4hrbt2/production/fc4e47cd302ace22f34282f338c98e537d9aed7a-2880x1920.webp?auto=format) ## Clusters ### Build the view once, export it as it is views *Views and export* A view is a job: the columns it needs, the families it covers, the filter that finds its rows. Save it under a name, duplicate it for the next channel, and export it as it is, in the column order you set. - Named, duplicated, updated · A view is yours to keep, rename, and hand to someone else. - One view per channel · The online store and the marketplace each get their own columns. - The export is the view · Same columns, same order, ready to send. ![The grid with the view switcher open: saved custom views, all products, errored products, and save as new view.](https://cdn.sanity.io/images/xd4hrbt2/production/42732d07f4ed5cc491b58213f6b67668e37e6dd0-1254x970.webp?auto=format) ### Every attribute you have, the few this job needs attributes *The columns* The picker searches your attribute model, and the model is not flat: Size on sneakers and Size on apparel are two attributes, so they are two columns. Show the ones this job needs and hide the rest. - Search instead of scrolling · Type the attribute you are working on and it is on screen. - Per family, not per catalog · Different families, different attributes, different columns. - Show only, or hide all · One attribute across every product, or a clean screen to start from. ![The attribute picker open over the grid, with attributes scoped by family and the show only, show all and hide all controls.](https://cdn.sanity.io/images/xd4hrbt2/production/312f093fa24cd2418031c437b84060f7ce242f0b-1148x842.webp?auto=format) ### Every problem is already marked issues *Review* An issue can be a value two sources disagree on, or a description that does not sound like you. The panel names the rule it breaks, shows the rewrite in your tone, and lists what it read to decide. Accept it or keep what you had, then on to the next one. - Your rules, checked · The agent is scored against your tone of voice on every value it writes. - The rewrite, ready · The panel shows the sentence it would publish instead. - What it read to decide · The sources are on the panel, so you can check the call it made. ![The review panel beside the grid: the current short description, the rule it breaks, the suggested rewrite, and the related attributes.](https://cdn.sanity.io/images/xd4hrbt2/production/4f126f984186173c87085870047cf04b4ad6bb64-1538x1140.webp?auto=format) ### A thousand rows in one gesture bulk *At catalog scale* Filter to the products that share a problem, select them, and act on all of them at once: run an agent, move the status, tag them, build a parent from the variants. The cells are typed, so a choice list edits as a choice list. - Filters that hold a condition · Status, category, code, tag, date, combined and kept. - Typed cells · Text, number, boolean, single and multi select, rich text, images. - Run an agent on the selection · The grid is where the work starts, and where it comes back. ![The selection toolbar over the grid: six products selected, run agent, and the more menu.](https://cdn.sanity.io/images/xd4hrbt2/production/1260bb4c8ffb8a339e96a25d4d6a59e05e73a3f0-1420x1062.webp?auto=format) ## Configured with you, then yours *How it gets set up* Your attribute model, your families, the sources each attribute may read from, the threshold that decides what waits for a person. We set it up with you, and your team changes it from then on, without opening a ticket. ## Open it on your own catalog Bring one category and the attributes you never got to fill. Half an hour in the grid tells you more than a deck. [Book a demo](/book-a-demo) ## Questions ### Is this just a spreadsheet with a different name? It does what a team already does in a spreadsheet, on the catalog itself rather than on a copy of it. Cells are typed, the rules are checked as you work, and the work is already in the catalog when you are done. ### We have hundreds of attributes. Do we see all of them? Only the ones the job needs. Columns are searched rather than scrolled, chosen per family rather than per catalog, and a view can show only a handful or hide everything else. ### Does the export match what we see? Yes. The export is the view: the same columns in the same order you built. Views are named, duplicated and updated, so the file a channel gets always comes from a saved view. ### How do we find the products that need attention? You walk the problems instead of hunting for them. Values that failed a rule are marked in the cell, the suggested rewrite is ready, and what the agent read to decide is one click away. ### Can we act on thousands of rows at once? Yes. A filter holds a condition instead of a one-off search, the selection stays as you work, and you can run an agent on it in one gesture. ## Sources - [Product data specification](https://support.google.com/merchants/answer/7052112) · Google Merchant Center - [Product structured data](https://developers.google.com/search/docs/appearance/structured-data/product) · Google Search Central - [Product](https://schema.org/Product) · Schema.org - [GTIN, the GS1 global trade item number](https://www.gs1.org/standards/id-keys/gtin) · GS1 --- # The people who run enterprise catalogs. > Seven customer stories, each with the figure that customer signed off. STIGA: seven marketplaces, five languages, four months. Yeppon: 80,000 SKUs and 24,000 marketplace listings at eight times the speed. Cisalfa Sport: eighty per cent less time to publish. 1000Farmacie: 38,000 pages rebuilt, +30% revenue month over month. Next: 500+ pages a month, +30% organic visibility. Nestlé Purina: cat food recommended by AI assistants. Caddy’s: compliance checked as the record is built. Enterprise retailers and brands with catalogs too large to hold by hand. Here is what they run with us, and what they said about it. **Wall caption:** Trusted by 50+ enterprise retailers and brands ## Wall tiles - Prada - Nestlé · [/customers/purina](/customers/purina) - Cisalfa Sport · [/customers/cisalfa-group](/customers/cisalfa-group) - Arcaplanet - Intersport - Dr. Max - STIGA · [/customers/stiga](/customers/stiga) - Prénatal - Atida eFarma - Gruppo Hippocrates - Yeppon · [/customers/yeppon](/customers/yeppon) - Scalapay - Next Hardware & Software · [/customers/next-hardware-software](/customers/next-hardware-software) - Caddy’s · [/customers/caddys](/customers/caddys) - BeautyStar - Farma.it - Top Farmacia - ClimaConvenienza ## Quotes - For the first time, we are going to be compliant before the deadline. · Gianfranco Bello · IT · Arcaplanet - The team validates output now. They have stopped writing. · Alberto Colnaghi · Cisalfa Sport - We are live. Yesterday we sent five, today we are sending five hundred. · Gianfranco Bello · IT · Arcaplanet - Now we find out whether your output answers for the German market too, at the quality level we reached in Italy. · Francesco Gasca · Cisalfa Sport - There is support in real time, even. · Gianfranco Bello · IT · Arcaplanet - I do not think we are far off building the right pages for the products. · Matia Tosti · Yeppon - You are what lets us be relaxed enough to say: fine, let us triple the selection. · Alberto Colnaghi · Cisalfa Sport ## Catalogs at the scale where hands stop working *The customers* Seven retailers and brands, and what the agents actually run for each of them. - STIGA · 7 marketplaces · live in five languages, in four months · [/customers/stiga](/customers/stiga) - Yeppon · 8x · faster to market, 80,000 SKUs processed end to end · [/customers/yeppon](/customers/yeppon) - Cisalfa Sport · 80% · less time to publish a new product · [/customers/cisalfa-sport](/customers/cisalfa-sport) - 1000Farmacie · +30% · revenue month over month, 38,000 pages rebuilt · [/customers/1000farmacie](/customers/1000farmacie) - Next Hardware & Software · 500+ · new product pages a month, thirty per cent more visibility · [/customers/next-hardware-software](/customers/next-hardware-software) - Nestlé Purina · 2 · categories live in AI answers, cat food first · [/customers/purina](/customers/purina) - Caddy’s · Thousands · of SKUs checked against the rules as they are built · [/customers/caddys](/customers/caddys) ## See what the agents do with your catalog Bring one messy feed, or the backlog of products you never got to. We will show you the rest. [Book a demo](/book-a-demo) ## Customer stories - [How 1000Farmacie reorganized its catalog and grew revenue thirty per cent a month](/customers/1000farmacie) · Online pharmacy - [How Caddy’s made compliance a property of its product data](/customers/caddys) · Beauty and household retail - [How Cisalfa Group got its catalog ready for the new season without growing the team](/customers/cisalfa-group) · Sports retail - [How Purina got its catalog ready to be recommended by AI agents](/customers/purina) · Pet nutrition - [How enriched product pages lifted search and conversion for Next](/customers/next-hardware-software) · IT and gaming distribution - [How STIGA launched on seven marketplaces in four months](/customers/stiga) · Outdoor power equipment - [How Yeppon got a 150,000 SKU catalog marketplace-ready, from its own store outward](/customers/yeppon) · Consumer electronics ## Questions ### Which customers has CommerceClarity published a story about? Seven: STIGA, Yeppon, Cisalfa Sport, 1000Farmacie, Next Hardware & Software, Nestlé Purina and Caddy’s. Each story carries the figures that customer agreed we may publish, and the card on this page opens on the same figure the story does. ### What kind of retailers and brands do you work with? Enterprise retailers and brands with catalogs too large to keep by hand. The published stories span sports retail, consumer electronics, online pharmacy, pet nutrition, beauty and household, and outdoor power equipment. Catalog sizes run from a few thousand SKUs to over 150,000. ### How long before the numbers move? It depends on the work, and the published stories give the range. Cisalfa Sport went from weeks to hours on publication time. STIGA went live on seven marketplaces in four months against the one to two years the sequential route takes. 1000Farmacie reached thirty per cent revenue growth month over month within six months. ### Which agents do these customers run? Each story names the agents actually running on that catalog, and they are the same agents the solution pages describe: Data Enrichment, Matching, Taxonomy Mapping, Content Translation, the Amazon and Leroy Merlin listing optimizers, Keyword Optimizer, GEO Content Writer, GEO Share of Voice and Catalog Auditing. ### Can we speak to a reference customer? Ask, and we arrange it with that customer’s consent. Reference calls are agreed with the customer each time, never offered on their behalf. ### Do you publish customers who have not agreed to be named? No. A story is published only with that customer’s sign-off, on the figures and on the quotes. The wall of marks above carries names we may state publicly, and nothing on this page is a claim a customer has not seen. ## Entities - STIGA · Organization, about - Yeppon · Organization, about - Cisalfa Sport · Organization, about - 1000Farmacie · Organization, about - Next Hardware & Software · Organization, about - Nestlé Purina · Organization, about - Caddy’s · Organization, about ## Sources - [Product structured data markup](https://developers.google.com/search/docs/appearance/structured-data/product) · Google Search Central - [Product data specification](https://support.google.com/merchants/answer/7052112) · Google Merchant Center - [Product](https://schema.org/Product) · schema.org - [Listing quality and product data requirements](https://sellercentral.amazon.com/help/hub/reference/GT4CDCFTFPWKMYNW) · Amazon Seller Central --- # The CommerceClarity blog > The CommerceClarity blog covers what changes when AI agents own the product data pipeline: product attributes and taxonomy, channel and marketplace rules, catalog compliance, on-site search, SEO and generative engine optimization, and agentic commerce. Articles are written by the people who run the agents on enterprise catalogs, and each one carries its own Q&A. From the CommerceClarity blog: what changes when agents own the product data pipeline. ## Get the latest from CommerceClarity New explainers, regulation updates, and research, straight to your inbox. [Book a demo](/book-a-demo) ## Articles - [ChatGPT product feed: nine required fields, and the optional ones that decide what gets surfaced](/blog/chatgpt-product-feed-required-fields) · 2026-09-18 · A ChatGPT product feed needs nine required fields. Optional attributes give OpenAI the data to filter, compare and surface the right products. - [Your ecommerce has an AI agent score.](/blog/llms-txt-webmcp-ecommerce) · 2026-09-14 · We ran Lighthouse’s experimental Agentic Browsing checks on our own site and read what they measure. What a retailer should do with the result, what to try with WebMCP and llms.txt, and where the real preparation for agentic commerce happens: the catalog. - [Claude Commerce Agents: the agent got easy to build. The catalog decides what it can enforce.](/blog/claude-commerce-agents-catalog-constraint) · 2026-09-03 · Anthropic and Shopify made the shopping agent easy to build. We read the code to show what it asks of the catalog underneath, and why creating the field is only the start. - [Meet the Retail Engineer](/blog/meet-the-retail-engineer) · 2026-09-01 · One person owns an AI agent deployment in retail, from the first business question to the live catalog. This is what a Retail Engineer does, and where the forward deployed engineer fits. ## Questions ### What does the CommerceClarity blog cover? Retail catalog operations and the agents that run them: product attributes and taxonomy, channel and marketplace rules, catalog compliance, on-site search, SEO and generative engine optimization, and agentic commerce. ### Who writes it? The people who do the work: Retail Engineers, product and go-to-market at CommerceClarity. Every article carries a named author with their own page, so you can see what else they wrote. ### Are the claims sourced? Yes. Every claim carries a source, and customer numbers appear only with the customer’s sign-off. Where something is not yet proven, the article says so rather than rounding it up. ### Can we quote or republish an article? Quote it with a link back. For a full republication or a syndication, write to press@commerceclarity.com and we will sort out the terms. ## Entities - [Google](https://www.google.com) · Organization, mentions - [OpenAI](https://openai.com) · Organization, mentions - [GS1](https://www.gs1.org) · Organization, mentions ## Sources - [Creating helpful, reliable, people-first content](https://developers.google.com/search/docs/fundamentals/creating-helpful-content) · Google Search Central - [Product structured data](https://developers.google.com/search/docs/appearance/structured-data/product) · Google Search Central - [Commerce documentation](https://developers.openai.com/commerce/) · OpenAI - [GTIN, the GS1 global trade item number](https://www.gs1.org/standards/id-keys/gtin) · GS1 --- # Meet us where the work happens. *Events* > CommerceClarity meets retailers at the events where catalog work is actually discussed: Netcomm Forum in Milan, the eTail series run by Worldwide Business Research in the UK and the United States, Shoptalk Europe in Barcelona, and the Ecommerce Tech Summit run by Casaleggio Associati. Upcoming entries carry the dates, the venue, the people who will be there and how to book time; past ones stay up with what came out of them. Conferences we show up to, webinars you can join live, and every session you can watch on demand. ## Also on the calendar Where else you can find us this year. ## Sessions you can watch back ## Rooms we have been in ## Rather not wait for the next one? Half an hour on your own catalog tells you more than a session does. [Book a demo](/book-a-demo) ## Events - [Meet CommerceClarity at Netcomm Forum 2027](/events/netcomm-forum-2027) · 2027-04-21 · Netcomm Forum is the Italian ecommerce industry in one room, and it is where we met most of the retailers we work with today. In 2027 we come back with a bigger stand and a main-hall workshop. - [Meet CommerceClarity at the Ecommerce Tech Summit 2026](/events/ecommerce-tech-summit-2026) · 2026-10-20 · The Ecommerce Tech Summit is Casaleggio Associati’s day on the technology under Italian ecommerce. We are the main partner of the 2026 edition: Federico Sargenti on stage, and the team in the room for the whole day. - [Meet CommerceClarity at eTail Autumn Interactive 2026](/events/etail-autumn-interactive-2026) · 2026-09-29 · eTail Autumn Interactive is two days of pre-booked one-to-one meetings rather than a show floor, which is why we go: thirty minutes, one catalog, a real conversation. We are there as a partner, and the schedule is set before the doors open, so the useful thing to do now is put a slot in the diary. - [We were at eTail Boston 2026](/events/etail-boston-2026) · 2026-08-10 · Our first eTail in the United States. A different market, the same catalog problem, and the conversation that came back was the same one we have in Milan and in Manchester. - [We were at eTail UK 2026](/events/etail-uk-2026) · 2026-06-16 · Two days in Manchester with UK retail, on the format we like: scheduled conversations rather than a stand to walk past. Most of what we do in the UK today traces back to rooms like this one. - [We were at Shoptalk Europe 2026](/events/shoptalk-europe-2026) · 2026-06-09 · Shoptalk Europe is where the retailers we do not yet know are in one place. Barcelona was our first pass at it, and the conversations that came out of it were with retailers outside the markets we had been working, which is exactly what we went for. - [We were at Netcomm Forum 2026](/events/netcomm-forum-2026) · 2026-05-06 · The 21st Netcomm Forum, and our first as an exhibitor. Two days at the Allianz MiCo with the Italian ecommerce industry in one building. Most of the retailers we talk to today, we met in a room like that one, which is why we go back in 2027 with a bigger stand and a workshop. - [We were at eTail Connect Spring 2026](/events/etail-connect-spring-2026) · 2026-03-03 · Our first eTail of the year, at Twickenham, and the first time we took the catalog argument to a UK room. What we learned about which format works for a conversation like ours is what shaped how we pick these events now. ## Questions ### Which events does CommerceClarity attend? Market events where catalog and ecommerce leaders are in the room: Netcomm Forum in Milan, the eTail series in the UK and the United States, Shoptalk Europe in Barcelona, and the Ecommerce Tech Summit in Milan. Private dinners and customer sessions are not listed here. ### How do we book time with the team at an event? Every upcoming event page carries a booking link. At the one-to-one formats the schedule is set before the doors open, so the useful thing is to take a slot early rather than to look for the stand on the day. ### What do you actually show? A live agent run, on your own catalog rather than on a demo dataset where we can. Bring a category and one messy feed and there is enough to work with. ### Why do past events stay on the site? Because what came out of them is the useful part: which format works for a conversation about catalogs, and what a retail team asked in the room. A past event page says what happened rather than pretending the date has not passed. ## Entities - [Consorzio Netcomm](https://www.consorzionetcomm.it) · Organization, mentions - [Worldwide Business Research](https://www.wbresearch.com) · Organization, mentions - [Shoptalk](https://shoptalk.com) · Organization, mentions - [Casaleggio Associati](https://www.casaleggio.it) · Organization, mentions ## Sources - [Netcomm Forum](https://www.consorzionetcomm.it/netcomm-forum/) · Consorzio Netcomm - [eTail UK](https://etailuk.wbresearch.com/) · Worldwide Business Research - [Shoptalk Europe](https://europe.shoptalk.com/) · Shoptalk - [Ecommerce Tech Summit 2026](https://www.ecommerceitalia.info/report/ecommerce-tech-summit-2026/) · Casaleggio Associati --- # The CommerceClarity press area ## What we announced, as we announced it *Announcements* ## In the press ### E-commerce, la rete di agenti di intelligenza artificiale (su cui investono Elkann e Berlusconi) Corriere della Sera, L’Economia November 2025 Entro il 2030 fino a 5.000 miliardi di dollari di vendite retail globali potrebbero essere mediati da agenti di intelligenza artificiale. [https://www.corriere.it/economia/finanza/25_novembre_13/e-commerce-la-rete-di-agenti-di-intelligenza-artificiale-su-cui-investono-elkann-e-berlusconi-37686617-81a2-48dc-a88d-c7324ae5bxlk.shtml](https://www.corriere.it/economia/finanza/25_novembre_13/e-commerce-la-rete-di-agenti-di-intelligenza-artificiale-su-cui-investono-elkann-e-berlusconi-37686617-81a2-48dc-a88d-c7324ae5bxlk.shtml) ### CommerceClarity raccoglie 2,7 milioni per potenziare l’e-commerce con l’IA Forbes Italia November 2025 Il futuro del commercio sarà guidato da reti di agenti intelligenti capaci di collaborare tra loro per ottimizzare ogni decisione. [https://forbes.it/2025/11/13/commerceclarity-raccoglie-27-milioni-per-potenziare-le-commerce-con-lia](https://forbes.it/2025/11/13/commerceclarity-raccoglie-27-milioni-per-potenziare-le-commerce-con-lia) ### CommerceClarity completes €2.7M funding to power the agentic era of e-commerce Tech.eu November 2025 As the complexity of product data keeps growing and AI agents begin influencing what people buy, e-commerce needs intelligent infrastructure to remain visible, discoverable, and competitive. [https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/](https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/) ### CommerceClarity chiude un round da 2,7 milioni: operazione guidata dai fondi Iff e Entourage Milano Finanza November 2025 La startup permette ai team e-commerce di scalare e automatizzare le operazioni di catalogo con precisione, velocità e controllo, creando nuove opportunità di crescita. [https://www.milanofinanza.it/news/commerceclarity-chiude-un-round-da-2-7-milioni-operazione-guidata-dai-fondi-iff-e-entourage-202511131501462937](https://www.milanofinanza.it/news/commerceclarity-chiude-un-round-da-2-7-milioni-operazione-guidata-dai-fondi-iff-e-entourage-202511131501462937) ### CommerceClarity, round investimento per potenziare piattaforma IA ANSA November 2025 Costruire l’infrastruttura intelligente del commercio online moderno, permettendo ai team e-commerce di scalare e automatizzare le operazioni con precisione, velocità e controllo. [https://www.ansa.it/sito/notizie/economia/2025/11/13/commerceclarityround-investimento-per-potenziare-piattaforma-ia_0a81cc7d-37d5-42d6-b745-aef02d827e65.html](https://www.ansa.it/sito/notizie/economia/2025/11/13/commerceclarityround-investimento-per-potenziare-piattaforma-ia_0a81cc7d-37d5-42d6-b745-aef02d827e65.html) ### CommerceClarity bags €2.7M to help retailers cut costs and boost sales with AI Tech Funding News November 2025 CommerceClarity has the potential to redefine how digital commerce operates globally. It’s building the infrastructure for the agentic future of commerce. [https://techfundingnews.com/commerceclarity-raises-2-7m-ai-driven-ecommerce-catalog-management/](https://techfundingnews.com/commerceclarity-raises-2-7m-ai-driven-ecommerce-catalog-management/) ### L’Intelligenza artificiale per e-commerce di CommerceClarity porta a casa round da 2,7 milioni StartupItalia November 2025 Con CommerceClarity abbiamo costruito l’infrastruttura che permette a brand e retailer di trasformare i propri dati in azioni intelligenti. [https://startupitalia.eu/startup/commerceclarity-round-da-2-7-milioni-euro/](https://startupitalia.eu/startup/commerceclarity-round-da-2-7-milioni-euro/) ### CommerceClarity Raises €2.7 million to Build AI Operating System for Agentic E-commerce Startup Researcher November 2025 In the agentic era of commerce, the retailers that win will be those able to operate at scale with structured, consistent data. [https://www.startupresearcher.com/news/commerceclarity-raises-eur2-7-million-to-build-ai-operating-system-for-agentic-e-commerc](https://www.startupresearcher.com/news/commerceclarity-raises-eur2-7-million-to-build-ai-operating-system-for-agentic-e-commerc) ## Press kit Everything you need to write about us, without asking us first. - Logos · Wordmark and monogram, light and dark, in every format. · Download · [https://drive.google.com/drive/folders/1FKM7D7EBHdzm8Y-VWyZJUZNBC9x6cfOk?usp=sharing](https://drive.google.com/drive/folders/1FKM7D7EBHdzm8Y-VWyZJUZNBC9x6cfOk?usp=sharing) - Company profile · Who we are, what we do, the facts you can quote. · Download · [https://drive.google.com/drive/folders/1FKM7D7EBHdzm8Y-VWyZJUZNBC9x6cfOk?usp=sharing](https://drive.google.com/drive/folders/1FKM7D7EBHdzm8Y-VWyZJUZNBC9x6cfOk?usp=sharing) - Founders · Photos and short bios of the founding team. · Download · [https://drive.google.com/drive/folders/1FKM7D7EBHdzm8Y-VWyZJUZNBC9x6cfOk?usp=sharing](https://drive.google.com/drive/folders/1FKM7D7EBHdzm8Y-VWyZJUZNBC9x6cfOk?usp=sharing) - All assets · Everything above, in one archive. · Download · [https://drive.google.com/drive/folders/1FKM7D7EBHdzm8Y-VWyZJUZNBC9x6cfOk?usp=sharing](https://drive.google.com/drive/folders/1FKM7D7EBHdzm8Y-VWyZJUZNBC9x6cfOk?usp=sharing) ## Press enquiries Interviews, quotes, assets. We answer. [Contact us](mailto:press@commerceclarity.com) ## Announcements - [We have acquired Katalogo.ai and welcome Luca Cozzolino as Founding Chief Product Officer](/press/commerceclarity-acquires-katalogo-ai) · 2026-06-04 · Eighteen months after its founding and six months after the €2.7M round, the AI agent platform for catalog operations completes its first M&A deal and expands the founding team. - [CommerceClarity raises €2.7M ($3.1M) pre-seed to power the AI agentic era of e-commerce](/press/commerceclarity-raises-3m-ai-commerce) · 2025-11-13 · We’ve raised €2.7M ($3.1M) to build the intelligent infrastructure for the agentic era of commerce. --- # Trust Center *Trust* > The CommerceClarity Trust Center states what happens to a retailer’s catalog data. CommerceClarity is certified to ISO/IEC 27001:2022 and compliant with the GDPR. Catalog data is stored in EU regions, and model calls processed outside the EU are covered by Standard Contractual Clauses. Product data is encrypted in transit and at rest, internal access is least-privilege, every sub-processor works under a data processing agreement, and no customer data trains shared or third-party models. Your catalog data stays yours. Here is how we protect it, what we are certified against, and who to ask when you need more information. **Contact email:** security@commerceclarity.com ## Certified and compliant *Compliance* - ISO/IEC 27001:2022 · Certified · [https://cdn.sanity.io/files/xd4hrbt2/production/220cb98e8dfa06b81551e899b3733cd7e6874f91.pdf](https://cdn.sanity.io/files/xd4hrbt2/production/220cb98e8dfa06b81551e899b3733cd7e6874f91.pdf) · Certificate (PDF) - GDPR · Compliant ## How we protect your data - Encrypted everywhere · Data is encrypted in transit (TLS 1.2+) and at rest (AES-256). - Least-privilege access · Least-privilege internal controls, and a full audit trail. - Catalog data in the EU · Your catalog data is stored in EU regions. Model calls can be processed outside the EU, under Standard Contractual Clauses. - No training on your data · Your product data is never used to train shared or third-party models. - Monitoring and logging · Continuous monitoring, alerting, and a full audit trail of every run. - No output publishes unchecked · Your rules score every output. A human reviews anything uncertain, and nothing reaches your catalog without passing your rules first. ## You stay in control *Your data* - You own your data · Your catalog data is yours. We process it to do the job and nothing else. - Scoped sub-processors · Every sub-processor works under a data processing agreement with the same obligations we take on toward you. Ask us for the current list. - Export any time · Export your product data whenever you need it. What happens to it when the contract ends is agreed when you sign. ## Questions we get asked ### Do you train AI models on our catalog data? No. Your product data is never used to train shared or third-party models. It is processed only to run the jobs you ask for. ### Where is our data stored and processed? Your catalog data is stored in EU regions, encrypted at rest and in transit. Model calls are the exception. Most of them are processed outside the EU today, because the model we use most runs on a global service with no EU region. Those transfers are covered by Standard Contractual Clauses and supplementary measures in line with EDPB guidance. ### Are you certified? Yes. CommerceClarity is ISO/IEC 27001:2022 certified, so information security runs as an audited management system. An independent security assessment and penetration test was run in August 2026. ### How do you handle access control? Least-privilege internal controls, and a full audit trail. ### What happens to our data when the contract ends? You can export your product data at any time. Retention and deletion at the end of the contract are agreed when you sign. Without that clause, the account and its data are deleted when the contract ends. ### Can you fill in our security questionnaire? Yes. Send it over and we will complete it. ### Who is liable for what the AI produces? The AI Act splits the roles: CommerceClarity is the provider of the system, the customer is the deployer and stays responsible for what it publishes to the consumer. We do not take on the customer’s liability. What we do take on is making the output checkable: your rules score every output, anything uncertain goes to a human, and every run carries an audit trail showing which rules fired, where each value came from, and what the system flagged as a conflict. ### How do we report a vulnerability? Email security@commerceclarity.com with what you found and how to reproduce it. We will confirm we have it and tell you what we do about it. ## Need documentation or more detail? Anything your security review needs and this page does not cover, ask us. [Contact us](mailto:security@commerceclarity.com) ## Entities - [ISO/IEC 27001](https://www.iso.org/standard/27001) · Thing, about - [General Data Protection Regulation](https://eur-lex.europa.eu/eli/reg/2016/679/oj) · Thing, about - [EU Artificial Intelligence Act](https://eur-lex.europa.eu/eli/reg/2024/1689/oj) · Thing, mentions ## Sources - [ISO/IEC 27001, information security management](https://www.iso.org/standard/27001) · ISO - [Regulation (EU) 2016/679, the General Data Protection Regulation](https://eur-lex.europa.eu/eli/reg/2016/679/oj) · EUR-Lex - [Regulation (EU) 2024/1689, the Artificial Intelligence Act](https://eur-lex.europa.eu/eli/reg/2024/1689/oj) · EUR-Lex - [AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework) · NIST --- # We ran these operations *About* > CommerceClarity builds AI agents that run retail catalog operations: enrichment, translation, category rules and channel feeds, on the retailer's own catalog and with a person accountable for every value published. The founding team ran retail operations at Amazon, Everli, Zalando and Shopify before automating them. The company is headquartered in Milan, with offices in Barcelona and London. **Title second line:** before we automated them. ![An open workspace in low afternoon light, a long table under tall windows, and people crossing the floor in motion blur.](https://cdn.sanity.io/images/xd4hrbt2/production/a96d827206743cdba4a54e1b6c65098fd76212d6-1774x887.jpg?auto=format) ## We started at the catalog because everything else is attached to it. *Our story* - This didn’t start with a thesis about AI. It started with a problem we kept running into, across 15+ years and every retail company we built or ran. - We’ve run FMCG at Amazon across Italy and Spain, on more than €3B in P&L. We took Everli from 10 people to 300 across four countries and past €100M in revenue, on more than $100M raised. We built an electronics retailer from zero to €30M, the order management system behind Zalando’s Connected Retail, used by 10,000 European merchants, and the omnichannel layer inside Shopify Core. - Different companies. Same week. - A launch blocked on missing attributes. A marketplace rejecting a feed because a rule had changed. A new market opening with 100,000 products that needed accurate content in a language nobody on the team spoke. - Every time, the answer looked the same: spreadsheets, agencies, offshore teams, and someone still fixing things at 8pm. - Eventually we understood why. The catalog sits underneath everything. Search reads it. Ads depend on it. Marketplaces validate against it. Merchandising sorts by it. Shelf labels are generated from it. Every downstream system assumes the product record is correct, and quietly breaks when it isn’t. - Retail already spends billions on this complexity: PIMs to store the data, feed managers to distribute it, then agencies and offshore teams to do what the software can’t. The software was built to manage data, not to do the work. - Now there’s a new reader of that data: AI agents. They are starting to discover, compare and buy on behalf of consumers, and by 2030 they could mediate up to $5T in global retail sales. They don’t interpret messy data the way a person does. They read the record. - If the information isn’t there, the product isn’t just worse. To the agent, it may as well not exist. And every retailer is building that future on the catalog they have today. - So we built the layer that fixes it. Specialised agents that do the work of retail catalog operations, running on our platform, in production on customer catalogs, with a person accountable for every value they publish. - Not another tool to configure. Not an agency that delivers a project and walks away. - A product company. The agents are the product. And the Retail Engineers deploying them work alongside the engineers building them. ## Our principles *What we value* Six of them, and they are the ones we hire, review, and argue against. - Own the Outcome · You always own it, even when the AI did it. Delegating work never delegates responsibility for the result. - Feedback First · Feedback is the shortest path to the truth: go get it, and give it. - Craft · Average is free now. We are not. Know the difference between good and great, and close that gap. - Fast by Default · Fast where we can undo, careful where we cannot. - Low Ego · Fall in love with the outcome, never with the solution. - Customer First · Burn tokens, not trust. When our economics and the customer's value conflict, the customer wins. ## Three offices, one team. *Where we work* Three days a week in the office, minimum. Close enough to solve things in the same room, flexible enough for the rest of your life. - Milan · Italy - Barcelona · Spain - London · United Kingdom ## What we have announced. *News* - We have acquired Katalogo.ai and welcome Luca Cozzolino as Founding Chief Product Officer · 2026-06-04 · Company news - CommerceClarity raises €2.7M ($3.1M) pre-seed to power the AI agentic era of e-commerce · 2025-11-13 · Funding ## The work is the operation, not a feature We are hiring across engineering, product, and go to market, in Milan, Barcelona and London. Come and take one of these processes off a retailer's hands. [/careers](/careers) ## Questions ### What does CommerceClarity do? It builds and runs AI agents specialised on retail catalog work: enrichment, matching, translation, taxonomy, compliance checks and channel feeds. They run on CommerceClarity's own platform, on the retailer's own catalog. Dedicated Retail Engineers bring each workflow into production alongside the retailer's team, and a person is accountable for every value published. ### Who founded CommerceClarity? A founding team of retail operators. Federico Sargenti ran the FMCG categories at Amazon for Italy and Spain and was CEO of Everli from 2016 to 2023; Alessandro Angelini was its chief operating and financial officer, through a hundred million in revenue and a $100 million Series C; Daniele Vella built an electronics retailer from zero to thirty million; Michele Sampieri owns the engineering; Luca Cozzolino joined as Chief Product Officer when CommerceClarity acquired Katalogo.ai. ### Where is CommerceClarity based? Milan is the headquarters, with offices in Barcelona and London. ### Is CommerceClarity a PIM? No. It sits on top of the PIM, ERP and DAM a retailer already runs. A PIM stores attributes and a feed manager ships them; neither writes, checks or decides, and that is the work the agents do. ## Entities - [Amazon](https://www.amazon.com) · Organization, mentions - [Everli](https://www.everli.com) · Organization, mentions - [Zalando](https://corporate.zalando.com) · Organization, mentions - [Shopify](https://www.shopify.com) · Organization, mentions - [Product information management](https://en.wikipedia.org/wiki/Product_information_management) · Thing, mentions - Federico Sargenti · mentions - Alessandro Angelini · mentions - Daniele Vella · mentions - Michele Sampieri · mentions - Luca Cozzolino · mentions - [Agentic commerce](https://en.wikipedia.org/wiki/Agentic_commerce) · Thing, mentions - [AI agent](https://en.wikipedia.org/wiki/AI_agent) · Thing, mentions ## Sources - [The agentic commerce opportunity: how AI agents are ushering in a new era for consumers and merchants](https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-agentic-commerce-opportunity-how-ai-agents-are-ushering-in-a-new-era-for-consumers-and-merchants) · McKinsey & Company - [Everli, the European marketplace for online grocery shopping, bags $100M Series C](https://techcrunch.com/2021/03/29/everli/) · TechCrunch - [Agentic commerce](https://en.wikipedia.org/wiki/Agentic_commerce) · Wikipedia - [Product information management](https://en.wikipedia.org/wiki/Product_information_management) · Wikipedia --- # Join us and take the manual work out of retail. *Careers* CommerceClarity helps enterprise retailers and brands make the transition to AI, starting from the catalog. Come and do it with us. ![The Milan office.](https://cdn.sanity.io/images/xd4hrbt2/production/26baaf1251e89f69397a281249dead513c5059ac-1600x900.jpg?auto=format) ## We ran the operations we are now automating. *Who we are* - As a team, we’ve run FMCG at Amazon across Italy and Spain, taken Everli from 10 to 300 people and past €100M in revenue, built an electronics retailer from zero to €30M, built the order management system behind Zalando’s Connected Retail, and shipped the omnichannel layer inside Shopify Core. - We kept seeing the same thing: companies don’t struggle to sell more products. They struggle to keep the data behind those products correct. - A launch blocked by missing attributes. A marketplace rejecting a feed because a rule changed. A new market opening with 100,000 products that need accurate content in a language nobody on the team speaks. - This work is repetitive, rules-driven, and painfully manual. So we built agents that do it instead - and they’re already running on customer catalogs in production. **Who quote:** Join us and shape the next generation of Commerce. ## The agent layer for modern commerce. **Text** E-commerce is entering the AI Agentic Commerce Era, a shift where specialized AI agents collaborate with humans to plan, merchandise, price, and grow retail businesses faster and more intelligently. These agents will run key back-office tasks, learn from outcomes, and interact directly with shoppers, compressing the distance between product and purchase. **Text** While dozens of AI tools are emerging, what's missing is an intelligent network of commerce-specific agents that can work together, learn from each other, and plug into existing e-commerce systems without heavy implementation. This new coordination layer between humans, agents, and tools will reshape the way commerce is operated. That's where CommerceClarity comes in. **Text** We are building the agent layer for modern commerce: a connected network of AI agents, each specialized in tasks like catalog creation and assortment optimization at AI speed. Not just automating workflows, but enabling e-commerce teams to expand their offering, personalize it for every market or customer, and operate at a scale that was previously impossible. Orchestrated by a central brain that balances cost, speed, and quality. **Text** We started with the most painful and highest-leverage task, product catalogs, because that's where we've spent the past decade as operators. We have taken businesses from their very first SKU all the way to 200M€ in GMV, and we've also run large-scale Amazon operations exceeding 3Bn€+ in annual revenue. After wrestling with thousands of SKUs and the daily pricing and media firefight that comes with that scale, we know firsthand how slow, manual catalog workflows become the critical bottleneck. Catalog content remains one of the most time-consuming and under-optimized areas in retail, and one where automation drives immediate ROI. **Text** CommerceClarity eliminates that drag with an AI-run back office. Our Catalog Agent ingests supplier data, enriches it, writes product content in brand voice, and structures it for both SEO and LLM-native discovery. What once took weeks now takes minutes, slashing costs and supercharging traffic and conversion for our clients. **Text:** What once took weeks now takes minutes **Pull:** yes **Text:** And we're just getting started: every new agent we add builds on shared memory and cross-agent intelligence, compounding value for our customers. **Text** Just as Shopify abstracted storefront complexity, we're abstracting operational complexity. Not by replacing systems, but by augmenting them with always-on, self-optimizing AI agents that make teams faster, leaner, and more resilient. **Text** We believe the winners in this era won't be those who build the flashiest agent, but those who build the most useful and reliable network. One that enables people to do things that weren't possible before, while keeping them in control. CommerceClarity is that network. **Manifesto closing:** Join us and shape the next generation of commerce. **Manifesto signed name:** The founding team **Manifesto doc no:** No. 1 ## A team from the world’s most ambitious companies - Amazon · 23 · [https://cdn.sanity.io/images/xd4hrbt2/production/4f68393dd2b6b847d8a83c0e2a2bbee4cf35d580-603x182.svg?auto=format](https://cdn.sanity.io/images/xd4hrbt2/production/4f68393dd2b6b847d8a83c0e2a2bbee4cf35d580-603x182.svg?auto=format) - Shopify · 24 · [https://cdn.sanity.io/images/xd4hrbt2/production/0c2a88b051cefe609028eae5e74ee987bd6fff57-612x192.svg?auto=format](https://cdn.sanity.io/images/xd4hrbt2/production/0c2a88b051cefe609028eae5e74ee987bd6fff57-612x192.svg?auto=format) - Zalando · 20 · [https://cdn.sanity.io/images/xd4hrbt2/production/86977eab7f9d8dc52b36af2eb02534ff2a3a6486-132x24.svg?auto=format](https://cdn.sanity.io/images/xd4hrbt2/production/86977eab7f9d8dc52b36af2eb02534ff2a3a6486-132x24.svg?auto=format) - Meta · 21 · [https://cdn.sanity.io/images/xd4hrbt2/production/c5de271cbba587a6c1259f5aabd1dcc7e324b15a-948x191.svg?auto=format](https://cdn.sanity.io/images/xd4hrbt2/production/c5de271cbba587a6c1259f5aabd1dcc7e324b15a-948x191.svg?auto=format) - Everli · 26 · [https://cdn.sanity.io/images/xd4hrbt2/production/860fea3d3af29f1c8756cabbe3640855b6aba459-115x40.svg?auto=format](https://cdn.sanity.io/images/xd4hrbt2/production/860fea3d3af29f1c8756cabbe3640855b6aba459-115x40.svg?auto=format) - McKinsey · 23 · [https://cdn.sanity.io/images/xd4hrbt2/production/0083871114ffd2471bdb668d52f398e2fca0fd6a-360x45.svg?auto=format](https://cdn.sanity.io/images/xd4hrbt2/production/0083871114ffd2471bdb668d52f398e2fca0fd6a-360x45.svg?auto=format) - Bain & Company · 21 · [https://cdn.sanity.io/images/xd4hrbt2/production/ffdeb9f3c6ac17bef6b352ed5a198d1708e1f370-279x31.svg?auto=format](https://cdn.sanity.io/images/xd4hrbt2/production/ffdeb9f3c6ac17bef6b352ed5a198d1708e1f370-279x31.svg?auto=format) - TravelPerk · 30 · [https://cdn.sanity.io/images/xd4hrbt2/production/96db9ae32748a461415815e6c3da326cb5c79028-78x40.svg?auto=format](https://cdn.sanity.io/images/xd4hrbt2/production/96db9ae32748a461415815e6c3da326cb5c79028-78x40.svg?auto=format) - Musixmatch · 23 · [https://cdn.sanity.io/images/xd4hrbt2/production/7c17a157f41a78bfb72417924375eb26ec68f8aa-883x202.png?auto=format](https://cdn.sanity.io/images/xd4hrbt2/production/7c17a157f41a78bfb72417924375eb26ec68f8aa-883x202.png?auto=format) - Rover · 23 · [https://cdn.sanity.io/images/xd4hrbt2/production/0c5bd78d1193fa37193030d5bd6faef56847facd-423x134.png?auto=format](https://cdn.sanity.io/images/xd4hrbt2/production/0c5bd78d1193fa37193030d5bd6faef56847facd-423x134.png?auto=format) ## Our principles *What we value* Six of them, and they are the ones we hire, review, and argue against. - Own the Outcome · You always own it, even when the AI did it. Delegating work never delegates responsibility for the result. - Feedback First · Feedback is the shortest path to the truth: go get it, and give it. - Craft · Average is free now. We are not. Know the difference between good and great, and close that gap. - Fast by Default · Fast where we can undo, careful where we cannot. - Low Ego · Fall in love with the outcome, never with the solution. - Customer First · Burn tokens, not trust. When our economics and the customer’s value conflict, the customer wins. ## Three offices, one team. *Where we work* Three days a week in the office, minimum. We want the hard conversations to happen in the same room. - Milan · Italy - Barcelona · Spain - London · United Kingdom ## Open roles ## Open roles - [AI Engineer](/careers/ai-engineer) · Engineering - [AI Engineer, Internship](/careers/ai-engineer-internship) · Engineering - [Founding Head of Engineering](/careers/founding-head-of-engineering) · Engineering - [Platform Engineer](/careers/platform-engineer) · Engineering - [Agent Product Manager](/careers/agent-product-manager) · Product - [Business Development Representative, Italy](/careers/business-development-representative-italy) · Sales & Partnerships - [Business Development Representative, UK](/careers/business-development-representative-uk) · Sales & Partnerships - [Enterprise Account Executive, Germany](/careers/enterprise-account-executive-germany) · Sales & Partnerships - [Enterprise Account Executive, Italy](/careers/enterprise-account-executive-italy) · Sales & Partnerships - [Enterprise Account Executive, UK](/careers/enterprise-account-executive-uk) · Sales & Partnerships - [Forward Deployed Engineer, Italy](/careers/forward-deployed-engineer-italy) · Service & Delivery - [Forward Deployed Engineer, UK](/careers/forward-deployed-engineer-uk) · Service & Delivery - [Retail Engineer (Deployment Strategist), Germany](/careers/retail-engineer-germany) · Service & Delivery - [Retail Engineer (Deployment Strategist), Italy](/careers/retail-engineer-italy) · Service & Delivery - [Retail Engineer (Deployment Strategist), UK](/careers/retail-engineer-uk) · Service & Delivery --- # Why retail teams run on CommerceClarity Built for the catalogs the team cannot keep up with. Book a demo and we will run it on one of your real feeds. **Trust caption:** Trusted by catalog teams ## Logos - Prada - Arcaplanet - Dr. Max - Cisalfa Sport - Prénatal ![](https://cdn.sanity.io/images/xd4hrbt2/production/0ad6be73d83212ee0b3a465bf3555e1bd95885b6-1400x1600.jpg?auto=format) --- # Make products readable by shopping agents. > Agentic Commerce is the CommerceClarity area that publishes a catalog to the surfaces autonomous shopping agents read. Two agents run here: ChatGPT Feed, built to OpenAI’s product feed spec, and Google Feed, which carries the conversational attributes Google’s AI surfaces consume. The work is attribute completeness rather than page design: an agent buying on a shopper’s behalf reads data, not layout. Read it as a transaction protocol rather than a traffic channel: where SEO and GEO/AEO end with a person browsing the site, an agent completes the purchase over an API with no page in the flow. Agents that publish your catalog where autonomous shopping agents can read it and buy from it: OpenAI’s product spec for ChatGPT, Google Merchant with the conversational attributes its AI surfaces consume, and whatever surface comes after them. ![A person at home in the evening speaking to their phone, the screen lighting their face in a dark room.](https://cdn.sanity.io/images/xd4hrbt2/production/387cbc4e2c0d4d7e621146fa2522902d990e4314-1774x887.jpg?auto=format) [Book a demo](/book-a-demo) ## A new buyer, with no eyes A shopping agent does not browse. It [reads a feed](/blog/chatgpt-product-feed-required-fields), compares structured attributes and completes a purchase, and it never sees the page you designed. Everything built to persuade a person, the photography, the layout, the copy above the fold, is invisible to it. It is worth naming what this is not. SEO and GEO/AEO both end with a person on your site: they win a click, or a citation that leads to one, and the human browses, compares and checks out. This does not. There is no session, no category page, no basket anyone built. It is a transaction protocol running over an API, and a catalog can be good at bringing people in while being unreadable to it. What it does read is a specification, and the requirements are somebody else’s. OpenAI publishes a product spec with its own attribute set and its own update cadence. Google wants the classic Shopping feed and, separately, the conversational attributes its AI surfaces consume. Neither one is your PIM’s schema. This is also the channel where being early is cheap and being late is not. The feed either validates and republishes on cadence or your products are not in the answer, and there is no ranking to slip slowly down: you are in the consideration set, or you are not in it at all. ## The jobs on the feed A few of the ones that run here, and the list grows every time a surface does. Each is a single workflow with its own output and its own numbers. Publish to one surface, or to all of them. - ChatGPT Feed · Publish your catalog to ChatGPT shopping. · Continuous - Google Feed · Publish to Google Merchant: the Shopping feed plus the conversational attributes Google’s AI shopping needs. · Continuous ## One pipeline, every surface that reads a feed *Publishing to the agents* For Google, both layers run on one pipeline. The classic mapping onto its required attributes outputs a feed that passes validation for Shopping ads and free listings. The conversational attributes are the second layer, the ones its AI surfaces read in order to recommend a product and complete a purchase. For ChatGPT, the same catalog is transformed to OpenAI’s product spec, its attribute set and its update cadence, then validated and republished on that cadence. Both run off the same source record, so the feed that ranks and the feed that gets recommended stay in step. And when a spec moves, which these specs are young enough to keep doing, we maintain the mapping. ## What retailers ask us ### What is agentic commerce? Buying done by software acting for a person, instead of by the person browsing. Read it as a transaction protocol rather than a traffic channel: the agent reads a published product spec, compares, and completes the purchase, and none of that happens on your site. It changes what a product record is for. A shopper forgives a missing attribute and looks at the photo. An agent comparing options cannot. What is not in the data does not exist, and the product is not in the comparison. ### How is agentic commerce different from SEO and GEO/AEO? SEO and GEO/AEO both end with a person. They win a click, or a citation that leads to one, and the human then browses, compares and checks out on your site. Agentic commerce ends without a person. The agent never lands on a page. It reads a published product spec over an API, compares what it can read, and completes the purchase inside the conversation. No session, no category page, no basket anyone built. So the two are not versions of the same work. One improves what a human finds. The other decides whether a machine can transact with you at all, and a catalog can be good at the first while being invisible to the second. ### Can you give me an example of agentic commerce? A person tells an assistant to reorder the dog food they buy, but cheaper, same protein, in the large bag. The assistant does not open a category page. It queries the products it can read, filters on protein source and pack size, checks price and availability, and completes the purchase. Every step of that runs on attributes. A record that says large bag without the weight, or that never states the protein, drops out of the comparison silently, and nobody at the retailer ever learns it happened. ### How is this different from the Google feed we already have? The Shopping feed exists to win a click that lands on your product page. The conversational layer exists so the agent never needs your product page. Same catalog underneath, different job. One is built to pass ad validation. The other carries the attribute set the AI surfaces read in order to recommend a product and complete a purchase, and a feed built only for Shopping ads does not carry it. ### Is ChatGPT shopping worth a workflow today? That is your call on timing, and it is a cheap one to take early. It is the same product knowledge mapped onto a published spec, so the work is the mapping and the cadence, not a new content project. What makes early cheap is the asymmetry. A shopper who did not find you today can come back tomorrow. An agent that could not read you did not include you in a comparison that is already over, and nobody tells you it happened. ### What happens when the spec changes? We maintain the mapping. This is one place where agentic commerce is easier than SEO rather than harder: a ranking algorithm is a black box you infer from results, while these are published, versioned specifications you can read. The catch is that they are young and they move. That is the argument for running this continuously instead of exporting once and calling it done. ### Do we lose control of how we are presented? Some of it, and it is worth being precise about which. With no page in the flow you lose the levers that lived on the page: photography, layout, badges, where the reviews sit, the copy that persuades. What is left is the data, and the data is the whole of what the agent reasons on. You control the input. You do not control the comparison, and that is true of any marketplace. ### Does this replace SEO? No, and they do not compete for the same visit. SEO and GEO/AEO bring a person to the site. Agentic commerce is a transaction that completes without one. The overlap is the product knowledge underneath, which is why these areas share it. ### Whose data is it? Yours. The corrections your team makes are what the agents learn from, and both the data and the corrections stay exportable at any point. ## Related areas - [SEO & GEO/AEO](/solutions/seo-geo) - [Channels & Markets](/solutions/channels-markets) ## Proven by - [Nestlé Purina](/customers/purina) · How Purina got its catalog ready to be recommended by AI agents ## Every correction makes the next run better *The feedback loop* Your team judges the output. A value is right, or it is wrong and gets fixed once. Each verdict changes the next run: the score on your own sample moves, and the agent starts from what the last one learned. So the share of values that need a person keeps falling, measured on your catalog and your rules rather than on an average of everybody else. ![The platform flagging a product description as too generic for the brand tone, with the suggested rewrite and the two answers.](https://cdn.sanity.io/images/xd4hrbt2/production/30815d23ee4d132352d037d84d401a32a7b48eec-2880x1620.webp?auto=format) ## Live on your catalog in weeks, not months *How we work* We run all four phases with your team, on the stack you already have, and each one ends with a number you agreed in advance. - Design · We map how your data comes in and how the result goes back out, then agree the business KPI the agent has to move. - Implement · We build the agent and test it on a sample of your own catalog, agreed with you. Every change is scored against that sample, so the tuning runs on numbers. - Prove · Done means the KPI reached target on the full catalog, and you have seen the numbers yourself. - Evolve · We stay on the account. Your corrections keep improving the agent, the KPI holds where you need it, and the next use case starts from the context this one already built. ## Enterprise, on a catalog you cannot break ![A department store floor.](https://cdn.sanity.io/images/xd4hrbt2/production/251441053d31b02d7111cf86ee1083ac6fd8f061-1600x1000.jpg?auto=format) [Book a demo](/book-a-demo) - It runs on the stack you have · It reads your PIM, ERP and DAM through their APIs and writes the result back to them. - Nothing uncertain reaches a channel · Every value is checked against your rules, and anything uncertain goes to review before it reaches a channel. - Your data stays yours · Your own isolated workspace. It is never pooled into shared models, and you can export it at any point. - Every run is logged · Each value carries its source, its run and its score, so you can see why a wrong one went wrong. ## Start with one use case One use case, run end to end on your own catalog. That is enough to see what it moves and what it costs per product, before you commit the rest. [Book a demo](/book-a-demo) ## Entities - [OpenAI](https://openai.com) · Organization, about - [Google](https://www.google.com) · Organization, about - [ChatGPT](https://chatgpt.com) · Product, mentions ## Sources - [Commerce documentation](https://developers.openai.com/commerce/) · OpenAI - [Product data specification](https://support.google.com/merchants/answer/7052112) · Google Merchant Center - [Product structured data](https://developers.google.com/search/docs/appearance/structured-data/product) · Google Search Central - [Product](https://schema.org/Product) · Schema.org --- # Manage the whole product catalog, not a sample of it. > Catalog Foundation is the CommerceClarity area that builds the product records every other area reads from. Six agents run here: Data Enrichment, Matching, Variant Grouping, ERP Codification, Catalog Migrator and Product Image Generation. They read the supplier feed, the PIM, the ERP and the DAM without replacing any of them, hold every value to the retailer’s own attribute model, and write the finished record back to the system of record. Agents that build the product records everything else reads from. SEO, translation and marketplace work all pull from the same fields, so this is where the work starts. ![A long-exposure shot down a store aisle.](https://cdn.sanity.io/images/xd4hrbt2/production/e447fd9addb331c4f03a57faaa79e6fcae3f6c98-1774x887.jpg?auto=format) [Book a demo](/book-a-demo) ## The catalog is the constraint A product record starts life in three or four systems at once. An ERP with codes and no copy. Supplier files in whatever shape the supplier felt like sending. A PIM built to hold specifications rather than to sell them. A spreadsheet holding the hand-offs together, maintained by the one person who understands it. None of that is wrong, and replacing it is not the job. The problem is narrower: the fields that decide whether a product is found and bought, fit, material, dimensions, the variant it belongs to, the category it sits in, are the fields no system owns. So they get typed in by hand, once per channel and once per market, by the team that is already the constraint. The cost shows up in assortment. Every channel and every market that needs the same fields retyped is a channel that opens later, with less of the catalog on it. Everything above this layer reads from the same records, so fixing the record once is what makes the other seven areas cheap. ## The jobs on the record A few of the ones that run here. Each is a single workflow with its own output and its own numbers, and a job you do not see named is usually a variant of one that is. Run one, or run the set. ### Data Enrichment Auto-enrich product pages from your own or supplier data, recovering missing specs online. Builds publishable product pages from your own data or your suppliers, and recovers the specs they left out from sources online. time per product Continuous ### Data Sourcing Find the product data you do not have, from outside sources, with the evidence behind every value. Goes outside your own systems for the values nobody can fill in, reads the manufacturer and the named sources, arbitrates where they disagree, and returns each attribute with its evidence and a confidence. attributes left blank Continuous ### Matching Match products across catalogs and marketplaces: offers to canonical, SKUs to listings, dedup. Decides whether an incoming record is a product you already carry, links it to the canonical entry, flags duplicates. duplicate rate Continuous ### Variant Grouping Turn flat SKUs into parent products with their variants linked. Turns a flat SKU list into a family: one parent with the shared content, the children linked to it. PDPs per product Continuous ### ERP Codification Codify new supplier products into your ERP automatically. Produces the codification your ERP needs before a supplier product reaches the PIM: code, dimensions, weight, packaging. days to sellable Continuous ### Catalog Migrator Migrate your catalog to a new platform, Magento to Shopify and similar. Maps an existing catalog onto a new platform schema, and surfaces the gaps that need a merchant decision. weeks of replatforming One-shot ## A supplier file in, a sellable product out *Product onboarding* Twenty suppliers, twenty formats: an XML feed, a CSV dropped on SFTP, an API, a spreadsheet somebody maintains by hand. Every field is mapped onto your own attribute model and normalized on the way in: currency on price, category paths flattened, image URLs rewritten, and a field with no home is flagged rather than quietly dropped. The specs the supplier left blank, fit, material, compatibility, get read off the datasheet and the supplier’s own text, then recovered from sources online and cross-checked. Every value carries where it came from. ## What retailers ask us ### What is product catalog management? Keeping every product record complete, consistent and current, on every channel that reads it. In a small catalog it is a job one person finishes. Past a few thousand SKUs, with suppliers sending twenty different formats, nobody finishes it. That is where it stops being data entry and becomes an operation. ### Who runs the agents, us or you? We do, on your data, category by category, and we stay on the account after go-live. What you are buying is the outcome on your catalog, not a licence and a manual. ### Do we have to replace our PIM or ERP? No. The agents read your PIM, ERP and DAM and write back to them. We sit on top of the stack you already run, and mapping how data enters and leaves is the first thing we do. ### What if the agent we need does not exist yet? Most of these already run for another retailer and adapt to your catalog, and that adaptation is part of the service. Where a workflow is not covered, we design the agent with you and scope it separately. ### How do you know the output is right? The KPI, agreed before any build, says whether the business moved on the real catalog. Underneath it we score the work itself against your rules on a sample of your catalog, and that score re-runs on every change. ### Whose data is it? Yours. The corrections your team makes are what the agents learn from, and both the data and the corrections stay exportable at any point. ## Related areas - [Channels & Markets](/solutions/channels-markets) - [Compliance & Quality](/solutions/compliance-quality) ## Proven by - [Cisalfa Group](/customers/cisalfa-group) · How Cisalfa Group got its catalog ready for the new season without growing the team ## Every correction makes the next run better *The feedback loop* Your team judges the output. A value is right, or it is wrong and gets fixed once. Each verdict changes the next run: the score on your own sample moves, and the agent starts from what the last one learned. So the share of values that need a person keeps falling, measured on your catalog and your rules rather than on an average of everybody else. ![The platform flagging a product description as too generic for the brand tone, with the suggested rewrite and the two answers.](https://cdn.sanity.io/images/xd4hrbt2/production/30815d23ee4d132352d037d84d401a32a7b48eec-2880x1620.webp?auto=format) ## Live on your catalog in weeks, not months *How we work* We run all four phases with your team, on the stack you already have, and each one ends with a number you agreed in advance. - Design · We map how your data comes in and how the result goes back out, then agree the business KPI the agent has to move. - Implement · We build the agent and test it on a sample of your own catalog, agreed with you. Every change is scored against that sample, so the tuning runs on numbers. - Prove · Done means the KPI reached target on the full catalog, and you have seen the numbers yourself. - Evolve · We stay on the account. Your corrections keep improving the agent, the KPI holds where you need it, and the next use case starts from the context this one already built. ## Enterprise, on a catalog you cannot break ![A department store floor.](https://cdn.sanity.io/images/xd4hrbt2/production/251441053d31b02d7111cf86ee1083ac6fd8f061-1600x1000.jpg?auto=format) [Book a demo](/book-a-demo) - It runs on the stack you have · It reads your PIM, ERP and DAM through their APIs and writes the result back to them. - Nothing uncertain reaches a channel · Every value is checked against your rules, and anything uncertain goes to review before it reaches a channel. - Your data stays yours · Your own isolated workspace. It is never pooled into shared models, and you can export it at any point. - Every run is logged · Each value carries its source, its run and its score, so you can see why a wrong one went wrong. ## Start with one use case One use case, run end to end on your own catalog. That is enough to see what it moves and what it costs per product, before you commit the rest. [Book a demo](/book-a-demo) ## Entities - [Akeneo](https://www.akeneo.com) · Organization, mentions - [Salsify](https://www.salsify.com) · Organization, mentions - [GS1](https://www.gs1.org) · Organization, mentions ## Sources - [GTIN, the GS1 global trade item number](https://www.gs1.org/standards/id-keys/gtin) · GS1 - [GDSN, the Global Data Synchronisation Network](https://www.gs1.org/standards/gdsn) · GS1 - [Product data specification](https://support.google.com/merchants/answer/7052112) · Google Merchant Center - [Product](https://schema.org/Product) · Schema.org --- # Feed every channel without rebuilding the data. > Channels and Markets is the CommerceClarity area that reshapes one product record for the place it is going. Three agents run here: Content Translation, Amazon Listing Optimizer and Leroy Merlin Listing Optimizer. Amazon is treated as one workflow per country rather than one integration, because the schema, the mandatory fields and the search differ by marketplace. The agents sit alongside an existing feed manager rather than replacing it. Agents that reshape the same product for the place it is going: translated into the languages you sell in, and rewritten to each marketplace’s schema and its search. Amazon is not one channel, it is one per country, and so is everything else here. ![A city shopping street at dusk, lit shop fronts and people crossing, caught on a slow shutter.](https://cdn.sanity.io/images/xd4hrbt2/production/a7c282b72199fcaf154a37d13e4aaef25e16de26-1774x887.jpg?auto=format) [Book a demo](/book-a-demo) ## The connector works. The data is what fails validation. The plumbing is rarely what stalls a launch. The connector exists, the feed goes out, the integration works. What stops it is the data arriving at the other end: fields the channel requires and your catalog does not carry, values in a format the operator rejects, one mandatory attribute nobody owns. And a channel is not one rulebook. Amazon Italy and Amazon Germany carry different attribute schemas, different validation and different operators, which makes them two jobs rather than one job run twice. That is a lesson from a live delivery, not a design preference: the rules cannot be pasted from one country into the next, so the template is studied country by country. The cost of that shows up as time, and not in the integration: it goes into producing data that fits the template. On the retailer’s side the same gap looks smaller and hurts more: a product with good copy and good specs that still cannot be created, because one required field is missing. ## The jobs on the channel A few of the ones that run here. Amazon alone is one workflow per country, so the real list is longer than the cards below, and a channel we have not named is a template to study rather than a new product. Each has its own output and its own numbers. - Content Translation · Translate your product content into your other languages. · Continuous - Amazon Listing Optimizer · Optimize your Amazon listings for Amazon search: keywords, titles, bullets, backend. Live on IT, UK, FR, DE and ES. · Continuous - Leroy Merlin Italy Listing Optimizer · Optimize your Leroy Merlin listings to pass validation and convert. · Continuous ## Passing validation, and reading like it was written there *Getting live on a channel* Two kinds of work land on the same product. The listing is mapped to that channel’s own schema for that country: mandatory attributes generated rather than left blank, values normalized into the formats the operator accepts, single-select fields like category resolved deterministically, and titles, bullets and backend keywords written for that marketplace’s search rather than lifted off your product page. The language is the second job: the content you already hold, taken into the markets you sell in against your own glossary and tone, so a launch never waits in a translation queue. Whatever genuinely cannot be resolved comes back as a decision before upload, because a listing that fails validation costs more than one that waits a day. ## What retailers ask us ### What is product feed management? Taking one product record and reshaping it for each channel that wants it. No two channels want the same thing: each has its own required fields, its own category tree, its own rules about what a title may contain. Feed management is satisfying all of them from one source, and noticing when one of them changes the rules. ### Is Amazon one integration, or one per country? One per country, and that is deliberate. Each marketplace country carries its own attribute schema, validation rules and operators, so each is its own workflow with its own template. Sharing one template across countries is how a listing passes in the first and fails in the rest. ### Do you replace our feed manager or our integration? No. Whatever moves the data keeps moving it. We work on what it carries, so the record that arrives at the channel already fits that channel’s schema. ### Is this translation or rewriting? Both, and they are separate jobs on purpose. Translation takes the content you already have into your other languages. A marketplace listing is rewritten for that marketplace’s rules and its search, so the sentences are different ones, written for that channel. ### What happens when a channel changes its rules? The template is per country and maintained as the rules move. That is the part we stay on after go-live, and it is the reason this runs continuously rather than as a migration you do once. ### What about mandatory fields we simply do not have? They get derived or recovered where that is possible, and flagged as a decision where it is not. Nothing is filled with a plausible guess to clear validation, because that defect comes back later as a return or a takedown. ### Whose data is it? Yours. The corrections your team makes are what the agents learn from, and both the data and the corrections stay exportable at any point. ## Related areas - [Catalog Foundation](/solutions/catalog-foundation) - [Agentic Commerce](/solutions/agentic-commerce) ## Proven by - [STIGA](/customers/stiga) · How STIGA launched on seven marketplaces in four months - [Yeppon](/customers/yeppon) · How Yeppon got a 150,000 SKU catalog marketplace-ready, from its own store outward ## Every correction makes the next run better *The feedback loop* Your team judges the output. A value is right, or it is wrong and gets fixed once. Each verdict changes the next run: the score on your own sample moves, and the agent starts from what the last one learned. So the share of values that need a person keeps falling, measured on your catalog and your rules rather than on an average of everybody else. ![The platform flagging a product description as too generic for the brand tone, with the suggested rewrite and the two answers.](https://cdn.sanity.io/images/xd4hrbt2/production/30815d23ee4d132352d037d84d401a32a7b48eec-2880x1620.webp?auto=format) ## Live on your catalog in weeks, not months *How we work* We run all four phases with your team, on the stack you already have, and each one ends with a number you agreed in advance. - Design · We map how your data comes in and how the result goes back out, then agree the business KPI the agent has to move. - Implement · We build the agent and test it on a sample of your own catalog, agreed with you. Every change is scored against that sample, so the tuning runs on numbers. - Prove · Done means the KPI reached target on the full catalog, and you have seen the numbers yourself. - Evolve · We stay on the account. Your corrections keep improving the agent, the KPI holds where you need it, and the next use case starts from the context this one already built. ## Enterprise, on a catalog you cannot break ![A department store floor.](https://cdn.sanity.io/images/xd4hrbt2/production/251441053d31b02d7111cf86ee1083ac6fd8f061-1600x1000.jpg?auto=format) [Book a demo](/book-a-demo) - It runs on the stack you have · It reads your PIM, ERP and DAM through their APIs and writes the result back to them. - Nothing uncertain reaches a channel · Every value is checked against your rules, and anything uncertain goes to review before it reaches a channel. - Your data stays yours · Your own isolated workspace. It is never pooled into shared models, and you can export it at any point. - Every run is logged · Each value carries its source, its run and its score, so you can see why a wrong one went wrong. ## Start with one use case One use case, run end to end on your own catalog. That is enough to see what it moves and what it costs per product, before you commit the rest. [Book a demo](/book-a-demo) ## Entities - [Amazon](https://www.amazon.com) · Organization, about - [Leroy Merlin](https://www.leroymerlin.com) · Organization, about - [Google](https://www.google.com) · Organization, mentions ## Sources - [Product data specification](https://support.google.com/merchants/answer/7052112) · Google Merchant Center - [Supplemental feeds](https://support.google.com/merchants/answer/6324461) · Google Merchant Center - [GTIN, the GS1 global trade item number](https://www.gs1.org/standards/id-keys/gtin) · GS1 - [Product](https://schema.org/Product) · Schema.org --- # Product compliance checked before you publish, not after. > Compliance and Quality is the CommerceClarity area that checks a catalog against one written rulebook. Two agents run here: Catalog Auditing, which scores the whole catalog on cadence and tells you where you stand, and Catalog Monitoring, which watches the live catalog and raises a defect as it appears. The same instrument covers editorial rules the retailer sets and regulation it does not, including the EU General Product Safety Regulation and EPREL energy labelling. One rulebook, and agents to run it: auditing the whole catalog against it to tell you where you stand, watching the live one to raise a defect as it appears. Your editorial rules and the regulation you do not get to negotiate, checked with the same instrument. ![An office at night seen through a window, someone working at a desk below a wall of pinned reports.](https://cdn.sanity.io/images/xd4hrbt2/production/02bf4bd1a0d00df4c9777a932a5331673e5e6518-1774x887.jpg?auto=format) [Book a demo](/book-a-demo) ## Nobody reads a hundred thousand products Quality in a catalog is a sampling problem rather than a review step. A person can check a hundred products properly. The rules that matter, brand tone, a mandatory statement, a claim you are not allowed to make, apply to all of them, and the ones nobody checked are the ones that reach a customer. And the rulebook is two rulebooks. Yours: tone of voice, glossary, translation rules, HTML and SEO standards, what the brand may and may not say. And the one you were given: anti-greenwashing wording, category regulation, the Digital Product Passport arriving for whole categories at once. Most retailers have the second written down and not the first, which is its own finding. The catalog also does not hold still once it is published. A marketplace quietly stops showing a product. A third party overwrites your image. An attribute drifts away from the system of record, a feed breaks and nothing says so. Each of those is found by a customer or by a buyer, unless something is watching for it. ## The jobs on quality A few of the ones that run here. Some run on cadence across the whole catalog, some continuously on the live one, and each has its own output and its own numbers. - Catalog Auditing · Audit the catalog against your brand rules and the regulation that applies. · One-shot - Catalog Monitoring · Watch the live catalog and raise a defect before your customers find it. · Continuous ## One rulebook, run on cadence and watched live *Keeping the catalog clean* Your rules stop being a document and become an evaluation set: each rule scored across the whole catalog, per rule and per product, conclusion first. What comes back is where you stand, which rule breaks most often and which products break it, rather than ten thousand warnings. The same run covers the rules you wrote and the ones you were given, so anti-greenwashing wording, a missing mandatory statement and brand tone are all checked with one instrument. Then the same rulebook stays on the live catalog: whether products are still live where they should be, whether attributes still agree with your source of truth, whether an image has been swapped by a third party or a description has drifted, whether the feeds are still going out. History comes first, so a change has a before. And where a rule is one you cannot afford to get wrong, it holds the value rather than flagging it. ## What retailers ask us ### What is product compliance? Every product record carrying what the market it sells in requires: declarations, labels, restricted substances, documentation. For a retailer the hard part is rarely reading the regulation. It is that every record has to carry the right fields, the rules differ by market, and nobody can check a catalog that size by hand. ### Do we have to write our rules down first? Ideally, and in practice often not. Where a formal guideline exists we run it as it is. Where it does not, the first pass is extracting the rules your team is already applying by hand, so that there is something to test against. ### Is this a report or a fix? The audit is a measurement, deliberately. It tells you how much of the catalog respects the rules and where it does not. Fixing is the other agents’ job, and keeping the two apart is what makes the score worth trusting. ### How is a regulatory check different from a brand check? Only in who wrote the rule. Both run as the same kind of test. The difference is the consequence, which is why regulatory rules are usually set to hold a value rather than to flag it. ### What does the monitoring actually watch? Four things: whether products are still live where they should be, whether attributes still agree with your source of truth, whether images or descriptions have been changed by someone else, and whether the feeds are still going out. Each one is checked against its own history, so a change has a before. ### Can it stop something from publishing? Yes, where you want it to. A rule can flag for review or hold the value outright, and the rules you cannot afford to get wrong are normally set to hold. ### Whose data is it? Yours. The corrections your team makes are what the agents learn from, and both the data and the corrections stay exportable at any point. ## Related areas - [Catalog Foundation](/solutions/catalog-foundation) - [Physical Retail](/solutions/physical-retail) ## Proven by - [Caddy’s](/customers/caddys) · How Caddy’s made compliance a property of its product data ## Every correction makes the next run better *The feedback loop* Your team judges the output. A value is right, or it is wrong and gets fixed once. Each verdict changes the next run: the score on your own sample moves, and the agent starts from what the last one learned. So the share of values that need a person keeps falling, measured on your catalog and your rules rather than on an average of everybody else. ![The platform flagging a product description as too generic for the brand tone, with the suggested rewrite and the two answers.](https://cdn.sanity.io/images/xd4hrbt2/production/30815d23ee4d132352d037d84d401a32a7b48eec-2880x1620.webp?auto=format) ## Live on your catalog in weeks, not months *How we work* We run all four phases with your team, on the stack you already have, and each one ends with a number you agreed in advance. - Design · We map how your data comes in and how the result goes back out, then agree the business KPI the agent has to move. - Implement · We build the agent and test it on a sample of your own catalog, agreed with you. Every change is scored against that sample, so the tuning runs on numbers. - Prove · Done means the KPI reached target on the full catalog, and you have seen the numbers yourself. - Evolve · We stay on the account. Your corrections keep improving the agent, the KPI holds where you need it, and the next use case starts from the context this one already built. ## Enterprise, on a catalog you cannot break ![A department store floor.](https://cdn.sanity.io/images/xd4hrbt2/production/251441053d31b02d7111cf86ee1083ac6fd8f061-1600x1000.jpg?auto=format) [Book a demo](/book-a-demo) - It runs on the stack you have · It reads your PIM, ERP and DAM through their APIs and writes the result back to them. - Nothing uncertain reaches a channel · Every value is checked against your rules, and anything uncertain goes to review before it reaches a channel. - Your data stays yours · Your own isolated workspace. It is never pooled into shared models, and you can export it at any point. - Every run is logged · Each value carries its source, its run and its score, so you can see why a wrong one went wrong. ## Start with one use case One use case, run end to end on your own catalog. That is enough to see what it moves and what it costs per product, before you commit the rest. [Book a demo](/book-a-demo) ## Entities - [General Product Safety Regulation](https://eur-lex.europa.eu/eli/reg/2023/988/oj) · Thing, about - [EPREL](https://eprel.ec.europa.eu/screen/home) · Thing, about - [GS1](https://www.gs1.org) · Organization, mentions ## Sources - [Regulation (EU) 2023/988 on general product safety](https://eur-lex.europa.eu/eli/reg/2023/988/oj) · EUR-Lex - [European Product Registry for Energy Labelling (EPREL)](https://eprel.ec.europa.eu/screen/home) · European Commission - [Regulation (EU) 2017/1369 framework for energy labelling](https://eur-lex.europa.eu/eli/reg/2019/2020/oj) · EUR-Lex - [Product data specification](https://support.google.com/merchants/answer/7052112) · Google Merchant Center --- # Turn product data into better commercial decisions. > Merchandising is the CommerceClarity area that presents one product record more than one way. Three agents run here: Bundle Builder, which assembles what sells together out of the catalog itself, Buyer-persona Content, which writes the version each audience needs, and Product Infographics. Every variant is generated from structured product knowledge the retailer already holds rather than from a creative brief. One product record carries more than one story. These agents write the version each audience actually needs and assemble what goes with what, from the structured knowledge already in the record rather than from a brief. ![A lit shop window at night, a dresser reaching up among the mannequin forms, the wet pavement in front of it.](https://cdn.sanity.io/images/xd4hrbt2/production/182386d7cb473847f80be5af4ce8c21bf58f9e0b-1774x887.jpg?auto=format) [Book a demo](/book-a-demo) ## One description, for everybody who buys it Most catalogs carry exactly one version of each product’s story, written for a general shopper who does not exist. A technical buyer wants the specification and the compatibility. A first-time buyer wants to know whether it is the right one and what else they will need. A procurement buyer wants neither, and reads past both. Merchandising is where that costs money rather than clarity. The bundle nobody built, the outfit nobody assembled, the specification that never became a panel a shopper could scan in three seconds: each one is order value left on the table, and each one is manual work that only ever gets done for the top of the range. The knowledge is already sitting in the record. Materials, dimensions, compatibility, what pairs with what. What is missing is the labour of turning it into more than one presentation, per product, across an assortment nobody has time to work through twice. ## The jobs on presentation A few of the ones that run here. Each is a single workflow with its own output and its own numbers, built from product knowledge you already hold rather than from a brief. Run one, or run the set. - Bundle Builder · Auto-create bundles, outfits and kits to grow basket size. · Continuous - Buyer-persona Content · Write per-audience versions of the same product copy. · Continuous - Product Infographics · Generate infographics and enhanced images from your product data. · Continuous ## The same record, presented more than once *Moving order value* You define the audiences you actually sell to, and the copy is written for each of them off the same record: the technical read for the buyer comparing specifications, the plain one for the first-timer who needs to know it fits, the procurement version that leads on terms rather than on lifestyle. Every version is written against your own tone of voice and glossary. The same knowledge decides what belongs with what, so bundles, outfits and kits are proposed out of your own assortment with the merchandising rationale and the margin logic attached. And the specifications a shopper will never read as a paragraph get laid out to be scanned. A merchandiser reviews the proposals and approves what publishes. ## What retailers ask us ### What is ecommerce merchandising? Deciding what to show, to whom, and in what order. The decisions are commercial, but they run on product data. A bundle has to know what goes with what. A description written for one kind of buyer needs attributes to write from. A category page needs its products actually in the category. Poor data caps how good the merchandising can get. ### Is this personalization? Not in the tracking sense. There is no profile and no cookie. These are versions of the copy tied to audiences you define, and they can be served by segment, by channel or by store. Same product, different story. ### Do we have to maintain three descriptions per product? No. You maintain the audiences and the rules. The versions are generated from the record, so a change to the product updates every version at once. ### How does a bundle get decided? From your own assortment and your own margin rules, with the merchandising rationale and the pricing logic attached, so a merchandiser reviews a proposal rather than assembling it. ### Does more copy hurt our SEO? Only if the versions are near-duplicates of each other on the same URL. They are written for different audiences and different surfaces, and where two would collide the canonical stays with one of them. ### Where do we start? With one category and the audiences you can already name, usually the two that argue with each other in a buying meeting. It is a cheap place to see whether the copy moves anything before it goes wider. ### Whose data is it? Yours. The corrections your team makes are what the agents learn from, and both the data and the corrections stay exportable at any point. ## Related areas - [SEO & GEO/AEO](/solutions/seo-geo) - [Search & Discovery](/solutions/search-discovery) ## Every correction makes the next run better *The feedback loop* Your team judges the output. A value is right, or it is wrong and gets fixed once. Each verdict changes the next run: the score on your own sample moves, and the agent starts from what the last one learned. So the share of values that need a person keeps falling, measured on your catalog and your rules rather than on an average of everybody else. ![The platform flagging a product description as too generic for the brand tone, with the suggested rewrite and the two answers.](https://cdn.sanity.io/images/xd4hrbt2/production/30815d23ee4d132352d037d84d401a32a7b48eec-2880x1620.webp?auto=format) ## Live on your catalog in weeks, not months *How we work* We run all four phases with your team, on the stack you already have, and each one ends with a number you agreed in advance. - Design · We map how your data comes in and how the result goes back out, then agree the business KPI the agent has to move. - Implement · We build the agent and test it on a sample of your own catalog, agreed with you. Every change is scored against that sample, so the tuning runs on numbers. - Prove · Done means the KPI reached target on the full catalog, and you have seen the numbers yourself. - Evolve · We stay on the account. Your corrections keep improving the agent, the KPI holds where you need it, and the next use case starts from the context this one already built. ## Enterprise, on a catalog you cannot break ![A department store floor.](https://cdn.sanity.io/images/xd4hrbt2/production/251441053d31b02d7111cf86ee1083ac6fd8f061-1600x1000.jpg?auto=format) [Book a demo](/book-a-demo) - It runs on the stack you have · It reads your PIM, ERP and DAM through their APIs and writes the result back to them. - Nothing uncertain reaches a channel · Every value is checked against your rules, and anything uncertain goes to review before it reaches a channel. - Your data stays yours · Your own isolated workspace. It is never pooled into shared models, and you can export it at any point. - Every run is logged · Each value carries its source, its run and its score, so you can see why a wrong one went wrong. ## Start with one use case One use case, run end to end on your own catalog. That is enough to see what it moves and what it costs per product, before you commit the rest. [Book a demo](/book-a-demo) ## Entities - [Google](https://www.google.com) · Organization, mentions ## Sources - [Ecommerce search query types](https://baymard.com/blog/ecommerce-search-query-types) · Baymard Institute - [Product structured data](https://developers.google.com/search/docs/appearance/structured-data/product) · Google Search Central - [Creating helpful, reliable, people-first content](https://developers.google.com/search/docs/fundamentals/creating-helpful-content) · Google Search Central - [Product](https://schema.org/Product) · Schema.org --- # Shelf labels that say what the website says. > Physical Retail is the newest of the eight CommerceClarity areas and carries one agent, Shelf Label Content. It writes what an electronic shelf label or an in-store screen needs from the same product record the website reads: a short name that fits, the attributes that can be read at arm’s length, and the icons a regulated category has to show. It feeds the display system already installed rather than replacing it. Electronic shelf labels and in-store screens refresh as often as your site does. Each one still needs its own writing: a short name, the attributes that matter at arm’s length, and the icons the category requires. ![A shopper walking past a long grocery shelf run, the shelves streaking behind her.](https://cdn.sanity.io/images/xd4hrbt2/production/b66298118f51ca7fdec1692753c463df18a2d9dd-1774x887.jpg?auto=format) [Book a demo](/book-a-demo) ## A store is a channel with less room In most retailers the digital catalog stops at the warehouse door, and the store runs on its own content: a name typed to fit the old paper label, a price, and whatever the category manager had time to add. Then the shelf edge became a screen, and could suddenly show anything. The constraint changed shape rather than going away. A shelf label is read at arm’s length in a couple of seconds, so it is not a shortened product page: it is a different piece of writing, with a hard character budget, the one or two attributes that actually decide the purchase in that category, and the allergen or safety marks the category makes mandatory. And it is needed for every product, in every store, in the local language, updated whenever the assortment moves. That is the real work, and it is why dynamic shelf content so often launches on one pilot aisle and stops there. ## The jobs in the store Two workflows today. This is the newest of the eight areas, and the list will grow. What runs here feeds the display system and the planogram software you already installed. - Shelf Label Content · Generate the content for in-store digital shelf labels. · Continuous - Planogram · Fill the SKU dimensions your planogram software needs to lay out shelves and position products in store. · Continuous ## Written for arm’s length *Shelf label content* The agent derives shelf content from the same master record the site reads, and writes it to the format the label actually has: a short name that fits the field rather than one truncated by it, the attributes that matter in that category, the promotional flag, and the allergen or safety icons where the category requires them. It runs per store and per language, and it pushes to the display system rather than producing a file for somebody to import. When the assortment or the price moves, the label follows without a person retyping it. ## What retailers ask us ### What goes on an electronic shelf label? Price and name, then whatever the shelf has room for: unit price, origin, allergens, an energy class, a promotion. The hardware handles the display. The content comes from the same catalog the website reads, which is why the two drift apart when only one of them is maintained. ### Do you replace our label system? No. The hardware and the display platform stay yours. We produce the content they show and push it there. ### Is this just a truncated product name? No. Truncation is the failure this avoids. A label is written to its own budget, so the words that survive are the ones that decide the purchase rather than the first thirty characters of a web title. ### What about price and stock? They stay with the systems that own them. The agent owns the descriptive content around them, which is the part that was never written per product in the first place. ### Can it run per store? Yes, and that is usually the requirement. Assortment, language and local rules differ store to store, and content that is right in one region can be wrong in the next. ### Who checks it before it reaches a shelf? You do, on the same review the other agents use. A category can be set to publish once it clears the rules, or to hold everything for a person. ### Whose data is it? Yours. The corrections your team makes are what the agents learn from, and both the data and the corrections stay exportable at any point. ## Related areas - [Catalog Foundation](/solutions/catalog-foundation) - [Compliance & Quality](/solutions/compliance-quality) ## Every correction makes the next run better *The feedback loop* Your team judges the output. A value is right, or it is wrong and gets fixed once. Each verdict changes the next run: the score on your own sample moves, and the agent starts from what the last one learned. So the share of values that need a person keeps falling, measured on your catalog and your rules rather than on an average of everybody else. ![The platform flagging a product description as too generic for the brand tone, with the suggested rewrite and the two answers.](https://cdn.sanity.io/images/xd4hrbt2/production/30815d23ee4d132352d037d84d401a32a7b48eec-2880x1620.webp?auto=format) ## Live on your catalog in weeks, not months *How we work* We run all four phases with your team, on the stack you already have, and each one ends with a number you agreed in advance. - Design · We map how your data comes in and how the result goes back out, then agree the business KPI the agent has to move. - Implement · We build the agent and test it on a sample of your own catalog, agreed with you. Every change is scored against that sample, so the tuning runs on numbers. - Prove · Done means the KPI reached target on the full catalog, and you have seen the numbers yourself. - Evolve · We stay on the account. Your corrections keep improving the agent, the KPI holds where you need it, and the next use case starts from the context this one already built. ## Enterprise, on a catalog you cannot break ![A department store floor.](https://cdn.sanity.io/images/xd4hrbt2/production/251441053d31b02d7111cf86ee1083ac6fd8f061-1600x1000.jpg?auto=format) [Book a demo](/book-a-demo) - It runs on the stack you have · It reads your PIM, ERP and DAM through their APIs and writes the result back to them. - Nothing uncertain reaches a channel · Every value is checked against your rules, and anything uncertain goes to review before it reaches a channel. - Your data stays yours · Your own isolated workspace. It is never pooled into shared models, and you can export it at any point. - Every run is logged · Each value carries its source, its run and its score, so you can see why a wrong one went wrong. ## Start with one use case One use case, run end to end on your own catalog. That is enough to see what it moves and what it costs per product, before you commit the rest. [Book a demo](/book-a-demo) ## Entities - [GS1](https://www.gs1.org) · Organization, mentions - [EPREL](https://eprel.ec.europa.eu/screen/home) · Thing, mentions ## Sources - [European Product Registry for Energy Labelling (EPREL)](https://eprel.ec.europa.eu/screen/home) · European Commission - [Energy label and ecodesign product database](https://commission.europa.eu/energy-climate-change-environment/standards-tools-and-labels/products-labelling-rules-and-requirements/energy-label-and-ecodesign/product-database_en) · European Commission - [Regulation (EU) 2023/988 on general product safety](https://eur-lex.europa.eu/eli/reg/2023/988/oj) · EUR-Lex - [GTIN, the GS1 global trade item number](https://www.gs1.org/standards/id-keys/gtin) · GS1 --- # Found on the first try. > Search and Discovery is the CommerceClarity area that works the layer a search engine reads. Two agents run here: Taxonomy Mapping, which puts every product in a valid category, and On-site Search Enrichment, which fills the attributes and synonyms an index needs to reason on. It does not replace Algolia, FactFinder, Coveo or a platform’s own search: the relevance ceiling is set by the catalog underneath, and that is what the agents raise. Agents working the layer your search engine reads: every product in the right category, and the attributes and synonyms the index needs to reason on. Algolia, FactFinder, Coveo or the search that came with your platform, the relevance ceiling is set by the catalog underneath it. ![A shopper standing still with her phone while the supermarket aisle around her streaks past.](https://cdn.sanity.io/images/xd4hrbt2/production/5ed42dfefc8284243213d0374dddb3da838635a1-1774x887.jpg?auto=format) [Book a demo](/book-a-demo) ## The engine can only match what the catalog says The queries that fail are the ordinary ones. Asked for the worst searches on their own site, a grocery retailer sent back 4: still water, soft bread, yellow peppers, tuna. Plain words, and 3 of them a product name with no attribute attached to it. What goes wrong is a matching failure rather than a search failure. Ask for red trousers and a product named carrot comes back, because the word matched. Ask for an elegant shirt and you collide with a brand that is literally called Elegant. The engine did exactly what it was told, on the only thing it was given. Underneath both sits the taxonomy, and it is usually two taxonomies. One retailer handed us their tree and told us which half of it was real: 1,000 live leaf categories against 3,000 counting the disabled legacy ones, plus a set of leaves still alive under a parent that had been switched off. Search filters, SEO landing pages and marketplace acceptance all read that tree, so a category that does not resolve costs you all three at once. ## The jobs on discovery A few of the ones that run here, and the area is not limited to them. Each is a single workflow with its own output and its own numbers, reading the same records the rest of the catalog reads. - Taxonomy Mapping · Classify every product into the right category automatically. · Continuous - On-site Search Enrichment · Fix on-site search relevance with better attributes, on Algolia, FactFinder and the rest. · Continuous ## The category, and the words around it *Findability* What decides whether a product surfaces sits underneath the query, and that is where the agents work. The category comes first: every product classified at the right depth of your own tree, continuously, with the drift between the tree you designed and the tree you actually use reported back to you. A boundary call stays a category argument rather than a keyword one, so a fixed balcony solar kit is a different product from a portable panel however similar the words look. Products with no valid slot come back as the list of categories your taxonomy is missing. Then the attribute layer the index reads: missing attributes generated, existing ones normalized, and the synonym and variant lists the engine needs, so a search for red trousers returns trousers. Algolia, FactFinder, Coveo or an agentic engine, that layer is the ceiling on all of them. ## What retailers ask us ### What is ecommerce site search? The search box on your own store, and the product data it reads. Most search failures get blamed on the engine and are not the engine. It returns nothing because the product carries no colour, no material, or a category nobody would search for. We do not replace the search engine. We fix what it has to read. ### Do you replace our search engine? No. We work on the layer it indexes. Algolia, FactFinder, Coveo or the search that came with your platform: the agent improves the attributes, synonyms and categories the engine reasons on, and the engine stays where it is. ### How do you know the classification is right? It is measured, not asserted. Correctness is scored against an evaluation set your own experts sign off, and reported as a percentage with the errors listed product by product. ### What happens to products with no valid category? They are surfaced, not forced. The run hands back the categories your taxonomy is missing as a decision for you to take, kept separate from the products that were genuinely put in the wrong place. ### Our taxonomy keeps changing. Does that break it? No, it is the normal condition. Classification runs against the tree that is live, and when the tree moves the evaluation set and the agreed bar get re-pinned rather than assumed to carry over. ### Who decides the edge cases? You do. Where a product type belongs is a merchandising decision, so the boundary you draw is encoded as an explicit rule. ### Whose data is it? Yours. The corrections your team makes are what the agents learn from, and both the data and the corrections stay exportable at any point. ## Related areas - [SEO & GEO/AEO](/solutions/seo-geo) - [Catalog Foundation](/solutions/catalog-foundation) ## Proven by - [1000Farmacie](/customers/1000farmacie) · How 1000Farmacie reorganized its catalog and grew revenue thirty per cent a month ## Every correction makes the next run better *The feedback loop* Your team judges the output. A value is right, or it is wrong and gets fixed once. Each verdict changes the next run: the score on your own sample moves, and the agent starts from what the last one learned. So the share of values that need a person keeps falling, measured on your catalog and your rules rather than on an average of everybody else. ![The platform flagging a product description as too generic for the brand tone, with the suggested rewrite and the two answers.](https://cdn.sanity.io/images/xd4hrbt2/production/30815d23ee4d132352d037d84d401a32a7b48eec-2880x1620.webp?auto=format) ## Live on your catalog in weeks, not months *How we work* We run all four phases with your team, on the stack you already have, and each one ends with a number you agreed in advance. - Design · We map how your data comes in and how the result goes back out, then agree the business KPI the agent has to move. - Implement · We build the agent and test it on a sample of your own catalog, agreed with you. Every change is scored against that sample, so the tuning runs on numbers. - Prove · Done means the KPI reached target on the full catalog, and you have seen the numbers yourself. - Evolve · We stay on the account. Your corrections keep improving the agent, the KPI holds where you need it, and the next use case starts from the context this one already built. ## Enterprise, on a catalog you cannot break ![A department store floor.](https://cdn.sanity.io/images/xd4hrbt2/production/251441053d31b02d7111cf86ee1083ac6fd8f061-1600x1000.jpg?auto=format) [Book a demo](/book-a-demo) - It runs on the stack you have · It reads your PIM, ERP and DAM through their APIs and writes the result back to them. - Nothing uncertain reaches a channel · Every value is checked against your rules, and anything uncertain goes to review before it reaches a channel. - Your data stays yours · Your own isolated workspace. It is never pooled into shared models, and you can export it at any point. - Every run is logged · Each value carries its source, its run and its score, so you can see why a wrong one went wrong. ## Start with one use case One use case, run end to end on your own catalog. That is enough to see what it moves and what it costs per product, before you commit the rest. [Book a demo](/book-a-demo) ## Entities - [Algolia](https://www.algolia.com) · Organization, mentions - [Coveo](https://www.coveo.com) · Organization, mentions - [FactFinder](https://www.fact-finder.com) · Organization, mentions ## Sources - [Ecommerce search query types](https://baymard.com/blog/ecommerce-search-query-types) · Baymard Institute - [Product structured data](https://developers.google.com/search/docs/appearance/structured-data/product) · Google Search Central - [Product data specification](https://support.google.com/merchants/answer/7052112) · Google Merchant Center - [Product](https://schema.org/Product) · Schema.org --- # AI visibility for the products, not just the brand. > SEO and GEO/AEO is the CommerceClarity area that works both demand surfaces off one product record. Four agents run here: Keyword Optimizer, GEO Content Writer, GEO Share of Voice and PDP Q&A. Generative and answer engine optimization are measured as share of voice in AI answers rather than as rank alone, and the cost per product is low enough to cover the long tail, which is where most unclaimed search demand sits. Agents working the same product knowledge for both demand surfaces: the words shoppers actually type, the content the AI engines lift into an answer, and a question block on every page. At a cost per product low enough to reach the tail, which is where the unclaimed demand is. ![A woman at a cafe window at dusk working on a laptop, the lit shop fronts and the people on the street outside streaking past her.](https://cdn.sanity.io/images/xd4hrbt2/production/85887a0f4592f854b74090404a47b9bda06d843a-1774x887.jpg?auto=format) [Book a demo](/book-a-demo) ## Your catalog does not speak the way shoppers ask A product page is usually written the way the business files the product: the supplier’s name for it, the specification in the manufacturer’s words, a description written once for a brochure. Shoppers do not phrase it that way, and neither do the engines that increasingly answer on their behalf. There are two demand surfaces now and they reward different things. Google still reads a title and a description. An AI engine lifts whatever is structured enough to quote: a direct answer, a comparison, the trade-off, the question somebody actually asked. A catalog optimized for the first is not optimized for the second. What makes this economic is the tail. Rewriting the top hundred products is a morning of work for one person, and it is also the part that already ranks. The rest of the catalog is where the demand is unclaimed, and it only ever gets written if the cost per product is close to zero. ## The jobs on demand A few of the ones that run here, and the list grows as the surfaces do. Each is a single workflow with its own output and its own numbers, working off the same product knowledge underneath. Run one, or run the set. - Keyword Optimizer · Research keywords and optimize titles and descriptions to rank. · Continuous - GEO Content Writer · Write the content that gets your products cited by AI engines. · Continuous - GEO Share of Voice · Track how often the AI engines cite you. · Continuous - PDP Q&A · Add a question and answer block that answers what shoppers actually ask. · Continuous ## Written for the list, and for the answer *Capturing demand* The same product knowledge, worked for two surfaces. For search: how people actually phrase a query in your categories, on Google and inside marketplace search, with titles and descriptions rewritten against those phrasings rather than against the supplier’s name for the product. For the answer engines: the comparisons, trade-offs and direct answers in the shapes an engine will lift. A question block on every page earns on both surfaces, built from the questions your pages never answered: Google reads it for People Also Ask, and an AI engine quotes it when it recommends the product. Underneath both sits the measurement, rankings on one side and citation share on the other. All of it runs per product, which is the only reason it reaches past the hundred products somebody already optimized by hand. ## What retailers ask us ### What is AI visibility? Whether a model answering a question about products like yours names yours. It is not the same as ranking on Google, and it is not brand monitoring. For a retailer the unit is the product, not the company. A model recommends a specific item, and it can only do that if the item carries enough data to be recommended. ### Is GEO/AEO just SEO with a new name? No. Ranking a page and being quoted in an answer are different mechanics. SEO optimizes a title and a description for a list of results. GEO structures the content an engine will lift, and the format that gets lifted is not the format that ranks. ### Do we have to rewrite the whole catalog? No. The work goes where the demand is unclaimed, which is usually the tail rather than the products somebody already optimized by hand. ### How is it measured? On the demand surface rather than on the output. Rankings and traffic for the search side, citation share for the answer side. Tracking that citation share is its own workflow, separate from the writing. ### Will the content read like a machine wrote it? It is written against your own rules: tone of voice, glossary, the claims you are allowed to make. Anything the rules do not cover is held for review instead of published. ### Does a generated question block clash with our real customer questions? No, it fills the gap ahead of them. Where you already have customer questions and answers, those stay. The agent covers the pages where nobody has asked anything yet. ### Whose data is it? Yours. The corrections your team makes are what the agents learn from, and both the data and the corrections stay exportable at any point. ## Related areas - [Agentic Commerce](/solutions/agentic-commerce) - [Search & Discovery](/solutions/search-discovery) ## Proven by - [Next Hardware & Software](/customers/next-hardware-software) · How enriched product pages lifted search and conversion for Next ## Every correction makes the next run better *The feedback loop* Your team judges the output. A value is right, or it is wrong and gets fixed once. Each verdict changes the next run: the score on your own sample moves, and the agent starts from what the last one learned. So the share of values that need a person keeps falling, measured on your catalog and your rules rather than on an average of everybody else. ![The platform flagging a product description as too generic for the brand tone, with the suggested rewrite and the two answers.](https://cdn.sanity.io/images/xd4hrbt2/production/30815d23ee4d132352d037d84d401a32a7b48eec-2880x1620.webp?auto=format) ## Live on your catalog in weeks, not months *How we work* We run all four phases with your team, on the stack you already have, and each one ends with a number you agreed in advance. - Design · We map how your data comes in and how the result goes back out, then agree the business KPI the agent has to move. - Implement · We build the agent and test it on a sample of your own catalog, agreed with you. Every change is scored against that sample, so the tuning runs on numbers. - Prove · Done means the KPI reached target on the full catalog, and you have seen the numbers yourself. - Evolve · We stay on the account. Your corrections keep improving the agent, the KPI holds where you need it, and the next use case starts from the context this one already built. ## Enterprise, on a catalog you cannot break ![A department store floor.](https://cdn.sanity.io/images/xd4hrbt2/production/251441053d31b02d7111cf86ee1083ac6fd8f061-1600x1000.jpg?auto=format) [Book a demo](/book-a-demo) - It runs on the stack you have · It reads your PIM, ERP and DAM through their APIs and writes the result back to them. - Nothing uncertain reaches a channel · Every value is checked against your rules, and anything uncertain goes to review before it reaches a channel. - Your data stays yours · Your own isolated workspace. It is never pooled into shared models, and you can export it at any point. - Every run is logged · Each value carries its source, its run and its score, so you can see why a wrong one went wrong. ## Start with one use case One use case, run end to end on your own catalog. That is enough to see what it moves and what it costs per product, before you commit the rest. [Book a demo](/book-a-demo) ## Entities - [Google](https://www.google.com) · Organization, mentions - [OpenAI](https://openai.com) · Organization, mentions ## Sources - [Creating helpful, reliable, people-first content](https://developers.google.com/search/docs/fundamentals/creating-helpful-content) · Google Search Central - [Product structured data](https://developers.google.com/search/docs/appearance/structured-data/product) · Google Search Central - [FAQ structured data](https://developers.google.com/search/docs/appearance/structured-data/faqpage) · Google Search Central - [Commerce documentation](https://developers.openai.com/commerce/) · OpenAI --- # How 1000Farmacie reorganized its catalog and grew revenue thirty per cent a month > 1000Farmacie, an Italian online pharmacy marketplace bringing more than seventy independent pharmacies onto one storefront with a catalog of over 100,000 products, rebuilt 38,000 product pages and lifted revenue thirty per cent month over month within six months. The work ran in order: taxonomy first, so every product sits in the right category, then the missing attributes and details, then titles and descriptions rewritten for the search a customer actually types. ![A 1000Farmacie billboard.](https://cdn.sanity.io/images/xd4hrbt2/production/c8a36cadadd376cf1b836ddf14b806f2ff66b9bc-800x600.jpg?auto=format) **Hero figure value:** +30% **Hero figure label:** Revenue, month over month **Industry:** Online pharmacy ## Key figures - 38,000 · product pages rebuilt and optimized - +30% · revenue, month over month - 6 mo · from start to those numbers - 70+ · independent pharmacies on one catalog ## Agents - Taxonomy Mapping - Data Enrichment - Keyword Optimizer ## Overview 1000Farmacie is one of the fastest growing online pharmacy platforms in Italy, a marketplace bringing more than seventy independent pharmacies onto one storefront with a catalog of over 100,000 products, from over-the-counter medicines and supplements to dermocosmetics and mother and baby. A marketplace of that shape inherits its data from everyone on it, which is exactly where the growth was stuck. ## The challenge Categorization was inconsistent and descriptions were thin, and the two compounded each other. - Customers could not find products that were misclassified or half described. - Search surfaced the wrong items, which diluted relevance and cost organic traffic. - Pages that did get the visit did not convert it. ## How we solved it 1. **Taxonomy first.** The whole catalog structure was reviewed and rebuilt so every product sits in the right category and the hierarchy holds. 1. **Then the gaps.** Missing attributes and product details were filled, which is what makes a category page usable once the taxonomy is right. 1. **Then the copy.** Titles and descriptions rewritten for the search a customer actually types. ## What changed 38,000 product pages rebuilt, whole categories unlocked that the old structure had buried, and revenue up thirty per cent month over month within six months. The new taxonomy also made the catalog cheaper to run, which is the part that keeps paying. The next objective is the one Purina is already on: making the catalog discoverable, selectable and recommendable by AI agents. ## Before and after Six months, one catalog, seventy pharmacies behind it. - Taxonomy · Inconsistent, products in the wrong place · Rebuilt, every product in its category - Product pages · Thin descriptions, missing attributes · 38,000 rebuilt and optimized - Revenue · Constrained by what customers could find · Up thirty per cent month over month - Catalog management · Manual and hard to scale · Efficient on the new structure ## Proves area [Search & Discovery](/solutions/search-discovery) ## More customer stories - [Next Hardware & Software](/customers/next-hardware-software) · How enriched product pages lifted search and conversion for Next - [Nestlé Purina](/customers/purina) · How Purina got its catalog ready to be recommended by AI agents ## See what the agents do with your catalog Bring one messy feed, or the backlog of products you never got to. We will show you the rest. [Book a demo](/book-a-demo) ## Questions ### What did 1000Farmacie achieve? 38,000 product pages rebuilt and optimized, whole categories unlocked that the old structure had buried, and thirty per cent revenue growth month over month, reached within six months. ### Why was the catalog holding growth back? A marketplace inherits its data from everyone selling on it. Categorization was inconsistent and descriptions were thin, so customers could not find products that were misclassified or half described, search surfaced the wrong items, and the pages that did get a visit did not convert it. ### Why start with the taxonomy rather than the copy? Because the order is load-bearing. Filling attributes on a product filed in the wrong category improves nothing a customer can reach, and rewriting copy for a page nobody lands on is wasted. Taxonomy, then the gaps, then the copy. ### How big is the catalog? Over 100,000 products, from over-the-counter medicines and supplements to dermocosmetics and mother and baby, contributed by more than seventy independent pharmacies and parapharmacies on one storefront. ### What is next for 1000Farmacie? The same objective Purina is already on: making the catalog discoverable, selectable and recommendable by AI agents, from shopping assistants to conversational platforms like ChatGPT. ## Entities - [1000Farmacie](https://www.1000farmacie.it) · Organization, about - [Search & Discovery](/solutions/search-discovery) · Thing, about - [ChatGPT](https://en.wikipedia.org/wiki/ChatGPT) · Product, mentions ## Sources - [1000Farmacie](https://www.1000farmacie.it) · 1000Farmacie - [Product structured data markup](https://developers.google.com/search/docs/appearance/structured-data/product) · Google Search Central - [Product data specification](https://support.google.com/merchants/answer/7052112) · Google Merchant Center - [Product](https://schema.org/Product) · schema.org --- # How Caddy’s made compliance a property of its product data > Caddy’s, part of the DMO group and one of Italy’s better known beauty, personal care and household retailers with hundreds of stores, automated product attribute enrichment with compliance validation built into the same pass. Thousands of SKUs were brought up to standard, and the requirements of Italian consumer protection law are now checked as the record is created rather than reviewed page by page afterwards. No existing process was dismantled to do it. ![Caddy’s own-brand products on a shelf.](https://cdn.sanity.io/images/xd4hrbt2/production/c0cb0329aa7c89b84206ad399900ff4e00a04c72-800x600.jpg?auto=format) **Industry:** Beauty and household retail ## Key figures - Hundreds · of stores, and a growing online catalog - Thousands · of SKUs enriched and checked against the rules ## Agents - Data Enrichment - Catalog Auditing ## Overview Caddy’s is part of the DMO group and one of Italy’s better known retail brands in beauty, personal care and household goods, with hundreds of stores and an online catalog growing fast behind them. It came to us with a problem that is easy to state and hard to hold: every product page, online and in store, has to carry information that is accurate, complete and compliant with Italian consumer protection rules. ## The challenge Attributes are what a product is made of, in data terms. A retailer tracks stock by code; a customer searches in their own words, for a shampoo that suits sensitive skin or a cream without a particular ingredient. Size, material, use and the regulated details are what connect the two. At the scale of thousands of pages, keeping that set complete by hand is not a quality problem, it is an exposure. Compliance details are the ones a person is most likely to skip and a regulator most likely to check. ## How we solved it The programme centred on automatic attribute enrichment with the rules built into it: the agents populate the attribute set and validate it against the compliance requirements in the same pass, so a record either meets the standard or gets flagged before it is published. Nothing in the existing process had to be dismantled for it. The check sits where the data is created, which is the only place it holds as the catalog grows. ## What changed Attribute quality improved across thousands of SKUs, and compliance stopped being a periodic review and became a property of the data. The ecommerce team spends its time on the customer experience rather than on repetitive data entry. ## Before and after Where the rules live, before and after. - Compliance · Checked by hand, page by page · Validated as the record is built - Attributes · Incomplete and uneven across the catalog · Complete and consistent across thousands of SKUs - The ecommerce team · Repetitive data entry · Customer experience - Existing processes · n/a · Unchanged, the check sits inside them ## Proves area [Compliance & Quality](/solutions/compliance-quality) ## More customer stories - [1000Farmacie](/customers/1000farmacie) · How 1000Farmacie reorganized its catalog and grew revenue thirty per cent a month - [Nestlé Purina](/customers/purina) · How Purina got its catalog ready to be recommended by AI agents ## See what the agents do with your catalog Bring one messy feed, or the backlog of products you never got to. We will show you the rest. [Book a demo](/book-a-demo) ## Questions ### What did Caddy’s automate? Product attribute enrichment, with the compliance rules built into the same pass. The agents populate the attribute set and validate it against the requirements, so a record either meets the standard or is flagged before it publishes. ### Why do product attributes decide compliance? Because they are what a product is, in data terms. A retailer tracks stock by code, a customer searches in their own words, and size, material, use and the regulated details are what connect the two. At the scale of thousands of pages, keeping that set complete by hand is not a quality problem, it is an exposure. ### Which rules is the catalog held to? Italian consumer protection law, the Codice del Consumo, which implements the European consumer rules on the information a seller has to give before a purchase. The validation runs against those requirements as the record is built. ### Did the existing process have to change? No. The check sits where the data is created, inside what Caddy’s already ran. That is also the only place it keeps holding as the catalog grows: a periodic review falls behind, a check at the point of creation does not. ### Are there hard numbers on this one? Not published ones. What Caddy’s has agreed we may say is the shape of the outcome: attribute quality improved across thousands of SKUs, compliance is verified rather than reviewed, and the ecommerce team spends its time on the customer experience instead of repetitive data entry. A percentage we cannot source is a percentage we do not print. ## Entities - [Caddy’s](https://www.caddys.it) · Organization, about - [Compliance & Quality](/solutions/compliance-quality) · Thing, about - [DMO S.p.A.](https://www.dmospa.it) · Organization, mentions ## Sources - [Caddy’s](https://www.caddys.it) · Caddy’s - [DMO S.p.A.](https://www.dmospa.it) · DMO - [Codice del Consumo, decreto legislativo 6 settembre 2005 n. 206](https://www.normattiva.it/uri-res/N2Ls?urn:nir:stato:decreto.legislativo:2005-09-06;206) · Normattiva - [Consumer protection law](https://commission.europa.eu/law/law-topic/consumer-protection-law_en) · European Commission --- # How Cisalfa Group got its catalog ready for the new season without growing the team > Cisalfa Group, the largest sports retailer in Italy and Germany with 240+ stores and roughly 20,000 SKUs, cleared the FW26 season step change without growing its catalog team. Close to 20% of the catalog is private label and the qualifying attributes do not live in the ERP, so the content has to be generated rather than copied, and produced twice, once for the Cisalfa Sport site and once for the Intersport-network catalog. An agent pipeline mirrored 1:1 on the customer’s Akeneo normalizes, enriches, applies the rules and generates both channel variants, with a CommerceClarity Retail Engineer embedded in the team. In the first quarter it saved 84.7% of the time per style against a 50% target, generated 2,563 styles, and lifted the mandatory-attribute match from 86.0% to 93.2%. ![The footwear and football department of a Cisalfa Group store.](https://cdn.sanity.io/images/xd4hrbt2/production/fe30d3dc718e2cf99a045881aa4da8aba4d89e80-2000x1334.jpg?auto=format) **Hero figure value:** 84.7% **Hero figure label:** Time saved per style, against a 50% target **Industry:** Sports retail ## Key figures - 84.7% · time saved per style, against a 50% target - 2,563 · styles generated in the first quarter - 93.2% · mandatory-attribute match, up from 86.0% ## Agents - Data Enrichment ## The challenge Cisalfa Group is the largest sports retailer operating in Italy and Germany: 240+ stores across 18 regions, around 5,200 employees, and a catalog of roughly 20,000 SKUs spanning 280+ third-party brands plus its own private-label portfolio, ABC, Admiral, Arena, Dack’s, Fila and Performer. In Italy it sells through two distinct storefronts, the Cisalfa Sport brand site and the Intersport-network catalog, each with its own schema, naming and language conventions. The platform stack runs on an Akeneo PIM feeding Salesforce Commerce Cloud. The catalog work was hard for reasons that compounded. 1. **Close to 20% of the catalog is private label**, so the content cannot be pulled from a brand feed. It has to be constructed in-house from sparse internal data. 1. The attributes that actually make a sports product sellable, the **qualifying attributes** like fit, composition and material, mostly **do not live in the ERP**, so they have to be generated rather than copied. 1. On the third-party side, only about **half of the brand catalog cleanly matched** Cisalfa Group’s own schema, which left a large manual reconciliation gap. 1. Every product has to be produced **twice**, once for Cisalfa Sport and once for Intersport, against two different sets of field names and rules. On top of the catalog itself, time was the binding constraint. Time-to-market at Cisalfa Group is measured as the share of SKUs complete against warehouse arrivals, with a bar above 95%. The FW26 season meant a step change in volume, and the catalog had historically been run by a small, junior-heavy copy team of one to two people for 20,000 SKUs. Hitting FW26 the old way would have meant multiplying the manual effort rather than removing it. Cisalfa Group wanted the opposite: to clear the FW26 step change at a higher quality bar, without growing the team. More catalog, not more people. ### The setup before CommerceClarity Cisalfa Group ran its catalog on the systems built for its retail and store business. Product data lived in Akeneo, designed to hold specifications and feed the storefronts, not to generate sellable, channel-ready content. Around that system ran a fully manual operation. - Catalog copy and attributes were produced **100% in-house**, by a small team of one to two people for around 20,000 SKUs. - Quality was tracked as **binary completeness**: an attribute was either present or absent, with no notion of richness or accuracy. - There was no internal capacity to build the dashboards that would have exposed the gap. - Every additional brand, channel or attribute group meant another full round of manual enrichment and checking, run once for Cisalfa Sport and once for Intersport. **The model worked, but it did not scale.** ## How CommerceClarity solved it: AI agents and Retail Engineers CommerceClarity slotted into the existing stack with no rip-and-replace. The configuration was set up as a 1:1 mirror of Cisalfa Group’s Akeneo, so generated content exports straight back in a compatible format, and supplier-portal files are ingested directly instead of being reworked by hand. ### The agents bring the scale Each product starts as a single canonical record, and an agent pipeline carries it from raw input to channel-ready content, every step building on the one before it. 1. **Normalize.** Map heterogeneous inputs, ERP fields, supplier-portal files and brand feeds, into Cisalfa Group’s taxonomy and hierarchy. 1. **Enrich.** Generate the qualifying attributes the ERP does not carry, fit, composition, material and the like, which is where Cisalfa Group sees the real added value. 1. **Apply rules.** Enforce brand voice, channel constraints and the agreed data model, so the taxonomy-driven attributes come back correct by construction. 1. **Generate channel variants.** Reshape the single source into both the Cisalfa Sport and the Intersport formats, each with its own schema and naming. 1. **Validate.** Score quality, run the automated checks, and surface only the exceptions that need a human eye. Cisalfa Group did not switch all of this on at once. They started where the pain was sharpest, end-to-end generation for both Cisalfa Sport and Intersport, then layered in direct supplier-file ingestion and the Akeneo mirror, with per-attribute enrichment and exception-only review following on. ### The Retail Engineers co-build the outcome The agents do not run alone. A CommerceClarity Retail Engineer works as an extension of Cisalfa Group’s team, the same person scoping, building and shipping inside the way the company already operates rather than handing work across siloes. That flipped the operating model. The old way was manual and regression-prone, a small team checking every product and every change by hand. With the agents and the Retail Engineer in place it became governed and exception-only: the team reviews only what the pipeline flags, every run is traceable, and unreliable sources are excluded by construction. The agents bring the speed and the scale, and the people keep the governance over what actually ships. ## What is next With the foundation in place, the work has shifted from building it to building on it: scaling FW26 volumes through direct supplier-file upload with no rework, driving rework toward zero with exception-only review and per-attribute correction, and going beyond text with an image workflow built to Cisalfa Group’s guidelines and coordinated outfits drawn from the catalog. From here the partnership continues the way it started, side by side. A CommerceClarity Retail Engineer and Cisalfa Group’s team find the next bottleneck together, build the agents that clear it, and turn more of the catalog into revenue. ## Before and after The first quarter of work in a single view. - Time per style · Fully manual · 84.7% saved, about 18 minutes a style, against a 50% target - Throughput · 1 to 2 people for around 20,000 SKUs · About 30 products a day at 3 to 5 minutes each, 2,563 styles in Q1 - Mandatory-attribute match · Binary completeness only · 86.0% to 93.2%, scored per attribute - Channels · Produced by hand, per storefront · Cisalfa Sport and Intersport generated from one source - Operating model · Internal team, reactive, regression-prone · Agents and an embedded Retail Engineer, exception-only review, fully traceable ## More catalog, not more people *In their words* ### Francesco Gasca As our catalog continued to grow in both volume and complexity, we needed a new way to scale without simply adding resources. Together with CommerceClarity, we built a model that combines AI-powered automation with strong governance, enabling us to manage more products, improve data quality and accelerate time-to-market without expanding the team. The real value goes beyond efficiency: it allows our people to focus on higher-value activities while giving us a stronger, more scalable foundation for future growth. Chief Digital & Omnichannel Officer, Cisalfa Group ![Francesco Gasca, Chief Digital and Omnichannel Officer at Cisalfa Group.](https://cdn.sanity.io/images/xd4hrbt2/production/4cb0ccd37e8cef820c99808250c16846db1ff8db-1600x1600.jpg?auto=format) ## Proves area [Catalog Foundation](/solutions/catalog-foundation) ## More customer stories - [STIGA](/customers/stiga) · How STIGA launched on seven marketplaces in four months - [Yeppon](/customers/yeppon) · How Yeppon got a 150,000 SKU catalog marketplace-ready, from its own store outward ## See what the agents do with your catalog Bring one messy feed, or the backlog of products you never got to. We will show you the rest. [Book a demo](/book-a-demo) ## Questions ### What changed for Cisalfa Group? In the first quarter, 84.7% of the time per style against a 50% target, which is about 18 minutes saved on every style. 2,563 styles were generated, and the mandatory-attribute match went from 86.0% to 93.2%. The team did not grow. ### Why was this catalog harder than most? Close to 20% of it is private label, so the content cannot be pulled from a brand feed and has to be built in-house from sparse internal data. The attributes that make a sports product sellable, fit, composition and material, mostly do not live in the ERP. And every product has to be produced twice, once for the Cisalfa Sport site and once for the Intersport-network catalog, against two different schemas. ### How are two storefronts handled without doing the work twice? Each product is one canonical record. The pipeline generates the channel variants from it, reshaping the same source into the Cisalfa Sport and the Intersport formats with their own field names and rules, so the work is done once and shaped twice. ### Did Cisalfa Group have to change its systems? No. CommerceClarity slotted into the Akeneo and Salesforce Commerce Cloud stack that was already running, with no rip-and-replace. The configuration is a 1:1 mirror of Cisalfa Group’s Akeneo, so generated content exports straight back in a compatible format, and supplier-portal files are ingested directly instead of being reworked by hand. ### How big is the catalog behind this? Cisalfa Group runs 240+ stores across 18 regions with around 5,200 employees, and a catalog of roughly 20,000 SKUs spanning 280+ third-party brands plus its own private-label portfolio. ## Entities - [Cisalfa Group](https://www.cisalfagroup.com) · Organization, about - [Cisalfa Sport](https://www.cisalfasport.it) · Organization, mentions - [Catalog Foundation](/solutions/catalog-foundation) · Thing, about - Francesco Gasca · Person, mentions ## Sources - [Cisalfa Group](https://www.cisalfagroup.com) · Cisalfa Group - [Product data specification](https://support.google.com/merchants/answer/7052112) · Google Merchant Center - [Product](https://schema.org/Product) · schema.org - [Product structured data markup](https://developers.google.com/search/docs/appearance/structured-data/product) · Google Search Central --- # How enriched product pages lifted search and conversion for Next > Next Hardware & Software, an Italian distributor of IT and gaming products founded in 1996, publishes 500+ enriched product pages a month with CommerceClarity and records thirty per cent more organic search visibility. It carries 200+ brands and 10,000 items in stock, and its store reaches over 100,000 unique visitors a month. In 2026 it was named among the E-commerce Stars, the Corriere della Sera and Statista ranking of Italy’s best online shops. ![The Next Hardware & Software building.](https://cdn.sanity.io/images/xd4hrbt2/production/10cf0196902973def374708bd8a485e63bd7d6fc-800x400.jpg?auto=format) **Hero figure value:** 500+ **Hero figure label:** New product pages a month **Industry:** IT and gaming distribution ## Key figures - 500+ · new product pages a month - 30% · more organic search visibility - 10,000 · items in stock, 200+ brands - 100k+ · unique visitors a month ## Agents - Data Enrichment - Keyword Optimizer ## Overview Next Hardware & Software was founded in Italy in 1996 and distributes IT and gaming products: 200+ brands, 10,000 items in stock, and an e-commerce reaching over 100,000 unique visitors a month. In 2026 it was named among the E-commerce Stars by Corriere della Sera and Statista. It was one of our first customers, and it came to us in a period of expansion with a specific problem: the catalog could not grow at the speed the business was growing. ## The challenge Catalog management was manual end to end. Every listing was enriched, corrected and uploaded by hand, which capped how much could ship in a month. Their own performance data made the second problem plain: pages with thin or badly structured content had markedly higher bounce rates. Incomplete product data was costing them both search visibility and the visit itself. ## How we solved it Every product page gets completed automatically: structured attributes, copy that reads, and descriptions written for search. The keyword work is part of the same pass rather than a later clean-up, so intent lands in titles, metadata and descriptions as the page is built. The marketing team keeps the pen. Generated content is theirs to review, edit and adapt by category or campaign, which is what let them scale production without handing over the brand’s voice. ## What changed 500+ new product pages a month, and thirty per cent more organic search visibility. Pages that carry real detail hold the visit: more time on site, more pages per session, fewer people leaving from the first one. ## Before and after What the catalog operation looked like on either side of the change. - New pages a month · Whatever the team could type · 500+, generated and reviewed - Organic visibility · Held back by thin pages · Up thirty per cent - Page quality · Minimal content, high bounce · Structured attributes and copy that answers - The marketing team · Enriching and uploading by hand · Editing, adapting and approving ## Proves area [SEO & GEO/AEO](/solutions/seo-geo) ## More customer stories - [1000Farmacie](/customers/1000farmacie) · How 1000Farmacie reorganized its catalog and grew revenue thirty per cent a month - [Yeppon](/customers/yeppon) · How Yeppon got a 150,000 SKU catalog marketplace-ready, from its own store outward ## See what the agents do with your catalog Bring one messy feed, or the backlog of products you never got to. We will show you the rest. [Book a demo](/book-a-demo) ## Questions ### What did Next Hardware & Software achieve? 500+ new product pages a month and thirty per cent more organic search visibility, with visitors spending longer on site, opening more pages and leaving from the first one less often. ### How did they know thin product pages were costing them? Their own performance data said so. Pages with minimal or badly structured content showed markedly higher bounce rates, which made the case for a data-rich, automated approach rather than more manual editing. ### Is the SEO work a separate pass? No, it happens as the page is built. The keyword research and the search intent land in titles, metadata and descriptions in the same pass that fills the structured attributes and writes the copy, rather than as a clean-up afterwards. ### Does the marketing team still control the content? Yes. Generated content is theirs to review, edit and adapt by category or by campaign. That is what let them scale production without handing over the brand’s voice. ### What are the E-commerce Stars? Le Stelle dell’Ecommerce, the ranking Corriere della Sera publishes with Statista, which tests hundreds of Italian online shops on around forty technical criteria plus consumer scoring. Next was named among the 2026 edition. ## Entities - [Next Hardware & Software](https://www.nexths.it) · Organization, about - [SEO & GEO](/solutions/seo-geo) · Thing, about - [Corriere della Sera](https://www.corriere.it) · Organization, mentions - [Statista](https://www.statista.com) · Organization, mentions ## Sources - [Next Hardware & Software](https://www.nexths.it) · Next Hardware & Software - [Le Stelle dell’E-commerce 2026, ranking methodology](https://cdn.statista-rankings.com/e-commerce-and-retail/rankings/onlineshops-it-2026/methodology-le-stelle-dell-ecommerce-top-e-commerce-companies-italy-2026.pdf) · Statista and Corriere della Sera - [Product structured data markup](https://developers.google.com/search/docs/appearance/structured-data/product) · Google Search Central - [Product](https://schema.org/Product) · schema.org --- # How Purina got its catalog ready to be recommended by AI agents > Purina, part of the Nestlé group and selling in more than ninety countries, prepared its product data to be read and recommended by AI assistants rather than only ranked by search engines. The work was structured data enrichment, metadata rewritten for machine comprehension, and one category at a time so discoverability could be measured. Cat food went first, and Purina products are now discoverable and recommended inside LLM-based platforms. No existing system was replaced. ![A dog beside a Purina Pro Plan pack, from Purina’s own material.](https://cdn.sanity.io/images/xd4hrbt2/production/2ce981bfe7f9e8f6c7c8a28eae8b46d6faecd0cf-800x600.jpg?auto=format) **Industry:** Pet nutrition ## Key figures - 90+ · countries the brand sells in - 2 · categories proven first, cat food and dog nutrition ## Agents - GEO Content Writer - GEO Share of Voice ## Overview Purina is part of the Nestlé group and one of the most established names in pet nutrition, selling in more than ninety countries. Its product pages were already optimized for search. The question it came to us with was about the next intermediary. People are starting to discover, compare and buy through AI assistants. That multiplies the number of places a product has to be legible, and an assistant does not read a page the way a search engine does. ## The challenge A page can rank and still be invisible to an agent. What an assistant needs is semantic structure, attributes that are actually there, and context it can reason over. Traditional product content, however well optimized for keywords, often carries none of that. The second constraint was practical: close that gap without disturbing what already works for search, and without a replatform. ## How we solved it 1. **Structured data enrichment.** Product pages carry semantic attributes, so a machine can tell what the product is, what it is for, what is in it and who it suits. 1. **Metadata written for comprehension.** Titles, descriptions and tags rewritten so a conversational platform can place the product in context rather than match a string. 1. **One category at a time.** Cat food and dog nutrition first, so discoverability inside AI environments could be measured before the pattern was rolled wider. ## Three layers, not a replacement Search is not going away, it is getting another two layers on top. Classic visibility is keywords, links and authority. Generative visibility is schema, structure and semantic comprehension. Answer visibility is whether an assistant trusts the product enough to include it in what it tells someone. Each layer stands on the one under it. ## What changed Purina’s products, cat food first, are now discoverable and recommended inside LLM-based platforms, which makes it one of the first brands of its size to have done the work before the channel matters. ## Before and after Same catalog, read by a different kind of reader. - Optimized for · Search engines · Search engines and AI assistants - Product data · Readable by people, thin for machines · Semantic attributes a model can reason over - Discoverability in AI answers · Unmeasured · Measured by category, starting with cat food - Systems touched · n/a · None replaced ## Proves area [Agentic Commerce](/solutions/agentic-commerce) ## More customer stories - [1000Farmacie](/customers/1000farmacie) · How 1000Farmacie reorganized its catalog and grew revenue thirty per cent a month - [Caddy’s](/customers/caddys) · How Caddy’s made compliance a property of its product data ## See what the agents do with your catalog Bring one messy feed, or the backlog of products you never got to. We will show you the rest. [Book a demo](/book-a-demo) ## Questions ### Purina was already optimized for search. What was missing? The structure a machine needs to reason. An assistant does not match strings, it evaluates: what the product is, what it is for, what is in it, who it suits. A page can rank on keywords and still carry none of that, which makes it invisible to the systems that now sit between the shopper and the shelf. ### What did the work involve? Three things. Product pages enriched with semantic attributes. Titles, descriptions and tags rewritten so a conversational platform can place the product in context. And one category at a time, cat food and dog nutrition first, so discoverability inside AI environments could be measured before the pattern went wider. ### Does this replace SEO? No, it adds two layers on top of it. Classic visibility is keywords, links and authority. Generative visibility is schema, structure and semantic comprehension. Answer visibility is whether an assistant trusts the product enough to include it in what it tells someone. Each layer stands on the one under it. ### What is the measurable result? Purina products, cat food first, are discoverable and recommended inside LLM-based platforms. That is the result the category was chosen to test, and it is measured by category rather than claimed for the whole catalog. ### Did anything have to be replatformed? No. The constraint was explicit from the start: close the gap without disturbing what already works for search, and without replacing a system. ### Is this Purina or Nestlé? Purina is Nestlé’s pet care brand, and the work is on the Purina catalog. The mark on this page is the Nestlé one because Purina’s own logo is a solid panel with the name reversed out of it, which a white treatment turns into a block. ## Entities - [Nestlé Purina](https://www.purina.it) · Organization, about - [Agentic Commerce](/solutions/agentic-commerce) · Thing, about - [Nestlé](https://www.nestle.com) · Organization, mentions - [ChatGPT](https://en.wikipedia.org/wiki/ChatGPT) · Product, mentions - [OpenAI](https://openai.com) · Organization, mentions ## Sources - [Purina Italia](https://www.purina.it) · Nestlé Purina PetCare - [AI features and your website](https://developers.google.com/search/docs/appearance/ai-features) · Google Search Central - [Product](https://schema.org/Product) · schema.org - [Merchant listing structured data](https://developers.google.com/search/docs/appearance/structured-data/merchant-listing) · Google Search Central - [ChatGPT](https://en.wikipedia.org/wiki/ChatGPT) · Wikipedia --- # How STIGA launched on seven marketplaces in four months > STIGA, Europe’s market leader in ride-on mowers, went live on seven marketplaces in five native languages in four months, with 500+ product pages generated, validated and published. The channels include Amazon in five countries, Allegro, Otto, Kaufland, ManoMano, Leroy Merlin and cDiscount. The sequential route, one country at a time, takes one to two years. Product data stayed in Akeneo, STIGA’s own PIM, with nothing replaced. ![A STIGA machine in a garden.](https://cdn.sanity.io/images/xd4hrbt2/production/f17c5962644bda093d12da9e166e2dfdcba8f217-1000x750.jpg?auto=format) **Hero figure value:** 500+ **Hero figure label:** Product pages generated, validated and live **Industry:** Outdoor power equipment ## Key figures - 7 · marketplaces live - 500+ · PDPs generated, validated and live - 5 · native languages - 4 mo · to live, against one to two years ## Agents - Data Enrichment - Content Translation - Amazon Listing Optimizer - Leroy Merlin Listing Optimizer ## Overview For over ninety years STIGA has engineered powered garden equipment, and today it is Europe’s market leader in ride-on mowers, sold through specialist dealers, garden centres and mass retail. Garden equipment has always been an offline-first industry. Now the market is moving online: shoppers research long before they set foot in a store, and the marketplaces are opening the garden and outdoor category. STIGA saw the opening early and set an ambitious target, go live across every target marketplace, in every language, in parallel. ## The challenge The constraint was capacity, not ambition. Three things made the catalog genuinely hard. 1. **A deep catalog.** 500+ SKUs, each carrying technical specifications that had to be rendered accurately. 1. **A high average selling price.** The content could not just describe the products, it had to carry the value and justify the price. 1. **Proprietary technology.** Much of that value lived in engineering language, Vista, AGS, ePower, Multiclip, which says nothing to a marketplace shopper until someone rewrites it as a concrete benefit. On top of the catalog, time was binding. The traditional way to expand is sequential, one country and then the next, roughly six months each and one to two years in total. STIGA wanted the opposite. ### The setup before CommerceClarity Product data lived in Akeneo, a PIM designed to hold specifications and feed the trade channel, not to produce marketplace-ready content. Around it ran a manual, multi-vendor operation: source copy written in-house, translation tools for the other languages, a local marketing manager reviewing every product by hand, and spreadsheets holding the hand-offs together. The model worked. It did not scale. ## How we solved it, agents and engineers CommerceClarity replaced that chain with one agent platform, slotted into STIGA’s existing stack with no rip-and-replace of the PIM. The job: turn raw product data into channel-ready content, copy and structured attributes both, for Amazon in five countries, Allegro, Otto, Kaufland, ManoMano, Leroy Merlin and the direct eShop. ### The agents bring the scale Each product starts as a single canonical record, and a pipeline carries it from raw input to channel-ready content. 1. **Normalize** heterogeneous inputs into STIGA’s own taxonomy. 1. **Enrich** the missing attributes from STIGA’s internal sources. 1. **Apply the rules,** brand voice, channel constraints and compliance, so the output is correct by construction rather than corrected after. 1. **Translate and localize** natively, not word for word. 1. **Generate channel variants** in the exact shape each marketplace expects. 1. **Validate,** score the quality, run the automated checks, escalate what needs a human eye. STIGA started where the pain was sharpest, Channel Adaptation, then extended into Translation and Localization and Search Enrichment. ### The engineers co-build the outcome A CommerceClarity engineer works as an extension of STIGA’s team, the same person scoping, building and shipping. That flipped the operating model from decentralized and reactive to centralized and proactive. HQ owns the full content lifecycle and ships content already pre-validated and brand-ready. The agents bring the speed and the scale, the humans keep the governance over what actually goes live. ## Before and after Four months of work in a single view. - Time to launch · One to two years, country by country · Four months, in parallel - Marketplaces live · None · Seven - Product pages · Manual, per market · 500+ generated, validated, live - Languages · Generic translation tools, reviewed by hand · Five native localizations - Operating model · Internal team, tools, local managers, spreadsheets · One operator, centralized, end to end - Validation · Local managers, reactive · Automated checks, escalation only ## The people who own the rollout *In their words* - It feels like an extension of our team, like having their thinking embedded in an AI that works alongside us every day. · Asif Azam · Marketplace Manager, STIGA - Without CommerceClarity we would still be building the foundations. We are already building the next layer. · Gabriele Nofrini · Group E-Commerce Director, STIGA ## Proves area [Channels & Markets](/solutions/channels-markets) ## More customer stories - [Yeppon](/customers/yeppon) · How Yeppon got a 150,000 SKU catalog marketplace-ready, from its own store outward - [Cisalfa Group](/customers/cisalfa-group) · How Cisalfa Group got its catalog ready for the new season without growing the team ## See what the agents do with your catalog Bring one messy feed, or the backlog of products you never got to. We will show you the rest. [Book a demo](/book-a-demo) ## Questions ### What did STIGA achieve with CommerceClarity? Seven marketplaces live in five native languages in four months, with 500+ product pages generated, validated and published. Done the traditional way, country after country, the same expansion takes one to two years. ### Which marketplaces did STIGA launch on? Amazon in Italy, the United Kingdom, France, Germany and Poland, plus Allegro, Otto, Kaufland, ManoMano, Leroy Merlin and cDiscount, alongside STIGA’s own direct-to-consumer eShop. Each channel gets the product in the shape it requires, copy and structured attributes both. ### Did STIGA have to replace its PIM? No. Product data stayed in Akeneo, the PIM STIGA already ran for its dealer and retail business. The agents slot in around it: no rip-and-replace, and no second system for the team to keep in sync. ### Who did the work, agents or people? Both, and the split is the point. The agents carry the scale: normalize, enrich, apply the rules, translate, generate the channel variants, validate. A CommerceClarity engineer works as an extension of STIGA’s team, scoping and shipping inside the way STIGA already operates. People keep the governance over what goes live. ### Is the content translated or written for each market? Localized, not translated word for word. Five native localizations, with the proprietary technology names, Vista, AGS, ePower and Multiclip, rewritten as concrete benefits a marketplace shopper understands rather than left as engineering language. ### Are these figures approved by STIGA? Yes. Every figure and both quotes on this page are published with STIGA’s sign-off, and the two people quoted, Asif Azam and Gabriele Nofrini, are named with their consent. ## Entities - [STIGA](https://www.stiga.com) · Organization, about - [Channels & Markets](/solutions/channels-markets) · Thing, about - Asif Azam · Person, mentions - Gabriele Nofrini · Person, mentions - [Amazon](https://www.amazon.com) · Organization, mentions - [Leroy Merlin](https://www.leroymerlin.it) · Organization, mentions - [ManoMano](https://www.wikidata.org/wiki/Q48757787) · Organization, mentions - [Kaufland](https://en.wikipedia.org/wiki/Kaufland) · Organization, mentions - [Cdiscount](https://en.wikipedia.org/wiki/Cdiscount) · Organization, mentions - [Akeneo](https://www.akeneo.com) · Organization, mentions ## Sources - [Stiga Group](https://en.wikipedia.org/wiki/Stiga) · Wikipedia - [STIGA, garden machinery and outdoor power equipment](https://www.stiga.com) · STIGA Group - [Listing quality and product data requirements](https://sellercentral.amazon.com/help/hub/reference/GT4CDCFTFPWKMYNW) · Amazon Seller Central - [Akeneo, product information management](https://www.akeneo.com) · Akeneo --- # How Yeppon got a 150,000 SKU catalog marketplace-ready, from its own store outward > Yeppon, one of Italy’s leading electronics ecommerce players, runs 150,000+ SKUs across more than twelve European marketplaces. With CommerceClarity it processed around 80,000 SKUs end to end at roughly eight times its previous speed and activated around 24,000 products across channels, about 19,000 on Leroy Merlin and 5,000 on Amazon. Listing quality went from an average score near 42, below the auto-publish bar, to a target above 70. Every generated listing still passes human review. ![Inside a Yeppon warehouse.](https://cdn.sanity.io/images/xd4hrbt2/production/f1de727b7b2eba60c4e90ff637283725139561b7-800x600.jpg?auto=format) **Hero figure value:** 8x **Hero figure label:** Faster to market than the manual process **Industry:** Consumer electronics ## Key figures - 80k · SKUs processed end to end - 8x · faster go to market than the manual process - 24k · products activated across marketplaces - 150k+ · SKU catalog in scope ## Agents - Data Enrichment - Taxonomy Mapping - Matching - Leroy Merlin Listing Optimizer - Amazon Listing Optimizer ## Overview Yeppon is one of Italy’s leading electronics ecommerce players: 150,000+ SKUs across electronics, appliances, DIY and home furniture, sold on its own store and on more than twelve European marketplaces. Every product is not one listing but a set of variants, different by channel and by language. The target was both surfaces at once: get the product data right at catalog scale on its own site, and get the catalog onto the marketplaces, starting with Leroy Merlin and Amazon. ## The challenge Before us the catalog was largely a manual job. A part-time resource added new SKUs, enrichment leaned on a third-party feed and on whichever suppliers sent good data, and filters like screen size, RAM and disk were pieced together by hand. Around seventy per cent of listings got created or updated, many with no photo, on an irregular cadence. For fifteen years the economics concentrated the real effort on the twenty per cent of the catalog around Apple, appliances and TVs. The rest went to the marketplaces matched on price and availability alone, never enriched. ### A partner already ahead on AI Yeppon is one of the most AI-forward teams we work with. They had already built their own ChatGPT agents to write titles, descriptions and bullet points. That instinct took them a long way, and then hit the limits of the tool: output that would not stay consistent run to run, gaps filled from the model’s training rather than from the product, images slow and expensive at catalog scale, and no proof a listing would clear a marketplace’s quality bar before it went up. A catalog is a harder problem than a front end. Tens of thousands of products, a quality bar that keeps moving, and results that have to stay provable run after run. That is a system to engineer, not a tool to run. ## How we solved it, agents and a Retail Engineer ### First, their own store Yeppon runs its store on Shopify, built by the team itself. The agents fill in each product, its details, specs and images, and file it under the right category, through an API integration into Yeppon’s own flows. It went live on the Apple range first, proved out on a batch, then rolled across the catalog, and today it runs on its own. ### Then, the marketplaces Every marketplace is a walled garden with its own rules. Three things made this hard. Leroy Merlin scores listings and only publishes above the bar, and Yeppon sat well under it, so close to half the catalog could not be listed cleanly. Nobody hands you the formula, not even the marketplace’s own team. And the numbers are big: around 37,000 brands to match to their codes, and 63,000 product details to fill. 1. **Match** Yeppon’s brands and barcodes to the marketplace’s own product codes. 1. **Create** listings for products not on the marketplace yet. 1. **Improve** existing listings on what actually lifts the score: filled-in details, images, specs, required fields. 1. **Check against the bar** before anything goes up. The same agents work on any category. They were proven on four to start, drills, planters, garden hoses and hose reels, then pointed at the rest. This phase is still running: around 24,000 products are live across channels so far, roughly 19,000 on Leroy Merlin and 5,000 on Amazon. ### The Retail Engineer The agents do not work alone. A CommerceClarity Retail Engineer works as an extension of Yeppon’s team, the same person scoping, building and shipping, inside the way Yeppon already operates. It also gave the catalog something it had gone without for fifteen years: a dedicated owner. Yeppon’s marketing team stays in command, and every generated listing passes human review before it goes online. ## Before and after The do-it-yourself approach against the system that replaced it. - Cost · Unpredictable, hundreds of thousands of tokens for one operation · A controlled pipeline with a predictable cost - Marketplace readiness · Average score around 42, half the catalog under the auto-publish bar · Target around 70, auto-publish on Leroy Merlin and Amazon - Scale and consistency · Unthinkable by hand across 150k+ SKUs · Any category, output scored against a fixed quality bar - Operating model · Could not realistically be shipped in house · Agents and an embedded engineer, people deciding what goes live ## A team that already knew where AI stops *In their words* ### Danilo Longo We had built our own agents on ChatGPT before this, so we knew where AI helps and where it stops. The agents handle the volume across the full catalog, but every listing still passes our team’s review before it goes live. We got the scale without giving up the final call on quality. General Manager, Yeppon ![Danilo Longo, General Manager at Yeppon](https://cdn.sanity.io/images/xd4hrbt2/production/0ae222ca3b1dbe30a40c5d7606c672ba4d77c25a-400x400.jpg?auto=format) ### Valerio Di Dio It connects straight into our flows through the API, so we shape content around what each product actually needs instead of running one template across the board. The same product can go out with a different angle on Amazon than on Leroy Merlin, and the speed holds even when we switch language. Reporting Analyst, Yeppon ![Valerio Di Dio, Reporting Analyst at Yeppon](https://cdn.sanity.io/images/xd4hrbt2/production/7904ce6621ffc12a5298693b69be828e988b7b0f-400x400.jpg?auto=format) ## Proves area [Channels & Markets](/solutions/channels-markets) ## More customer stories - [STIGA](/customers/stiga) · How STIGA launched on seven marketplaces in four months - [Next Hardware & Software](/customers/next-hardware-software) · How enriched product pages lifted search and conversion for Next ## See what the agents do with your catalog Bring one messy feed, or the backlog of products you never got to. We will show you the rest. [Book a demo](/book-a-demo) ## Questions ### What did Yeppon get out of it? Around 80,000 SKUs processed end to end at roughly eight times the previous speed, the full catalog in shape on its own store, and around 24,000 products activated across marketplaces, about 19,000 on Leroy Merlin and 5,000 on Amazon. ### Yeppon had already built agents on ChatGPT. Why was that not enough? It worked until catalog scale. The output was hard to keep consistent product to product and run to run, the model filled gaps from its training rather than from the product, images were slow and expensive across 150,000+ SKUs, and nothing proved a listing would clear a marketplace’s quality bar before it went up. A catalog is a system to engineer, not a tool to run. ### What is a marketplace listing quality score? Marketplaces grade the listings you send them and publish automatically only above a threshold. On Leroy Merlin, Yeppon’s catalog averaged around 42 and close to half of it could not be listed cleanly. The target is above 70. Nobody publishes the formula, not even the marketplace’s own team, so getting there is testing and improving rather than a one-shot upload. ### How much matching does a catalog this size need? Around 37,000 brands to line up against Leroy Merlin’s own codes and around 63,000 product details to fill. The agents were proven on four categories first, drills, planters, garden hoses and hose reels, then pointed at the rest. ### Does anything publish without a person seeing it? Not at Yeppon. The agents bring the volume, and Yeppon’s marketing team reviews every generated listing before it goes online. That was their condition, and it is how the work runs. ### Is the work finished? No. The marketplace phase is still running, and by Yeppon’s own reckoning the programme is around two thirds of the way there. Next up are organic growth, new European markets, and using the same system for both organic and agentic demand. ## Entities - [Yeppon](https://www.yeppon.it) · Organization, about - [Channels & Markets](/solutions/channels-markets) · Thing, about - Danilo Longo · Person, mentions - Valerio Di Dio · Person, mentions - [Leroy Merlin](https://www.leroymerlin.it) · Organization, mentions - [Amazon](https://www.amazon.com) · Organization, mentions - [Shopify](https://www.shopify.com) · Organization, mentions - [ChatGPT](https://en.wikipedia.org/wiki/ChatGPT) · Product, mentions - [Cdiscount](https://en.wikipedia.org/wiki/Cdiscount) · Organization, mentions ## Sources - [Yeppon](https://www.yeppon.it) · Yeppon - [Listing quality and product data requirements](https://sellercentral.amazon.com/help/hub/reference/GT4CDCFTFPWKMYNW) · Amazon Seller Central - [Leroy Merlin](https://en.wikipedia.org/wiki/Leroy_Merlin) · Wikipedia - [Mirakl, the marketplace platform behind European retail marketplaces](https://www.mirakl.com) · Mirakl - [Shopify](https://www.shopify.com) · Shopify --- # ChatGPT product feed: nine required fields, and the optional ones that decide what gets surfaced > OpenAI Ads Manager can build campaigns from a retailer's catalog. Its product feed spec requires nine fields per row. A Google-compatible feed is accepted, but it changes availability, seller identity and eligibility behavior. During the beta, products from an ads feed can appear in ads only, not organic ChatGPT conversations. A ChatGPT product feed needs nine required fields. Optional attributes give OpenAI the data to filter, compare and surface the right products. **Published at:** 2026-09-18T16:10:25Z A **ChatGPT product feed** needs nine required fields. Optional attributes give OpenAI the data to filter, compare and surface the right products. Uploading the file is the short part. Fixing the catalog underneath it is the work. ## What OpenAI shipped Retail advertisers can now upload a catalog into [OpenAI Ads Manager](https://ads.openai.com) and build product-feed campaigns from it. One limit matters before anything else: during the current beta, products from an ads feed can appear in ads only. OpenAI says they do not appear in organic ChatGPT conversations today. The feed still matters beyond the campaign. OpenAI uses the same product schema for search eligibility, checkout eligibility and ads eligibility. The flags change. The product data does not. That makes this an [agentic commerce readiness](https://commerceclarity.com/solutions/agentic-commerce) problem before it becomes a media-buying problem. The timeline matters because the product schema predates the current ads use case. That is why it still carries separate controls for search, checkout and ads. ![Four-milestone timeline. September 2025, Instant Checkout launches and people buy inside ChatGPT. March 2026, OpenAI retreats to discovery and merchants keep checkout. June 2026, feeds appear in Ads Manager and ads are generated from catalog rows. September 2026, the spec goes public with nine required fields documented.](https://cdn.sanity.io/images/xd4hrbt2/production/5ca87360f5ed698159c0410b750bd148c25555df-3200x1440.png?auto=format) *The feed OpenAI asks for today is the one it built to take orders. Nothing about the schema changed when the purpose did.* ## The nine required fields in the OpenAI product feed spec The **OpenAI product feed spec** requires these nine fields on every row. Nine sounds manageable until variants enter the picture. Each variant needs its own item_id, price, availability, URL and images. A proper variant group also needs group_id, listing_has_variations and variant_dict. If color and size live only in the front end, the exported rows lose the relationship the shopper sees on the page. Uploaded items also expire after two weeks. OpenAI recommends a hosted URL or automated SFTP for catalogs that change often. A manual upload is a test, not an operating model. ## Reuse the Google Shopping feed, then inspect what it drops OpenAI accepts a Google-compatible product feed after it confirms that format for the registered feed. Use it. It is the fastest way to get live. Then inspect the gaps. [Product feed management](https://commerceclarity.com/solutions/channels-markets) is the work of adapting one catalog to each channel's schema, and the compatibility path has its own rules: - availability uses preorder, not pre_order, and does not accept unknown. - availability_date becomes required for preorder and backorder. - seller_name comes from the registered merchant name, not the uploaded row. - accepted products have search enabled and checkout disabled. An uploaded is_eligible_search=false does not opt a product out. - structured title and description alternatives are not mapped. The Google feed is a useful starting point. It was built for a results grid, where the image and title do most of the work. A model filters before it explains. That puts more weight on attributes the Google feed may never have needed. ## Field by field: OpenAI vs Google Merchant The names look similar. The requirements are not. Google Merchant has a core set required for every product, then conditional requirements based on category, identifier, market and destination. OpenAI requires the same nine base fields on every row. Its Google-compatible path accepts only a defined subset of the full Merchant Center schema. The useful shortcut is the shared core: id, title, description, link, image, availability, price and brand. The dangerous assumption is that every other Google Merchant attribute comes across with it. It does not. OpenAI documents a compatibility profile, not a full import of the [Google Merchant product data specification](https://support.google.com/merchants/answer/13889434?hl=en). ## The optional fields carry the comparison The required fields tell OpenAI that a row is valid. Optional fields describe enough of the product to answer a specific request. Take a simple query: waterproof trail shoe, black, size 10, under 100 euro. The feed needs color, size, material, price and availability to answer it cleanly. The spec also supports gtin, mpn, condition, product_category, gender, age_group, dimensions, weight, sale_price, shipping_price, returns, review_count, star_rating and additional images. OpenAI is careful here: eligibility does not guarantee display. Still, a missing attribute gives the system no structured value to filter on. This is a [catalog foundation](https://commerceclarity.com/solutions/catalog-foundation) problem. The same constraint showed up when we read the [Claude Commerce Agents code](https://commerceclarity.com/blog/claude-commerce-agents-catalog-constraint): filters read fields first. Text ranks what survives. ![Two cards compare required and optional fields. The required card lists item_id, title, description, url, brand, seller_name, image_url, availability and price. The optional card lists identifiers, product attributes, dimensions, price and shipping fields, returns, reviews, images, custom labels and ads metadata used for filtering and comparison.](https://cdn.sanity.io/images/xd4hrbt2/production/b81355ff0db678c4e7fd8b98bf7985d7300262a6-3200x1960.png?auto=format) *Nine fields decide whether the row is accepted. Optional attributes give OpenAI more data to filter and compare it.* ## Ads and organic discovery are separate today The schema includes three controls: - is_eligible_search controls search eligibility. - is_eligible_checkout controls direct purchase and depends on search eligibility plus an enabled integration. - is_ads_eligible controls ads processing. But an ads feed does not create an organic ChatGPT presence during the beta. OpenAI says those products are eligible for ads only. So the immediate use case is clear: upload a compliant feed, mark the products eligible for ads, keep price and availability current, then select products in Ads Manager. The longer-term value is the catalog work behind it. The same complete product record can be mapped to another marketplace, on-site search, a PDP or a future AI shopping surface without rebuilding the knowledge each time. ## What this asks of your catalog Most teams will not struggle with the CSV. They will struggle with the rows. A title has to name the variant without turning into a keyword dump. Availability has to match the enum and be current. GTINs need their leading zeros. Dimensions need units. Material, color and size have to exist as fields rather than clues buried in prose. CommerceClarity agents generate and validate those attributes against the retailer's rules. [Governance](https://commerceclarity.com/platform/governance) keeps uncertain values in review before they reach a channel. The feed is the output. The catalog process decides whether it works. [Book a demo](https://commerceclarity.com/book-a-demo) and we will audit a sample of your catalog against the OpenAI product feed spec, field by field. ![Small silver fish seen from above in deep blue water, the surface ripples breaking the light into white patches.](https://cdn.sanity.io/images/xd4hrbt2/production/da396337df2ff69c7d5e0c1ecd5dd4ba4087bd84-1600x1000.jpg?auto=format) ## Author [Daniele Vella](/author/daniele-vella) · Co-founder, CommerceClarity ## Questions ### What is a product feed, and do I need one for ChatGPT? A product feed is a structured file with one row per SKU or variant. It carries the fields a platform uses to show, filter and price the product. You need one to run product-feed campaigns in ChatGPT Ads Manager. ### What are the nine required fields for a ChatGPT product feed? item_id, title, description, url, brand, seller_name, image_url, availability and price. Variants add conditional fields such as group_id, listing_has_variations and variant_dict. ### Can I reuse my Google Shopping feed? Yes, after OpenAI confirms the Google-compatible format for your registered feed. Check the different availability values, seller identity, eligibility controls and unsupported structured title and description fields before treating it as finished. ### What do I lose by reusing the Google feed? You lose per-item search controls on that compatibility path, checkout stays disabled, and only the supported profile reaches OpenAI. The larger risk is catalog depth: a feed designed for a visual shopping grid may be thin in the optional attributes a model needs to filter products. ### Will the feed make my products appear in normal ChatGPT answers? No, not during the current ads beta. OpenAI says products from an ads feed are eligible for ads only and do not appear in organic ChatGPT conversations. ### How do I run ads on ChatGPT from a product feed? Create a feed in Ads Manager, import it by file, hosted URL or SFTP, fix validation errors, select the feed in a product-feed campaign, filter the eligible products and launch the campaign. Keep the feed fresh because uploaded items expire after two weeks. ## Entities - [OpenAI](https://openai.com) · Organization, about - [ChatGPT](https://chatgpt.com) · Product, about - [Google](https://www.google.com) · Organization, mentions - [Agentic Commerce](/solutions/agentic-commerce) · mentions - [Catalog Foundation](/solutions/catalog-foundation) · mentions - [Channels & Markets](/solutions/channels-markets) · mentions - Claude Commerce Agents: the agent got easy to build. The catalog decides what it can enforce. · mentions ## Sources - [Create Campaigns from Product Feeds](https://help.openai.com/en/articles/20001268-create-campaigns-from-product-feeds) · OpenAI Help Center - [Product Feeds (Ads)](https://developers.openai.com/ads/product-feeds) · OpenAI Developers - [Products: file upload schema](https://developers.openai.com/commerce/specs/file-upload/products) · OpenAI Developers - [Product feeds (Agentic Commerce)](https://developers.openai.com/commerce/specs) · OpenAI Developers - [Free listings and required product attributes](https://support.google.com/merchants/answer/13889434?hl=en) · Google Merchant Center Help --- # Claude Commerce Agents: the agent got easy to build. The catalog decides what it can enforce. > On 2 September 2026 Anthropic published Claude Commerce Agents, an open blueprint for a shopping agent and a merchant agent, and Shopify shipped its own implementation over the Universal Commerce Protocol the same day. Reading the code shows what the catalog has to provide: a condition the shopper states must exist as a queryable field, hold a value a machine can compare, and carry the rules for the market and channel where it is used. Filters read fields, text only ranks. Anthropic and Shopify made the shopping agent easy to build. We read the code to show what it asks of the catalog underneath, and why creating the field is only the start. **Published at:** 2026-09-03T08:00:00Z On 2 September 2026, Anthropic published [Claude Commerce Agents](https://github.com/anthropics/commerce-agents): an open blueprint for a shopping agent in the storefront and a merchant agent for the back office. It includes reference implementations for retail, travel, telecom and ticketing, plus a Claude Code plugin to adapt the pattern to a company's own systems. [Shopify published its implementation](https://github.com/Shopify/claude-for-commerce-examples) on the same day, connecting a storefront agent to a live Shopify store through the Universal Commerce Protocol. This is why the conversation has changed. [The agent layer is becoming practical infrastructure](https://commerceclarity.com/blog/llms-txt-webmcp-ecommerce) for discovery, conversational shopping and purchase flows. Anthropic reports that retailers already running shopping agents on Claude have seen larger carts and higher completion rates. Those are Anthropic's figures, not ours. The useful signal is that a retailer can now start from a working architecture rather than a blank page. That is the operational side of [agentic commerce](/solutions/agentic-commerce). We cloned the repository, ran the retail example and read the code behind the shopping flow. It makes one thing very clear: the agent can only apply the product conditions that the catalog gives it in a usable form. ![The agent layer, holding the shopping agent and the merchant agent, sits on the catalog layer of product records, commerce systems and operating rules. An agent answers only what the catalog holds.](https://cdn.sanity.io/images/xd4hrbt2/production/67985d041ac4a192485fec2e5c3e9c13cea11575-3200x2600.png?auto=format) ## The blueprint ships the agent, not the catalog Claude Commerce Agents supplies the agent loop, the skills, the tool contracts, memory, presentation rules, safety gates and evaluation harnesses for both sides of commerce. A team can run that pattern on the Messages API, the Claude Agent SDK or Claude Managed Agents. The shopping agent covers discovery, purchase research, planning, customer care and personalisation. The merchant agent covers performance, listings, inventory, pricing and campaigns. The retailer's catalog is outside that blueprint. `StorefrontBackend` is an interface that each company connects to its own product, cart, order and policy systems. The retail example uses 87 fictional ACME products across 11 categories. The **AI shopping agent** sees only the product record that this connection returns. ## How the Claude Commerce Agents shopping agent searches your catalog The shopping agent does not inspect the catalog on its own. It calls `search_products` with a query, filters and a limit of eight results. The important part is how the blueprint tells it to construct that call. The search skill is direct: use the catalog's vocabulary for the query, and put each condition the shopper explicitly states into a filter. A request for a bed frame with nine inches of clearance and a 700 lb weight capacity becomes a search for "platform bed frame", a price ceiling and two product-attribute filters. The query finds candidates. The filters decide whether candidates qualify. The retail example shows why this distinction matters. It ranks text with the title weighted highest, then brand, category, attributes and description. It keeps results within half of the top score. Only afterwards does it apply filters. Those filters inspect attributes, titles and option values. They do not inspect the specification sheet or the description. If a filter would return nothing, the demo backend removes it and sends back the ranked list. The agent is told that those results may be related items, not the requested product. A shopper condition remains enforceable only when the catalog holds it as a field. ![Five steps from a shopper's sentence to the eight records the agent sees: the skill phrases the search, the backend ranks on text, applies the filters over attributes, title and option values, drops a filter that would empty the result, and returns compact records without the spec sheet.](https://cdn.sanity.io/images/xd4hrbt2/production/84501dd661ce0b949f0b40d4dcad2066ea5a3778-3200x2600.png?auto=format) A production backend may use Elasticsearch, Algolia, Factfinder, a vector index or another search service. The ranking will differ. The requirement does not: product conditions need fields, text ranks candidates, and the agent can only work with the record search returns. ## Three things this asks of your product data ### A stated condition has to be a field, in the words the agent will use In Anthropic's demo catalog, 63 facts sit in the spec sheet of 29 products and nowhere else: under-bed clearance, weight capacity, tent floor area and whether a suitcase has a built-in lock. A shopper can read those facts on a product page. The filter cannot use them, so a request that depends on one may return a shortlist of look-alikes instead of a qualified product. We moved those facts into [**product attributes**](/blog/chatgpt-product-feed-required-fields) with a 48-line script. The merchant agent's `catalog-listings` skill follows the same discipline: it can fill a missing attribute from a value already in the record, and it stops when the record does not contain the fact. That is a small example of the difference between writing product copy and creating a product fact an agent can use. ![One product record as six fields in one column. Above the line the search reads title, attributes, option values and description. Below it the specification sheet and the long description are read only on request, one product at a time, and the weight capacity sits there.](https://cdn.sanity.io/images/xd4hrbt2/production/36c9696725e4fd1d31565fa5760bd53bdb590fa5-3200x2600.png?auto=format) ### The value has to be comparable A field alone is not enough. The demo filter compares strings. `weight` appears on 14 products as 254 g, 2.1 kg, 35 lb, 2 lb and 2 x 9 oz. `size` can mean a bed size, a rug dimension, a storage cube, a bottle or a grocery pack. The bed frame carries `weight_capacity: 700 lb`, and a filter for "700 lbs" does not bind, so the backend drops it and returns a ranked list. **Structured product data** needs both a field and a comparable value. That means one unit, one format and one allowed vocabulary for each category. It is the difference between information a person can interpret and information a system can apply. ![The attribute weight on five products in five formats, 254 g, 2.1 kg, 35 lb, 2 lb and 2 x 9 oz, and a record value of 700 lb next to an agent filter of 700 lbs that does not match, so the filter is dropped.](https://cdn.sanity.io/images/xd4hrbt2/production/868333bc5e28b5d089fc7e7eaa12f05049863960-3200x2600.png?auto=format) ### What lives in the description is invisible until someone asks for it A search result gives the agent a title, price, attributes and short description. The specification sheet and long description arrive later, one product at a time, through `get_product_details`. In the small retail demo, the right product is usually among the eight results and a careful agent can check it before answering. An enterprise catalog changes the stakes. A product whose deciding fact lives only in prose has to win a text-ranking contest before the agent can inspect it. A fact that the filter cannot read never becomes a condition on the shortlist. ## Making it a field is the beginning On the demo catalog, promoting a fact into a field is a script. In an enterprise retail catalog, the same move raises the questions that decide whether the field can be trusted: which attributes a category needs, where each value comes from, which variant it belongs to, and which version may be used in each market and channel. Some facts are accurate in one context and unusable in another. The demo sunscreen has `water_resistant: 80 min`, a US-style claim. Its description also says "reef-safer formula" without an evidence field behind it. In Europe, environmental and product claims face tighter consumer-protection rules from 27 September 2026. The merchant-agent guardrails already reflect the principle: some compliance and tax information cannot be changed in bulk by an agent. That is **product data governance**. A retailer defines the attributes each category requires, connects every value to a source, keeps it current, normalises the units and allowed values, then applies the correct version by variant, market and channel. **Product data quality** is an operating discipline, not a cleanup exercise. It is what makes **product data for AI agents** reliable. ![Five checks from a field to a value an agent can trust: field, value, source, variant and market, with a return loop underneath because supplier files change, ranges move and marketplaces add required fields.](https://cdn.sanity.io/images/xd4hrbt2/production/07325f84b221a2a8e18ec2c088bcbca382526861-3200x2600.png?auto=format) An **agent ready catalog** is one where every fact a shopper may state as a condition exists as a field, holds a comparable value and carries the rules for where it can be used. Creating the field solves the interface problem. Keeping the value trustworthy across the catalog is the enterprise work. CommerceClarity works alongside the PIM, ERP, DAM and supplier systems a retailer already runs. [Catalog Foundation](/solutions/catalog-foundation) turns scattered product information into usable records. [Context](/platform/context) connects those records to company rules, systems, markets and category requirements. [Compliance and Quality](/solutions/compliance-quality) applies those rules before a value reaches a channel. ## Where to start Start with the product questions the agent needs to answer. Take them from on-site search, customer care, merchandising and the sales floor. For each question, ask whether the catalog can identify the right product rather than a plausible shortlist. Check whether the deciding facts are fields with comparable values. Then check whether those facts are valid for the right variant, market, channel and category rules. Fix one category, validate it against an evaluation set, then extend the work category by category and channel by channel. The agent is becoming easier to build. The catalog work decides what it can enforce. If you want to run that check on your own catalog, [book a pilot](/book-a-demo). ![A single track running along a hillside under a wide open sky, the land falling away out of focus.](https://cdn.sanity.io/images/xd4hrbt2/production/9bd3bfd1a13a29f9bcd2e6b590b56c0f86846296-1600x1000.jpg?auto=format) ## Author [Daniele Vella](/author/daniele-vella) · Co-founder, CommerceClarity ## Questions ### What are Claude Commerce Agents? Claude Commerce Agents is Anthropic's commerce agents blueprint, published on 2 September 2026: an open reference for building a shopping agent and a merchant agent on Claude. It includes the agent architecture, skills, tool contracts, safety patterns and evaluation harnesses, reference implementations for retail, travel, telecom and ticketing, and a Claude Code plugin. ### What is Claude Commerce? Claude Commerce, or Claude for Commerce, is Anthropic's umbrella for commerce use of Claude: the Claude Commerce Agents blueprint, the partner implementations from companies such as Shopify, and the solutions pages around them. The blueprint is the part a team can clone and run. ### Do Claude Commerce Agents include a product catalog? No. The blueprint ships a fictional demo catalog of 87 products for its retail example. A retailer implements the StorefrontBackend interface over its own catalog, cart, order and policy systems, and the agent knows about a product only what that interface returns. ### What is the difference between the Agentic Commerce Protocol and the Universal Commerce Protocol? The Agentic Commerce Protocol, developed by OpenAI and Stripe, is how ChatGPT hands a purchase to a merchant's checkout. The Universal Commerce Protocol is an open protocol for catalog, cart and checkout that Shopify exposes on its stores, and the one Shopify's Claude Commerce Agents implementation runs on. Both standardise the transaction between an agent and a merchant. Neither defines what a product record has to contain for the agent to answer a real question about it. ### What is enterprise agentic commerce? Agentic commerce is the shift from a person browsing a storefront to an agent querying the catalog on their behalf. The hard part in an enterprise is not the agent, which is now largely a solved build. It is that volume of product information used to be the advantage, and an agent rewards correctness instead: it applies conditions through filters, and a filter reads a field. Winning enterprise agentic commerce means publishing the right product information, not more of it. ### Why does structured product data matter for an AI shopping agent? Because the agent applies a shopper's conditions through filters, and filters read fields. Text influences ranking, and a requirement stated only in a description or a specification sheet can be read by a person and never applied as a constraint. Structured product data with comparable values is what turns a plausible shortlist into an answer. ### What is an agent ready catalog? A catalog in which every fact a shopper might state as a condition exists as a field, holds a value a machine can compare, and carries the rules for the markets and channels where it may be used. Adding product attributes is the first step. Product data governance is what keeps them true. ## Entities - [Claude Commerce Agents](https://github.com/anthropics/commerce-agents) · Thing, about - [Anthropic](https://www.anthropic.com/) · Organization, about - Product data governance · Thing, about - [Shopify](https://www.shopify.com/) · Organization, mentions - [OpenAI](https://openai.com/) · Organization, mentions - [Stripe](https://stripe.com/) · Organization, mentions - [Universal Commerce Protocol](https://shopify.dev/docs/agents) · Thing, mentions - [Product information management](https://en.wikipedia.org/wiki/Product_information_management) · Thing, mentions ## Sources - [Building commerce agents with Claude](https://claude.com/blog/claude-for-commerce-agents) · Anthropic - [anthropics/commerce-agents](https://github.com/anthropics/commerce-agents) · Anthropic - [commerce-agents at commit fd4d592, the tree we read](https://github.com/anthropics/commerce-agents/commit/fd4d59224ab96b43c6dc6888207c67b3bd5a24cf) · Anthropic - [Shopify/claude-for-commerce-examples](https://github.com/Shopify/claude-for-commerce-examples) · Shopify - [Build commerce agents with the Universal Commerce Protocol](https://shopify.dev/docs/agents) · Shopify --- # Your ecommerce has an AI agent score. > Chrome's Lighthouse has an experimental Agentic Browsing category with four areas: accessibility, layout stability, an llms.txt check and WebMCP integration. CommerceClarity implemented WebMCP on its own site and recorded 4 of 4. Google Search says llms.txt does not affect ranking; WebMCP is a W3C Community Group draft in a Chrome origin trial. For an enterprise retailer: run the test, pilot one action, keep llms.txt to key pages, and prepare catalog data and processes for agentic commerce. We ran Lighthouse’s experimental Agentic Browsing checks on our own site and read what they measure. What a retailer should do with the result, what to try with WebMCP and llms.txt, and where the real preparation for agentic commerce happens: the catalog. **Published at:** 2026-09-14T12:00:00Z We implemented WebMCP on CommerceClarity’s website and recorded a 4/4 result in Lighthouse’s experimental Agentic Browsing category. I wanted to understand what the checks could tell us before recommending any of this to a retailer. They tell you something useful about how an agent can interact with a page. They don’t tell you whether it can choose the right product, trust its specifications or make a purchase your business can actually fulfill. That distinction matters if you run an enterprise retailer. I would run the test, fix the obvious problems and make room for a small experiment. I would put most of the effort into preparing the catalog and the processes behind it for [agentic commerce](https://commerceclarity.com/solutions/agentic-commerce). ![PageSpeed Insights report for commerceclarity.com with Lighthouse's experimental Agentic Browsing category showing four of four checks passed, next to the Performance, Accessibility, Best Practices and SEO scores.](https://cdn.sanity.io/images/xd4hrbt2/production/d02a61abf2adde0a4f61db449944b264aa36dfdb-1410x1066.png?auto=format) *Our own result, recorded on 14 September 2026 with the experimental category enabled. A pass ratio, not a score out of 100.* ## What Google’s agentic browsing checks actually measure Google has added an experimental Agentic Browsing category to Lighthouse. Its documentation describes four areas: accessibility, layout stability, discoverability through llms.txt, and WebMCP integration. The report uses a pass ratio, rather than the familiar score out of 100. [Lighthouse scoring documentation](https://developer.chrome.com/docs/lighthouse/agentic-browsing/scoring) **1. Accessibility: can the agent identify what’s on the page?** Buttons, inputs and links need meaningful names and roles. A button labeled “Check availability” gives the agent more to work with than an unlabeled icon. **2. Layout stability: does the target stay where it was?** If a banner pushes the page down just before a click, the agent can hit the wrong control. Lighthouse uses Cumulative Layout Shift to assess this. **3. Discoverability: is there an llms.txt file?** This is a short guide to the site’s content. The check is limited: a server error is flagged, while a missing file returns N/A. A pass does not validate the entire file or prove that agents use it. [llms.txt audit](https://developer.chrome.com/docs/lighthouse/agentic-browsing/llms-txt) **4. WebMCP integration: has the site declared tools an agent can use?** This area includes checks for registered tools, forms missing declarative WebMCP and schema validity. These are separate audits within one area, which is why four areas should not be confused with four individual tests. [Google’s toolkit overview](https://developer.chrome.com/blog/agent-ready-toolkit) I would ask your team to test the homepage, a product page and a shipping or returns page. Try [PageSpeed Insights](https://pagespeed.web.dev/); if the experimental category isn’t available there, use the supported Lighthouse setup described by Google. The documentation requires Chrome 150 or later, with origin-trial registration for the WebMCP audits. Keep the individual results, including N/A, alongside the page URL and test environment. A fraction without that context is easy to overread. ![The four areas covered by Lighthouse Agentic Browsing checks: identify the control, keep the target still, find key content, call a declared action.](https://cdn.sanity.io/images/xd4hrbt2/production/3131bed807169fe22bf4f21929357787534efeb4-3200x1840.png?auto=format) *Experimental checks. Not a ranking, and not a sales forecast.* ## What is WebMCP, and why did we implement it? An agent using a website can inspect the interface, find a control and interact with it. That can work, but a complicated form leaves plenty of room for mistakes. WebMCP gives the agent another route. The website declares an action, explains what it does and specifies the information it needs. Instead of working out how to complete a booking form, the agent can use a tool designed for that booking. The current draft supports tools implemented in the page’s JavaScript. [WebMCP specification](https://webmachinelearning.github.io/webmcp/) MCP, short for Model Context Protocol, is the broader approach to connecting AI applications with tools and data. WebMCP brings a related tool interface into the browser. It does not mean creating an MCP server for every product. On our site, we implemented tools to read page content, discover the site’s pages and request a demo. We wanted to learn what happens when an agent can use our existing content and actions directly. CommerceClarity is an AI-native company; testing this on our own website is part of how we form an opinion before taking it to customers. Google is actively developing and testing WebMCP in Chrome. The specification also has editors from Microsoft. But the document is still a W3C Community Group draft, explicitly **not a W3C Standard**. Chrome’s [origin trial](https://developer.chrome.com/blog/ai-webmcp-origin-trial) is a time-limited experiment intended to gather feedback. Official documentation is evidence of serious work, not a promise that today’s API will remain unchanged. ![One action, two ways for an agent to reach it: the interface route and the declared WebMCP tool both land in the same demo request process, with user permission before submission.](https://cdn.sanity.io/images/xd4hrbt2/production/769b54b9c9b5592e038944051a36e51a204aa4ef-3200x2480.png?auto=format) *WebMCP exposes an action. The business still owns what happens next.* ## For physical-goods retail, I would start with a small pilot A booking, appointment or quote request is where I would look first. It has a clear task, a defined set of inputs and an outcome you can check. Retail has useful candidates too: checking store availability, configuring a product or adding the right variant to a cart. I would not rule those out. I would ask the team to demonstrate that a declared tool makes one of them more reliable than the existing flow. What I would avoid today is turning WebMCP adoption into a broad enterprise program just to improve the report. For physical goods, the action depends on information outside the page: current inventory, regional prices, variant relationships, delivery eligibility and returns conditions. A tool can expose “check availability.” It cannot make the underlying inventory reliable. You still need permissions, confirmation for consequential actions, failure handling and someone responsible when the result is wrong. Other approaches are developing alongside it. The [Universal Commerce Protocol](https://developers.googleblog.com/under-the-hood-universal-commerce-protocol-ucp/) describes commerce capabilities and interactions such as checkout. It is worth following, but it doesn’t make WebMCP irrelevant or validate your product data for you. My recommendation is to choose one useful action, test it with a supported agent and measure completion, errors and recovery. Expand when the evidence warrants it. ## What is an llms.txt file, and should your ecommerce have one? It is still being developed. The [proposal was updated to v2 in August 2026](https://llmstxt.org/), and it remains a proposal. An llms.txt file gives agents a concise introduction to a site and links to useful content. Markdown versions of those pages are a related choice. A full-text bundle, often called llms-full.txt, is another. You do not have to produce a Markdown copy of every product page to maintain a useful index. Google Search says llms.txt neither helps nor hurts visibility or ranking in its search results. Lighthouse checks it for a different purpose: interaction with agents. Those positions are compatible. [Google Search guidance](https://developers.google.com/search/docs/fundamentals/ai-optimization-guide) For an established retailer with a capable digital team, I would still do a limited version. Start with the pages that explain the business and help a buyer make a decision: the brand story, important category introductions, buying guides, delivery, returns and warranty information. Where practical, offer clean Markdown versions from the same content source. Give the file an owner and check its links when those pages change. The aim is to make the brand’s own explanation easier to retrieve. I would treat that as a small publishing experiment, not a proven ranking tactic. The broader work on [SEO and GEO for product catalogs](https://commerceclarity.com/solutions/seo-geo) concerns the content shoppers search for and the answers they need. I would not make Markdown versions of all 70,000 product pages the default next step. The cost depends on your architecture: automatic output from the same source is very different from a separately maintained copy. Either way, you need to keep prices, availability and product claims current wherever they appear. I would want evidence that an intended consumer uses that extra format before taking on the work. First, keep the product pages, structured data, feeds and commerce APIs accurate. Google already documents [product structured data and Merchant Center feeds](https://developers.google.com/search/docs/appearance/structured-data/product) as ways to provide product information. ### An llms.txt example for a retailer A small index might introduce the business and link to a buying guide and a returns policy. This illustrative llms.txt example uses placeholder URLs; replace them with working pages you maintain. `# Example Retailer` `> Home appliances, with buying guides and delivery information.` `## Buying guides` `- [Washing machine guide](https://example.com/guides/washing-machines.md): Sizes and features to compare.` `## Customer information` `- [Returns policy](https://example.com/help/returns.md): Conditions and steps for returning a purchase.` ### What should an llms.txt checker check? Check that the file follows the format your intended consumer supports, that its links resolve and that the linked content is current. If you provide an llms-full.txt bundle, check that output too. A file returning HTTP 200 can still contain broken links or outdated information. Lighthouse’s retrieval check does not replace those checks. ![Use a curated llms.txt index for key pages while keeping product outputs aligned to one catalog source.](https://cdn.sanity.io/images/xd4hrbt2/production/0e96d04fd5ef4bae9435ffa6c120540e0b760b27-3200x2120.png?auto=format) *One index for the pages you write. One source for the products you sell.* ## Enterprise agentic commerce starts with the catalog and its processes Imagine a shopper asking an agent for a washing machine that fits a particular space, meets an energy requirement and can be delivered upstairs next week. The answer needs more than a readable page. It needs dimensions with the right units, a reliable model identifier, accurate product information and a live delivery service that can check the address. Some of that belongs in the catalog. Some belongs in inventory, pricing and fulfillment systems. The systems have to agree on which product they mean. An agent may query those systems through APIs or feeds, or read the website itself. In each case, the quality of the answer depends on what the business has made available. We explored a concrete example in [our analysis of Claude Commerce Agents](https://commerceclarity.com/blog/claude-commerce-agents-catalog-constraint): a shopper’s condition can only work as a filter when the catalog exposes it in a usable field. That is the sense in which **the catalog is the new storefront**. Your product needs to be represented accurately enough to enter the comparison, meet the shopper’s constraints and support a purchase. The FAQ below spells out the phrase; the operational consequence is that catalog work now serves both people and software making choices for them. For an enterprise retailer, I would build that preparation into the existing catalog operation: - **Define the data each category needs.** Model variants, identifiers, units and the attributes that determine whether a product fits a buyer’s request. These are also the foundations of [on-site search and product discovery](https://commerceclarity.com/solutions/search-discovery). - **Verify it against a source.** Keep the [source information and category rules](https://commerceclarity.com/platform/context) behind a claim, identify conflicts and send uncertain cases for review rather than filling every blank with a plausible answer. - **Keep outputs aligned.** When an attribute changes, update the [channels and markets that use it](https://commerceclarity.com/solutions/channels-markets). Connect dynamic information such as stock and delivery to the systems that own it. - **Make quality a recurring process.** Assign owners, set [acceptance checks and human review rules](https://commerceclarity.com/platform/governance), and track exceptions from supplier input to published output. For physical products, preparation also includes applicable product safety and consumer-information obligations. Selling through an agent does not remove the need to supply accurate product, price and delivery information. Your teams need to determine the requirements for each market and category and include them in [catalog compliance and quality checks](https://commerceclarity.com/solutions/compliance-quality). [EU product safety guidance](https://europa.eu/youreurope/citizens/consumers/shopping/product-safety/index_en.htm), [EU consumer information guidance](https://europa.eu/youreurope/citizens/consumers/shopping/contract-information/index_en.htm) This work supports internal agents as well. If you want agents to classify products, enrich attributes or prepare channel content, you need to define the inputs, rules, evaluation and exception handling. Connecting an agent to an inconsistent catalog simply gives it an inconsistent catalog to work with. That is the work we do through [catalog foundation](https://commerceclarity.com/solutions/catalog-foundation): bring raw product information into complete, validated attributes, then run the processes that keep them useful across channels. Our article [Meet the Retail Engineer](https://commerceclarity.com/blog/meet-the-retail-engineer) explains who owns that work from the initial business question through deployment and ongoing evaluation. It gives enterprise agentic commerce something dependable to run on. ![A shopping agent needs consistent product data and live fulfillment information to answer a concrete purchase request.](https://cdn.sanity.io/images/xd4hrbt2/production/03fd35725c99383033cd818fa6afa9b1d445b2d9-3200x2600.png?auto=format) *Illustrative. Each condition in the request has an owner, and an operating loop keeps the answer right.* ## Start before it becomes a material share of sales I expect agentic commerce to become a substantial channel. I would not attach a guaranteed percentage and date to that expectation. [McKinsey estimates](https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-agentic-commerce-opportunity-how-ai-agents-are-ushering-in-a-new-era-for-consumers-and-merchants) that agents could orchestrate $3 trillion to $5 trillion in global commerce for goods by 2030. That is a scenario dependent on adoption and merchant readiness. It is not a measurement of current sales or a forecast that every retailer will transact 20% of revenue through agents. The preparation case does not depend on hitting that exact number. If this becomes material in two, three or four years, how long will it take your organization to get the data and operations ready? Think of a retailer in 1994 deciding when to take ecommerce seriously. Publishing a website was only one part of learning to sell through a new channel. The comparison is about preparation time, not a claim that the adoption curves will match. An enterprise can run a useful pilot now while spending longer fixing the systems and ownership that wider adoption will require. Better catalog data also improves the channels you already operate. Run the Lighthouse test. Then take one important category and ask whether the information behind it can answer the questions your customers actually ask. If you want help connecting those two assessments, [book a demo](https://commerceclarity.com/book-a-demo). Bring the report and a product category you know well. We can look at where the data or the process would stop an agent from making a reliable choice. ![Glass office towers seen from the street, looking straight up, their window grids converging on a strip of grey sky.](https://cdn.sanity.io/images/xd4hrbt2/production/aea4b0c8a2a9320d15c97191a41a6f9edb5fc5a9-1600x1000.jpg?auto=format) ## Author [Daniele Vella](/author/daniele-vella) · Co-founder, CommerceClarity ## Questions ### What is agentic commerce? Agentic commerce is shopping in which an AI agent takes on tasks for a customer, such as finding products, comparing options or carrying out an authorized purchase. Different systems support different levels of autonomy; an assistant recommending products is not necessarily completing transactions. ### What is enterprise agentic commerce? Preparing a large retail business to support those interactions across its catalog, markets, channels and operating systems. Alongside product discovery, it requires reliable data, integrations, permissions, quality checks and clear responsibility for failures. It is an operating challenge, not the name of a separate technical standard. ### What does “the catalog is the new storefront” mean? An agent can compare products by querying catalog data through an API or feed, instead of following the same visual journey as a person. It might ask which model fits a space, whether a variant exists or what material a product contains. Your attributes determine how well the product can answer those questions. Agents can also browse pages, and the visual storefront still matters for people. The phrase describes the growing role of product data in discovery and selection. ### Is WebMCP an official Google standard? No. Google supports its development and testing, but WebMCP remains a W3C Community Group draft, not a W3C Standard. Chrome’s experimental implementation and Lighthouse checks do not establish universal browser or agent support. ### What is an llms.txt file? A short Markdown guide to a website, with context and links to selected content. It helps a compatible agent find information without first exploring the whole site. It remains a proposal, and its usefulness depends on whether the agent retrieves and uses it. ### Does llms.txt help SEO? No. Google says llms.txt does not affect visibility or ranking in Google Search. Its inclusion in Lighthouse’s experimental agentic browsing checks serves a different purpose and does not make it a ranking factor. ### What is the Universal Commerce Protocol? The Universal Commerce Protocol, or UCP, is an open protocol for commerce interactions such as checkout. A merchant can publish a profile at /.well-known/ucp to describe supported capabilities and endpoints. That profile does not replace or validate the product catalog. ### Should I create Markdown versions of every product page? I would not make that the default. Start with key brand and editorial pages where the publishing cost is small. Add product Markdown when there is a consumer and a maintenance model that justify it. Product pages, feeds, structured data and APIs should remain aligned to the same underlying information. ### Does passing Lighthouse mean my store is ready for AI agents? It means the tested page passed the applicable checks in that environment. It does not validate product claims, stock accuracy, fulfillment or an entire purchase journey. Use the report to find technical issues, then test real tasks separately. ## Entities - [WebMCP](https://webmachinelearning.github.io/webmcp/) · Thing, about - [llms.txt](https://llmstxt.org/) · Thing, about - [Lighthouse Agentic Browsing](https://developer.chrome.com/docs/lighthouse/agentic-browsing/scoring) · Thing, about - [Agentic Commerce](/solutions/agentic-commerce) · about - [Google](https://www.google.com/) · Organization, mentions - [Google Chrome](https://www.google.com/chrome/) · Thing, mentions - [Lighthouse](https://developer.chrome.com/docs/lighthouse) · Thing, mentions - [World Wide Web Consortium](https://www.w3.org/) · Organization, mentions - [Microsoft](https://www.microsoft.com/) · Organization, mentions - [Model Context Protocol](https://modelcontextprotocol.io/) · Thing, mentions - [Universal Commerce Protocol](https://ucp.dev/) · Thing, mentions - [McKinsey & Company](https://www.mckinsey.com/) · Organization, mentions - [PageSpeed Insights](https://pagespeed.web.dev/) · Thing, mentions - [Catalog Foundation](/solutions/catalog-foundation) · mentions - [SEO & GEO/AEO](/solutions/seo-geo) · mentions - [Search & Discovery](/solutions/search-discovery) · mentions - [Channels & Markets](/solutions/channels-markets) · mentions - [Compliance & Quality](/solutions/compliance-quality) · mentions - Claude Commerce Agents: the agent got easy to build. The catalog decides what it can enforce. · mentions - Meet the Retail Engineer · mentions ## Sources - [Lighthouse Agentic Browsing: scoring](https://developer.chrome.com/docs/lighthouse/agentic-browsing/scoring) · Chrome for Developers - [Web Model Context Protocol (WebMCP), Draft Community Group Report](https://webmachinelearning.github.io/webmcp/) · W3C Web Machine Learning Community Group - [WebMCP origin trial](https://developer.chrome.com/blog/ai-webmcp-origin-trial) · Chrome for Developers - [AI features and your website](https://developers.google.com/search/docs/fundamentals/ai-optimization-guide) · Google Search Central - [Under the hood: Universal Commerce Protocol (UCP)](https://developers.googleblog.com/under-the-hood-universal-commerce-protocol-ucp/) · Google for Developers - [The agentic commerce opportunity: how AI agents are ushering in a new era for consumers and merchants](https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-agentic-commerce-opportunity-how-ai-agents-are-ushering-in-a-new-era-for-consumers-and-merchants) · McKinsey & Company --- # Meet the Retail Engineer > A Retail Engineer is the human owner of an AI agent deployment in retail operations. The role maps the retail process, agrees a business KPI and an evaluation standard with the customer, designs and builds the retail agents, and stays accountable for the result after go-live, running the DIPE loop of Design, Implement, Prove and Evolve. At CommerceClarity the Forward Deployed Engineer is a separate human seat on the same engagement, owning the integrations, transformations and pipelines at the edge of the customer’s stack. One person owns an AI agent deployment in retail, from the first business question to the live catalog. This is what a Retail Engineer does, and where the forward deployed engineer fits. **Published at:** 2026-09-01T09:00:00Z Enterprise software has been sold as a triangle for most of the time any of us have been working in it. The customer sat at the top. Service ran down one side: consultants and integrators paid to make the system fit. Software ran down the other: the vendor, the licence, the roadmap. The gap between the two sides had a name everyone in retail has heard. Not my scope. Service could point at the product. Software could point at the implementation. The customer was left holding the deadline and the operating result. ![The enterprise software triangle, with the customer at the top carrying the gap between service and software](https://cdn.sanity.io/images/xd4hrbt2/production/bd90c83632b56cf962cff520f4baa00c5eacb2f1-3200x2600.png?auto=format) We built our delivery model to remove that gap. The person who holds it together is called a Retail Engineer, and this is what the job actually is. ## What is a Retail Engineer? A Retail Engineer is a customer-facing engineering role for AI delivery in retail operations. One person maps the process with the retailer, defines the business result, [designs the retail agents](/platform/agents), sets the [evaluation standard](/platform/governance), builds the agent workflow and stays on the account after go-live. Same person from the first call to the live customer. That is the rule we run on, and it is the whole point of the title. Business translator, agent engineer, coder, operations builder, and sole owner of the result. We start from the catalog because that is where retail processes meet. Supplier data, attributes, categorisation, content and channel requirements all pass through the product record. Our [catalog foundation](/solutions/catalog-foundation) is built on the same fact: [every downstream workflow reads what the catalog says first](https://commerceclarity.com/blog/llms-txt-webmcp-ecommerce). ![The Retail Engineer, where retail business operations and technology meet on the CommerceClarity platform](https://cdn.sanity.io/images/xd4hrbt2/production/bb3c459a3225172dede125c3f4dd19d6687245ce-3200x2600.png?auto=format) ## Why the role exists now The work used to be split across several people: a consultant who understood the process, a solution engineer who configured the product, a development team that wrote the code. Every seam between them was a place for the result to fall through. AI is changing the cost of those seams. An agent can be adapted to a new category, policy or channel in days. One person can now change a prompt, a rule or a service and see the score move the same afternoon, without opening a project that runs a quarter. The amount of a customer outcome a single person can carry went up, and the roles are catching up to it. The titles have not caught up. Forward deployed engineer, GTM engineer, marketing engineer, solutions architect: every company putting AI inside somebody else’s operation has invented a name for the person who does it, and none of them mean quite the same thing. Marketing got there first, hiring people who build the systems the commercial motion runs on instead of writing a brief and waiting. It is one shape, and it appears wherever the operation and the tooling used to sit with different people. Retail Engineer is that shape pointed at retail operations, catalog first. ## And the Forward Deployed Engineer? A Forward Deployed Engineer is a customer-facing software engineer who turns a platform into a production system inside the customer’s environment. OpenAI describes the role as the intersection of customer delivery and core platform development, owned from technical scoping through [production rollout](https://openai.com/careers/forward-deployed-engineer-%28fde%29-sf-san-francisco/). Companies also use forward deployed software engineer and forward deployed AI engineer for the same work. We have that seat, and it is a human one, hired separately from the Retail Engineer and sitting on the same engagement. The Retail Engineer owns what correct means for this retailer. The Forward Deployed Engineer makes sure the data can get there. On a large account you need both, and neither one is a version of the other. ## How the work runs: DIPE Every engagement runs the same four phases, and we call it DIPE. **Design** ends with two numbers agreed with the customer, and with the boring question answered: how the data gets in, and how the result gets back out. **Implement** builds the agent on the customer’s real data, not on a sample. **Prove** runs it at production scale until the numbers hold on the full catalog, in front of the customer. **Evolve** does not end, because a catalog does not stand still. ![The DIPE loop: Design, Implement, Prove and Evolve, with the evaluation running around the cycle](https://cdn.sanity.io/images/xd4hrbt2/production/e104cff0f76b087dc6ef7139be5a4fcf2ca9a2e7-3200x2600.png?auto=format) ### The two numbers One part of Design is worth stopping on, because everything after it is judged against it. The KPI says whether the business got enough: products published, attributes completed, time to market, cost per product, whatever the live operation is actually measured on. The eval says whether the work is right. It scores the agent against an evaluation set of examples and rules built with the customer’s own experts. OpenAI describes contextual evals the same way: a business expectation turned into a [specific, measurable standard](https://openai.com/index/evals-drive-next-chapter-of-ai/). Neither number substitutes for the other. A workflow can be fast and wrong. It can also be accurate on a small sample and irrelevant to the operating target. ![The two numbers signed in Design: the business KPI beside the evaluation score](https://cdn.sanity.io/images/xd4hrbt2/production/de2a5b5c40346be29d339f23b666685f9ff12a29-3200x2200.png?auto=format) Nothing ships as done because the agent runs. It ships when both numbers reach target on the full catalog and the customer has looked at them, failures included. ## The three hard parts Everything difficult about those four phases collects into three problems. ### Agree what right means A language model is not deterministic and a retail catalog is a very wide input space. You cannot judge the output by reading a few good examples. The evaluation set has to come from the retailer’s own rules and edge cases. What is correct for a pharmacy is wrong for a fashion retailer. What passes on one marketplace fails on another. An eval that works for one customer is not authorised for the next one: each customer owns its own set. This is why the eval belongs to the customer and gets built before the agent rather than after it, on the same principle as our [compliance and quality agents](/solutions/compliance-quality): a rule is only useful once it can be checked across the whole catalog. Once the evaluation set exists it is the only thing that says whether a change helped. Every prompt or rule edit gets the eval run again and the score read. Never tune by feel. ### Get the data in and the result back out This is the unglamorous half, and it decides whether anything reaches production. Product data arrives as supplier feeds, ERP exports, spreadsheets and half-filled PIM records. The result has to land in the system the retailer already runs, in the format that system will accept. A correct product record that cannot enter the customer’s environment has no operating value. This is the half where the Forward Deployed Engineer work sits, and on enterprise accounts it is most of the calendar. Which is why it is mapped before anything is signed: import, attribute configuration and export, all three understood in Design. If the data cannot support the agent and cannot be fixed, we say no up front instead of promising around it. A broken import is a no-go, not a risk to manage later. ### Keep it right in month six Catalogs do not stand still. New suppliers arrive, categories move, marketplace rules change, and an agent left alone gets worse on its own. A score proved once erodes if nobody watches it. So the eval keeps running after go-live, and the customer gets both numbers in the same monthly report, the business KPI beside the correctness score. Reading one without the other is how a quiet regression survives a quarter. Nothing goes onto a live tenant without a dry run against a snapshot and a look at the diff. A new objective is not a tweak, it is a new agent, and it starts again at Design. And when the score moves, somebody named owns the response. That is the part a title cannot fake. ## What the work feels like A morning can start with a merchandising team explaining why a launch takes six weeks. By the afternoon the same person is inside taxonomy rules, prompts and the exceptions that actually block the flow. No requirements document, no handover. They change the workflow, test it, and bring the result back to the people who run it. That proximity is the job. Retail operations holds knowledge that never makes it into a clean specification: an approval that happens only for one category, a naming rule that lives in one person’s head, a marketplace exception buried in an old email. The role turns that into something an agent can run and an eval can check. ![Retail Engineers working together over a laptop](https://cdn.sanity.io/images/xd4hrbt2/production/c57975409a419ade2b36bb09e9963bc489d37819-2048x1152.png?auto=format) ## Who becomes a Retail Engineer? The strongest profiles come from solution consulting, solution delivery, ecommerce operations or retail technology. They know how to read a process, hold a customer room, and find the real problem under the request. They also have a technical side, and they want to use it. They would rather build the workflow than write a ticket for it and wait. AI coding assistants make that shift cheap in a way it was not three years ago, but they do not replace the judgment. The hard part is still understanding the operation well enough to know what to build and what counts as correct. Inside a lot of retailers, somebody is already doing part of this job without the title, holding the business rules and the systems together in their head. Retail Engineer gives that work a method. ## Where this goes The title only matters if the accountability sits with one person. Ours does, and that person can show the two numbers that say whether the work is finished. If this is the work you already do, the Retail Engineer roles are open in [Milan](/careers/retail-engineer-italy) and [London](/careers/retail-engineer-uk), and the Forward Deployed Engineer roles in [Milan](/careers/forward-deployed-engineer-italy) and [London](/careers/forward-deployed-engineer-uk). Everything else is on our [careers page](/careers). ![Somebody climbing a flight of concrete steps marked with yellow nosings, caught mid-stride.](https://cdn.sanity.io/images/xd4hrbt2/production/b88cff1f0c12d8e049bdd680bacea87c086e3284-1600x1000.jpg?auto=format) ## Author [Daniele Vella](/author/daniele-vella) · Co-founder, CommerceClarity ## Questions ### What is a forward deployed engineer? A forward deployed engineer is a customer-facing software engineer who turns a platform into a stable production system inside the customer’s own environment. The role usually owns integrations, custom services, data pipelines and technical reliability, working alongside the customer’s teams rather than from behind a roadmap. ### What is the difference between a Retail Engineer and a Forward Deployed Engineer? They are two seats, hired separately, working the same engagement. The Retail Engineer owns the account and the method: the business goal, the KPI and eval, the agent design, the result after go-live. The Forward Deployed Engineer owns the code at the edge: integrations, transformations and pipelines inside the customer’s stack. At CommerceClarity both are human roles and both are open. ## Entities - Retail Engineer · Thing, about - [Forward Deployed Engineer](https://en.wikipedia.org/wiki/Forward_deployed_engineer) · Thing, about - DIPE · Thing, mentions - [OpenAI](https://openai.com/) · Organization, mentions - [Product information management](https://en.wikipedia.org/wiki/Product_information_management) · Thing, mentions ## Sources - [Forward Deployed Engineer (FDE)](https://openai.com/careers/forward-deployed-engineer-%28fde%29-sf-san-francisco/) · OpenAI - [How evals drive the next chapter in AI for businesses](https://openai.com/index/evals-drive-next-chapter-of-ai/) · OpenAI - [Dev versus Delta: demystifying engineering roles at Palantir](https://blog.palantir.com/dev-versus-delta-demystifying-engineering-roles-at-palantir-ad44c2a6e87) · Palantir - [A Day in the Life of a Palantir Forward Deployed Software Engineer](https://blog.palantir.com/a-day-in-the-life-of-a-palantir-forward-deployed-software-engineer-45ef2de257b1) · Palantir - [Product information management](https://en.wikipedia.org/wiki/Product_information_management) · Wikipedia --- # Meet CommerceClarity at the Ecommerce Tech Summit 2026 > CommerceClarity is the main partner of the Ecommerce Tech Summit 2026 on 20 October 2026 at the Centro Svizzero in Milan, the Casaleggio Associati day on the technology under Italian ecommerce. Co-founder and CEO Federico Sargenti is on stage and the team is in the room for the whole day. The summit presents the new edition of the Casaleggio Associati research on the partners of Italian ecommerce companies. The Ecommerce Tech Summit is Casaleggio Associati’s day on the technology under Italian ecommerce. We are the main partner of the 2026 edition: Federico Sargenti on stage, and the team in the room for the whole day. **Kind:** Conference **Format:** in-person **State:** upcoming **Date:** 20 October 2026 **Starts at:** 2026-10-20T07:30:00Z **Ends at:** 2026-10-20T11:00:00Z **Location:** Milan, Italy **Meta:** Main Partner **Register url:** [/book-a-demo](/book-a-demo) **Event url:** [https://www.ecommerceitalia.info/report/ecommerce-tech-summit-2026/](https://www.ecommerceitalia.info/report/ecommerce-tech-summit-2026/) ## Learn - What an agent actually is, once you take the word back to the work it does. - Which catalog jobs are ready to hand over today, and which are not. - What changes inside a catalog team when the writing is no longer manual. **Audience:** Heads of ecommerce, digital and catalog leaders at Italian retailers and brands. ## Agenda - On stage · Our session, Federico Sargenti. Time to be announced - All day · The team is in the room, live agent runs on request - By appointment · 1:1 sessions with the team ## Facts - Event · Ecommerce Tech Summit 2026 - Date · 20 October 2026 - Format · In person, with a live stream - City · Milan, Italy - We are there as · Main Partner - On stage · Federico Sargenti - Hours · 09:30 to 13:00 - Venue · Centro Svizzero, Via Palestro 2 - Organizer · Casaleggio Associati ## People - [Daniele Vella](/author/daniele-vella) · Co-founder, CommerceClarity - [Federico Sargenti](/author/federico-sargenti) · Co-founder and CEO, CommerceClarity **Venue** ### Centro Svizzero Via Palestro 2 Milan 20121 IT Metro M1, the red line, to Palestro. **Geo** - **Lat:** 45.47267 - **Lng:** 9.19637 ## Questions ### What is the Ecommerce Tech Summit? A Casaleggio Associati event on the technologies, services and partners behind Italian ecommerce, held in Milan. It presents the new edition of their research on the partners of Italian ecommerce companies. ### When and where is it? 20 October 2026, at the Centro Svizzero in Milan. ### What is CommerceClarity’s role? Main partner of the 2026 edition. Co-founder and CEO Federico Sargenti is on stage, and the team is in the room for the full day. ### Who should come and find us? Heads of ecommerce, digital and catalog at Italian retailers and brands. If product data is the thing holding up a channel, a market or a compliance deadline, that is the conversation to have. ## Entities - [Casaleggio Associati](https://www.casaleggio.it) · Organization, mentions ## Sources - [Ecommerce Tech Summit 2026](https://www.ecommerceitalia.info/report/ecommerce-tech-summit-2026/) · Casaleggio Associati - [Casaleggio Associati](https://www.casaleggio.it/) · Casaleggio Associati - [CommerceClarity completes €2.7M funding to power the agentic era of e-commerce](https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/) · Tech.eu --- # Meet CommerceClarity at eTail Autumn Interactive 2026 > CommerceClarity is a partner at eTail Autumn Interactive 2026 on 29-30 September 2026 at Ashridge House in Berkhamsted, United Kingdom, an invitation-only event run by Worldwide Business Research for senior UK ecommerce leaders preparing for peak season. The format is two days of pre-booked thirty-minute one-to-one meetings rather than a show floor, and the schedule is set before the doors open, so a slot has to be taken in advance. Five people from CommerceClarity will be there. eTail Autumn Interactive is two days of pre-booked one-to-one meetings rather than a show floor, which is why we go: thirty minutes, one catalog, a real conversation. We are there as a partner, and the schedule is set before the doors open, so the useful thing to do now is put a slot in the diary. **Kind:** Summit **Format:** in-person **State:** upcoming **Date:** 29-30 September 2026 **Starts at:** 2026-09-29T09:00:00Z **Ends at:** 2026-09-30T18:00:00Z **Location:** Berkhamsted, United Kingdom **Meta:** Pre-booked 1:1 meetings **Register url:** [/book-a-demo](/book-a-demo) **Event url:** [https://etailconnectautumn.wbresearch.com/](https://etailconnectautumn.wbresearch.com/) ## Learn - What thirty minutes is enough to show: one agent, on a real catalog, start to finish. - Which catalog jobs are worth handing over first, and which are not. - Where the review step belongs once the writing stops being manual. **Audience:** Heads of ecommerce, digital and catalog leaders at UK retailers running large multi-brand assortments. ## Agenda - Both days · 1:1 meetings, thirty minutes each - On site · Live agent runs on request - Evenings · Everyone stays on site, so the conversations carry on ## Facts - Event · eTail Autumn Interactive 2026 - Dates · 29-30 September 2026 - Format · In person, by appointment - Venue · Ashridge House, Berkhamsted - We are there as · Partner - Organizer · WBR ## People - [Alessandro Angelini](/author/alessandro-angelini) · Co-founder and CRO, CommerceClarity - [Federico Sargenti](/author/federico-sargenti) · Co-founder and CEO, CommerceClarity - [Nathan Salter](/author/nathan-salter) · Sales Director, CommerceClarity - [Lorenzo Riccò](/author/lorenzo-ricco) · Go to Market Engineer, CommerceClarity - [Daniele Vella](/author/daniele-vella) · Co-founder, CommerceClarity **Venue** ### Ashridge House Ashridge Berkhamsted HP4 1NS GB A residential venue outside London. Attendees stay on site across both days. **Geo** - **Lat:** 51.79934 - **Lng:** -0.56104 ## Questions ### What is eTail Autumn Interactive? An invitation-only UK event run by Worldwide Business Research for senior ecommerce, digital and omnichannel leaders getting ready for peak season. It runs on pre-booked one-to-one meetings rather than a show floor. ### When and where is it? 29-30 September 2026, at Ashridge House in Berkhamsted, United Kingdom. ### How do we get a meeting with CommerceClarity there? Book it before the event. The schedule is fixed before the doors open, so the useful thing to do now is take a slot rather than look for us on the day. ### What should we bring to the thirty minutes? One category and one messy feed. That is enough to run an agent on your own data rather than on a demo dataset, which is the only version of the conversation worth having. ## Entities - [Worldwide Business Research](https://www.wbresearch.com) · Organization, mentions ## Sources - [eTail Autumn Interactive](https://etailconnectautumn.wbresearch.com/) · Worldwide Business Research - [eTail UK](https://etailuk.wbresearch.com/) · Worldwide Business Research - [CommerceClarity completes €2.7M funding to power the agentic era of e-commerce](https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/) · Tech.eu --- # We were at eTail Boston 2026 > CommerceClarity attended eTail Boston 2026 in August 2026 at the Sheraton Boston Hotel, the North American edition of the eTail conference series run by Worldwide Business Research. It was the company’s first eTail in the United States, with three people there, including its UK and go-to-market team. The finding was that the catalog problem does not change across the Atlantic: the conversation was the same one the company has in Milan and in Manchester. Our first eTail in the United States. A different market, the same catalog problem, and the conversation that came back was the same one we have in Milan and in Manchester. **Kind:** Summit **Format:** in-person **State:** past **Date:** August 2026 **Starts at:** 2026-08-10T09:00:00Z **Location:** Boston, United States **Event url:** [https://etaileast.wbresearch.com/](https://etaileast.wbresearch.com/) ## Learn - Whether the catalog problem changes across the Atlantic, and where it does not. - Which jobs a US retail team hands over first. - What an agent run looks like on a real feed, in front of the people who own it. **Audience:** Heads of ecommerce, digital and catalog leaders at North American retailers. ## Agenda - Across the event · Scheduled meetings and sessions - Throughout · Live agent runs on request ## Facts - When · August 2026 - Format · In person - Venue · Sheraton Boston Hotel - Organizer · WBR - Region · Our first eTail in North America ## People - [Alessandro Angelini](/author/alessandro-angelini) · Co-founder and CRO, CommerceClarity - [Nathan Salter](/author/nathan-salter) · Sales Director, CommerceClarity - [Lorenzo Riccò](/author/lorenzo-ricco) · Go to Market Engineer, CommerceClarity **Venue** ### Sheraton Boston Hotel 39 Dalton Street Boston 02199 US **Geo** - **Lat:** 42.34651 - **Lng:** -71.08387 ## Questions ### What is eTail Boston? The North American edition of the eTail conference series, also known as eTail East, run every August in Boston by Worldwide Business Research for senior ecommerce and omnichannel leaders. ### When and where was eTail Boston 2026? August 2026, at the Sheraton Boston Hotel, 39 Dalton Street, Boston. ### Why go to the United States at all? The US reaches CommerceClarity through inbound and partners rather than outbound, so the point was to test whether the catalog argument lands the same way in a market the company does not prospect. It did. ### Does the catalog problem look different in North America? The channels and the regulation differ. The problem underneath does not: product data that no team can keep complete by hand, and marketplaces that each want it their own way. ## Entities - [Worldwide Business Research](https://www.wbresearch.com) · Organization, mentions ## Sources - [eTail Boston](https://etaileast.wbresearch.com/) · Worldwide Business Research - [eTail UK](https://etailuk.wbresearch.com/) · Worldwide Business Research - [CommerceClarity completes €2.7M funding to power the agentic era of e-commerce](https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/) · Tech.eu --- # We were at eTail Connect Spring 2026 > CommerceClarity attended eTail Connect Spring 2026 on 3-4 March 2026 at Allianz Stadium Twickenham in London, an invitation-only event run by Worldwide Business Research for senior UK ecommerce leaders. It was the company’s first UK room for the catalog argument, and three people were there. What it taught us about which formats suit a conversation about product data is how we choose events now. Our first eTail of the year, at Twickenham, and the first time we took the catalog argument to a UK room. What we learned about which format works for a conversation like ours is what shaped how we pick these events now. **Kind:** Summit **Format:** in-person **State:** past **Date:** 3-4 March 2026 **Starts at:** 2026-03-03T09:00:00Z **Ends at:** 2026-03-04T18:00:00Z **Location:** London, United Kingdom **Event url:** [https://etailconnecteu.wbresearch.com/](https://etailconnecteu.wbresearch.com/) ## Learn - What the catalog bottleneck looks like on a UK multi-brand assortment. - Which jobs a retail team hands over first, and which it keeps. - What an agent run shows in half an hour, on a real feed. **Audience:** Heads of ecommerce, digital and catalog leaders at UK retailers. ## Agenda - Day 1 · Sessions and scheduled conversations - Day 2 · Meetings with retailers, live agent runs on request ## Facts - Dates · 3-4 March 2026 - Format · In person - Venue · Allianz Stadium Twickenham - Organizer · WBR - What came next · eTail UK 2026 in Manchester, three months later ## People - [Alessandro Angelini](/author/alessandro-angelini) · Co-founder and CRO, CommerceClarity - [Nathan Salter](/author/nathan-salter) · Sales Director, CommerceClarity - [Lorenzo Riccò](/author/lorenzo-ricco) · Go to Market Engineer, CommerceClarity **Venue** ### Allianz Stadium Twickenham 200 Whitton Road Twickenham TW2 7BA GB **Geo** - **Lat:** 51.45601 - **Lng:** -0.3415 ## Questions ### What is eTail Connect Spring? An invitation-only event run by Worldwide Business Research for senior ecommerce and digital leaders at UK retailers. It runs on pre-booked one-to-one conversations rather than a show floor. ### When and where was the 2026 edition? 3-4 March 2026, at Allianz Stadium Twickenham in London. ### Why did CommerceClarity go? The United Kingdom is the first market the company opened outside Italy, and this was the first UK room for the catalog argument. Most of what we do in the UK today traces back to conversations in rooms like it. ### Is CommerceClarity going again? The eTail series stays on the calendar: eTail UK in Manchester and eTail Autumn Interactive at Ashridge House both followed. The events page carries whatever is next. ## Entities - [Worldwide Business Research](https://www.wbresearch.com) · Organization, mentions ## Sources - [eTail Connect Spring](https://etailconnecteu.wbresearch.com/) · Worldwide Business Research - [eTail UK](https://etailuk.wbresearch.com/) · Worldwide Business Research - [CommerceClarity completes €2.7M funding to power the agentic era of e-commerce](https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/) · Tech.eu --- # We were at eTail UK 2026 > CommerceClarity attended eTail UK 2026 on 16-17 June 2026 at Emirates Old Trafford in Manchester, the UK edition of the eTail conference series run by Worldwide Business Research. Three people were there. The audience is heads of ecommerce, digital and catalog at UK retailers running large multi-brand assortments, and the format is scheduled conversations rather than a show floor, which is why the company keeps returning to it. Two days in Manchester with UK retail, on the format we like: scheduled conversations rather than a stand to walk past. Most of what we do in the UK today traces back to rooms like this one. **Kind:** Summit **Format:** in-person **State:** past **Date:** 16-17 June 2026 **Starts at:** 2026-06-16T09:00:00Z **Ends at:** 2026-06-17T18:00:00Z **Location:** Manchester, United Kingdom **Event url:** [https://etailuk.wbresearch.com/](https://etailuk.wbresearch.com/) ## Learn - What breaks in a catalog once it crosses several brands and a second market. - Where the bottleneck actually sits, which is rarely the PIM. - What an agent run looks like on a real feed, in front of the people who own it. **Audience:** Heads of ecommerce, digital and catalog leaders at UK retailers running large multi-brand assortments. ## Agenda - Day 1 · Scheduled meetings and sessions - Day 2 · Meetings with retailers, live agent runs on request ## Facts - Dates · 16-17 June 2026 - Format · In person - Venue · Emirates Old Trafford, Manchester - Organizer · WBR - What came next · eTail Autumn Interactive 2026, 29-30 September ## People - [Federico Sargenti](/author/federico-sargenti) · Co-founder and CEO, CommerceClarity - [Nathan Salter](/author/nathan-salter) · Sales Director, CommerceClarity - [Lorenzo Riccò](/author/lorenzo-ricco) · Go to Market Engineer, CommerceClarity **Venue** ### Emirates Old Trafford Talbot Road, Stretford Manchester M16 0PX GB **Geo** - **Lat:** 53.45641 - **Lng:** -2.28686 ## Questions ### What is eTail UK? The UK edition of the eTail conference series, run by Worldwide Business Research for senior ecommerce, digital and omnichannel leaders at British retailers. ### When and where was eTail UK 2026? 16-17 June 2026, at Emirates Old Trafford in Manchester. ### Who was in the room? Heads of ecommerce, digital and catalog at UK retailers running large multi-brand assortments, which is the profile CommerceClarity works with. ### Why this format rather than a trade fair? A catalog conversation needs half an hour and one real feed, not a leaflet. Scheduled one-to-one meetings give that, and a stand people walk past does not. ## Entities - [Worldwide Business Research](https://www.wbresearch.com) · Organization, mentions ## Sources - [eTail UK](https://etailuk.wbresearch.com/) · Worldwide Business Research - [eTail Autumn Interactive](https://etailconnectautumn.wbresearch.com/) · Worldwide Business Research - [CommerceClarity completes €2.7M funding to power the agentic era of e-commerce](https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/) · Tech.eu --- # Meet CommerceClarity at Netcomm Forum 2027 > CommerceClarity exhibits at Netcomm Forum 2027 on 21-22 April 2027 at the Allianz MiCo, Milano Convention Centre in Milan, the annual gathering of the Italian ecommerce industry run by Consorzio Netcomm. The company is at stand B10 with a main-hall workshop in room SKY 2 on day one at 16:00, and six people including the founding team will be there. Demos run on a visitor’s own catalog rather than on a demo dataset. Netcomm Forum is the Italian ecommerce industry in one room, and it is where we met most of the retailers we work with today. In 2027 we come back with a bigger stand and a main-hall workshop. **Kind:** Fair **Format:** in-person **State:** upcoming **Date:** 21-22 April 2027 **Starts at:** 2027-04-21T09:00:00Z **Ends at:** 2027-04-22T18:00:00Z **Location:** Milan, Italy **Meta:** Stand B10 **Register url:** [/book-a-demo](/book-a-demo) **Event url:** [https://www.consorzionetcomm.it/netcomm-forum/](https://www.consorzionetcomm.it/netcomm-forum/) ## Learn - A live agent run on your own catalog, not on a demo dataset. - Where a human belongs in the loop, and where they only slow it down. - What it takes to publish into a new market without adding headcount. **Audience:** Heads of ecommerce and catalog leaders at retailers who own product data across several channels and markets. ## Agenda - All day · Stand B10, live agent demos on request - 16:00 · Workshop in the main hall, room SKY 2 - By appointment · 1:1 sessions with the team ## Facts - Event · Netcomm Forum 2027 - Dates · 21-22 April 2027 - Format · In person - City · Milan, Italy - Find us · Stand B10 - Workshop · Main hall, room SKY 2, 16:00 - Organizer · Consorzio Netcomm ## People - [Daniele Vella](/author/daniele-vella) · Co-founder, CommerceClarity - [Federico Sargenti](/author/federico-sargenti) · Co-founder and CEO, CommerceClarity - [Alessandro Angelini](/author/alessandro-angelini) · Co-founder and CRO, CommerceClarity - [Luca Cozzolino](/author/luca-cozzolino) · Co-founder and CPO, CommerceClarity - [Lorenzo Riccò](/author/lorenzo-ricco) · Go to Market Engineer, CommerceClarity - [Matteo Codutti](/author/matteo-codutti) · Retail Engineer, CommerceClarity **Venue** ### Allianz MiCo, Milano Convention Centre Piazzale Carlo Magno 1 Milan 20149 IT Find us at stand B10. The workshop is in room SKY 2 on day one at 16:00. **Geo** - **Lat:** 45.48048 - **Lng:** 9.15582 ## Questions ### Where do we find CommerceClarity at Netcomm Forum 2027? Stand B10, both days, with live agent runs on request. The workshop is in the main hall, room SKY 2, on day one at 16:00. ### When and where is the 2027 edition? 21-22 April 2027, at the Allianz MiCo, Milano Convention Centre, Piazzale Carlo Magno 1, Milan. ### What does a live agent run at the stand look like? Bring a category and a feed and the agent runs on it in front of you: what it sources, what it fills, what it marks as unsettled and why. On your own catalog, not on a demo dataset. ### What is the workshop about? Where a human belongs in the loop and where they only slow it down, and what it takes to publish into a new market without adding headcount. ## Entities - [Consorzio Netcomm](https://www.consorzionetcomm.it) · Organization, mentions ## Sources - [Netcomm Forum](https://www.consorzionetcomm.it/netcomm-forum/) · Consorzio Netcomm - [Consorzio Netcomm](https://www.consorzionetcomm.it/) · Consorzio Netcomm - [CommerceClarity acquires Katalogo.ai](https://tech.eu/2026/06/04/commerceclarity-acquires-katalogoai/) · Tech.eu --- # We were at Netcomm Forum 2026 > CommerceClarity exhibited at the 21st Netcomm Forum on 6-7 May 2026 at the Allianz MiCo, Milano Convention Centre, the Italian ecommerce industry’s annual gathering run by Consorzio Netcomm. It was the company’s first Netcomm Forum as an exhibitor, with four people on the stand. Most of the retailers CommerceClarity works with today were met in rooms like it, which is why the company returns in 2027 with a bigger stand and a main-hall workshop. The 21st Netcomm Forum, and our first as an exhibitor. Two days at the Allianz MiCo with the Italian ecommerce industry in one building. Most of the retailers we talk to today, we met in a room like that one, which is why we go back in 2027 with a bigger stand and a workshop. **Kind:** Fair **Format:** in-person **State:** past **Date:** 6-7 May 2026 **Starts at:** 2026-05-06T09:00:00Z **Ends at:** 2026-05-07T18:00:00Z **Location:** Milan, Italy **Meta:** 21st edition **Event url:** [https://www.consorzionetcomm.it/netcomm-forum/](https://www.consorzionetcomm.it/netcomm-forum/) ## Learn - Live agent runs on real catalogs, at the stand, on request. - Where the review step belongs once an agent does the writing. - Which catalog jobs are worth handing over first. **Audience:** Heads of ecommerce and catalog leaders at Italian retailers who own product data across several channels and markets. ## Agenda - Day 1 · Stand, live agent runs through the day - Day 2 · Meetings with retailers and partners ## Facts - Event · Netcomm Forum 2026 - Dates · 6-7 May 2026 - Edition · 21st - Format · In person - Venue · Allianz MiCo, Piazzale Carlo Magno 1 - Organizer · Consorzio Netcomm - Next edition · Netcomm Forum 2027, 21-22 April ## People - [Daniele Vella](/author/daniele-vella) · Co-founder, CommerceClarity - [Federico Sargenti](/author/federico-sargenti) · Co-founder and CEO, CommerceClarity - [Alessandro Angelini](/author/alessandro-angelini) · Co-founder and CRO, CommerceClarity - [Luca Cozzolino](/author/luca-cozzolino) · Co-founder and CPO, CommerceClarity **Venue** ### Allianz MiCo, Milano Convention Centre Piazzale Carlo Magno 1 Milan 20149 IT Metro M5 to Portello or Domodossola, fifteen minutes from the city centre. **Geo** - **Lat:** 45.48048 - **Lng:** 9.15582 ## Questions ### What is Netcomm Forum? The annual gathering of the Italian ecommerce industry, run by Consorzio Netcomm in Milan. The 2026 edition was the 21st. ### When and where was Netcomm Forum 2026? 6-7 May 2026, at the Allianz MiCo, Milano Convention Centre, Piazzale Carlo Magno 1, Milan. ### Why does CommerceClarity exhibit there? Because most of the Italian retailers the company works with today were met at events like it. Milan is the operational headquarters, and Netcomm Forum puts the buyers of catalog work in one building for two days. ### Is CommerceClarity at Netcomm Forum 2027? Yes, on 21-22 April 2027 at the Allianz MiCo, with stand B10 and a main-hall workshop in room SKY 2 on day one at 16:00. ## Entities - [Consorzio Netcomm](https://www.consorzionetcomm.it) · Organization, mentions ## Sources - [Netcomm Forum](https://www.consorzionetcomm.it/netcomm-forum/) · Consorzio Netcomm - [Consorzio Netcomm](https://www.consorzionetcomm.it/) · Consorzio Netcomm - [CommerceClarity completes €2.7M funding to power the agentic era of e-commerce](https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/) · Tech.eu --- # We were at Shoptalk Europe 2026 > CommerceClarity attended Shoptalk Europe 2026 on 9-11 June 2026 at Fira Barcelona Gran Via, the European edition of the Shoptalk retail conference. Two people were there. The point of the trip was reach rather than volume: the conversations that came out of it were with retailers outside the markets the company had been working, which is why Barcelona was chosen over a home-market event that quarter. Shoptalk Europe is where the retailers we do not yet know are in one place. Barcelona was our first pass at it, and the conversations that came out of it were with retailers outside the markets we had been working, which is exactly what we went for. **Kind:** Conference **Format:** in-person **State:** past **Date:** 9-11 June 2026 **Starts at:** 2026-06-09T09:00:00Z **Ends at:** 2026-06-11T18:00:00Z **Location:** Barcelona, Spain **Event url:** [https://europe.shoptalk.com/](https://europe.shoptalk.com/) ## Learn - What an agent does to a catalog, shown rather than described. - Which catalog jobs are worth handing over first. - How the same catalog problem reads in a market we had not worked before. **Audience:** Heads of ecommerce, digital and catalog leaders at European retailers and marketplaces. ## Agenda - Three days · On the floor, meetings by appointment - Throughout · Live agent runs on request ## Facts - Event · Shoptalk Europe 2026 - Dates · 9-11 June 2026 - Format · In person - Venue · Fira Barcelona Gran Via ## People - [Alessandro Angelini](/author/alessandro-angelini) · Co-founder and CRO, CommerceClarity - [Luca Cozzolino](/author/luca-cozzolino) · Co-founder and CPO, CommerceClarity **Venue** ### Fira Barcelona Gran Via Avinguda Joan Carles I, 64 L’Hospitalet de Llobregat 08908 ES **Geo** - **Lat:** 41.35069 - **Lng:** 2.13156 ## Questions ### What is Shoptalk Europe? The European edition of the Shoptalk retail conference, held in Barcelona. It brings retailers, brands and marketplaces from across Europe into one event. ### When and where was Shoptalk Europe 2026? 9-11 June 2026, at Fira Barcelona Gran Via. ### Why Barcelona? Two reasons. Shoptalk Europe puts retailers the company has not met in one place, and CommerceClarity has an office in Barcelona alongside the Milan headquarters. ### What came out of it? Conversations with retailers outside the markets the company had been working, which was the reason for going. Proving the company can convert internationally is a stated goal for the year. ## Entities - [Shoptalk](https://shoptalk.com) · Organization, mentions ## Sources - [Shoptalk Europe](https://europe.shoptalk.com/) · Shoptalk - [Shoptalk](https://shoptalk.com/) · Shoptalk - [CommerceClarity acquires Katalogo.ai](https://tech.eu/2026/06/04/commerceclarity-acquires-katalogoai/) · Tech.eu --- # We have acquired Katalogo.ai and welcome Luca Cozzolino as Founding Chief Product Officer *For immediate release* > CommerceClarity has acquired Katalogo.ai, the Barcelona startup building styling agents on retail catalogs, and its founder Luca Cozzolino joins the founding team as Chief Product Officer. It is the company’s first acquisition, eighteen months after founding and six months after the €2.7 million round led by Italian Founders Fund and Entourage. Headquarters stay in Milan, and the United Kingdom is the first market CommerceClarity opens outside Italy. Eighteen months after its founding and six months after the €2.7M round, the AI agent platform for catalog operations completes its first M&A deal and expands the founding team. **Dateline:** Milan, 4 June 2026 **Published at:** 2026-06-04 **Category:** Company news ![Luca Cozzolino, Founding Chief Product Officer at CommerceClarity.](https://cdn.sanity.io/images/xd4hrbt2/production/0b4b75c633090effff4c2805f689a34bc4e01ed6-1600x1066.jpg?auto=format) **Hero caption:** Luca Cozzolino joins the founding team as Chief Product Officer. CommerceClarity, the AI agent platform for catalog operations used by enterprise retailers and brands including Prada, STIGA, Cisalfa Sport, Arcaplanet, Next Hardware & Software and Top Farmacia (Hippocrates Holding), today announces the acquisition of Katalogo.ai, a Barcelona-based startup founded by Luca Cozzolino. As part of the deal, Cozzolino joins CommerceClarity’s founding team as Chief Product Officer. The acquisition is the company’s first M&A transaction, closed eighteen months after its founding and six months after the €2.7 million round led by Italian Founders Fund (Koinos Capital SGR) and Entourage, with participation from Euregio+ (Alpine VC), Redstone, Vento, Vesper Holding, Ithaca Investments, and business angels from the tech and AI sector. ## Why now: the catalog is the leverage point of Agentic Commerce For a decade, the product catalog has been an operational problem: feeding the proprietary website and the major marketplaces with sufficiently structured data. The arrival of AI search and shopping agents is rewriting the criteria for visibility: a product exists only if its data is readable by a machine that decides what to show the consumer. In parallel, the multiplication of channels, markets and compliance requirements has pushed catalog management beyond the threshold of manual sustainability. What emerges is a product category of its own: not a PIM, not a feed tool, not a generic AI integration. It’s an intelligence layer that ingests catalog data, structures it, enriches it and validates it against the company’s rules. Whoever builds this layer over the next eighteen months secures a lead that will be hard to close, because they’re working from better data. CommerceClarity customers reduce catalog management operational costs by up to 90%, bring new products online in hours instead of weeks, and record an average +30% in traffic and sales. Not as the effect of a single deployment, but as a continuous capability to keep the catalog aligned with the business. ## The acquisition: thesis-fit before tech-fit Katalogo.ai was building styling agents on top of retail catalogs — systems capable of analyzing available products and generating look proposals designed to sell together. The product is concrete proof that the same thesis behind CommerceClarity — the catalog as a competitive asset, not as an archive — also holds in segments with high curation complexity such as fashion and lifestyle. ## Luca Cozzolino: ten years at the two companies that defined two eras of ecommerce Cozzolino brings to the founding team a track record spanning both sides of the market: a former merchant before becoming a product leader, he led product at Shopify (the platform that defined SMB ecommerce) and at Zalando (the retailer that defined fashion ecommerce in Europe), with a stop at Perk along the way. ## What changes: European ambition, execution unchanged The acquisition coincides with the launch of CommerceClarity’s international expansion, with the United Kingdom as the first market outside Italy. Cozzolino’s base in Barcelona extends the company’s European footprint, with headquarters remaining in Milan. The Service as a Software operating model — in which every customer is supported by dedicated AI Engineers who bring the agents into production category by category — remains unchanged. ## The facts, in one place - The deal · Acquisition of Katalogo.ai, Barcelona. First M&A transaction. - Announced · 4 June 2026 - Founded · CommerceClarity, eighteen months before this announcement - Last round · €2.7M, November 2025 - Headquarters · Milan. First market outside Italy: the United Kingdom. - Customers · Prada, STIGA, Cisalfa Sport, Arcaplanet, Next Hardware & Software, Top Farmacia (Hippocrates Holding) **Investors label:** Investors in the €2.7M round ## Investors - Italian Founders Fund (Koinos Capital SGR) - Entourage - Euregio+ (Alpine VC) - Redstone - Vento - Vesper Holding - Ithaca Investments **Investors note:** Plus business angels from the tech and AI sector. **Contact:** press@commerceclarity.com ## People - [Federico Sargenti](/author/federico-sargenti) · Co-founder and CEO, CommerceClarity - [Luca Cozzolino](/author/luca-cozzolino) · Co-founder and CPO, CommerceClarity ## Questions ### What did CommerceClarity acquire? Katalogo.ai, a Barcelona-based startup founded by Luca Cozzolino. It was building styling agents on top of retail catalogs: systems that read the products a retailer already has and propose looks designed to sell together. It is CommerceClarity’s first M&A transaction, announced on 4 June 2026. ### Who is Luca Cozzolino? A merchant before he was a product leader. He led product at Shopify, the platform that defined SMB ecommerce, and at Zalando, the retailer that defined fashion ecommerce in Europe, with a stop at Perk along the way, then founded Katalogo.ai in Barcelona. He joins CommerceClarity’s founding team as Chief Product Officer. ### Why is the product catalog the leverage point of agentic commerce? Because AI search and shopping agents now decide what a consumer sees, and they decide it from data. A product exists only if its data is readable by the machine doing the choosing. That turns the catalog into an intelligence layer a company has to build: not a PIM, not a feed tool, not a generic AI integration, but a layer that ingests catalog data, structures it, enriches it and validates it against the company’s own rules. ### What changes for CommerceClarity customers? Nothing in how they are served. The Service as a Software model holds: every customer works with dedicated Retail Engineers who bring the agents into production category by category. Customers reduce catalog management operational costs by up to 95%, bring new products online in hours instead of weeks, and record an average +30% in traffic and sales. ### Where is CommerceClarity based after the acquisition? Headquarters remain in Milan. Cozzolino’s base in Barcelona extends the European footprint, and the United Kingdom is the first market the company opens outside Italy. ### Were the terms of the deal disclosed? No. The financial terms of the acquisition were not disclosed. The release states that it is the company’s first M&A transaction, closed six months after the €2.7 million round led by Italian Founders Fund (Koinos Capital SGR) and Entourage. ## Entities - [Katalogo.ai](https://katalogo.ai) · Organization, about - Luca Cozzolino · Person, about - [Shopify](https://www.shopify.com) · Organization, mentions - [Zalando](https://www.zalando.com) · Organization, mentions - [Italian Founders Fund](https://www.italianfoundersfund.com) · Organization, mentions ## Sources - [CommerceClarity acquires Katalogo.ai](https://tech.eu/2026/06/04/commerceclarity-acquires-katalogoai/) · Tech.eu - [Why we invested in CommerceClarity](https://italianfoundersfund.substack.com/p/why-we-invested-in-commerceclarity) · Italian Founders Fund - [CommerceClarity completes €2.7M funding to power the agentic era of e-commerce](https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/) · Tech.eu - [CommerceClarity chiude un round da 2,7 milioni: operazione guidata dai fondi Iff e Entourage](https://www.milanofinanza.it/news/commerceclarity-chiude-un-round-da-2-7-milioni-operazione-guidata-dai-fondi-iff-e-entourage-202511131501462937) · Milano Finanza - [CommerceClarity raccoglie 2,7 milioni per potenziare l’e-commerce con l’IA](https://forbes.it/2025/11/13/commerceclarity-raccoglie-27-milioni-per-potenziare-le-commerce-con-lia) · Forbes Italia --- # CommerceClarity raises €2.7M ($3.1M) pre-seed to power the AI agentic era of e-commerce *For immediate release* > CommerceClarity raised a €2.7M ($3.1M) oversubscribed pre-seed round, announced on 13 November 2025, led by Italian Founders Fund (Koinos Capital SGR) and Entourage, with participation from Euregio+ (Alpine VC), Redstone, Vento (Exor), Ithaca Investments and Vesper Holding. The company builds AI agents for retail catalog operations: customers reduce catalog costs by up to 95% and record around +30% in traffic and sales. By 2030, over $5 trillion in global retail sales could be mediated by AI agents, according to McKinsey. We’ve raised €2.7M ($3.1M) to build the intelligent infrastructure for the agentic era of commerce. **Dateline:** Milan, 13 November 2025 **Published at:** 2025-11-13 **Category:** Funding ![The four CommerceClarity founders](https://cdn.sanity.io/images/xd4hrbt2/production/1f3f03bae786a5ea11e88f5d8ec50c165cb79d12-2560x1709.jpg?auto=format) **Hero caption:** The founding team: Ale, Dani, Fede and Mike. CommerceClarity has raised a €2.7M ($3.1M) oversubscribed pre-seed round, led by IFF (Italian Founders Fund) and Entourage, with participation from Euregio+ (Alpine VC), Redstone, Vento (Exor), Ithaca Investment, and Vesper Holding. This funding will accelerate our AI-driven e-commerce automation platform and support international expansion, with offices now in Milan, Rome, and London. ## The problem: e-commerce complexity has outgrown human capacity Modern e-commerce catalog management has become too complex to handle manually. Every product must be described, localized, categorized, optimized for search engines and recommendation systems, and adapted to marketplace rules. Most teams still rely on spreadsheets and manual workflows that don’t scale. The result is missing data, inconsistent catalogs, and products that never go live. Each delay means lost visibility and lost sales. Having managed thousands of SKUs and billions in GMV ourselves, we’ve seen it firsthand — catalog operations remain the slowest and least automated process in retail. ## The shift: from human commerce to agentic commerce AI is transforming how products are discovered, evaluated, and purchased. By 2030, over $5 trillion in global retail sales could be mediated by AI agents, according to McKinsey. But these systems depend on structured, consistent product data — and most catalogs today are incomplete or unstructured. In the next generation of commerce, products without structured data risk becoming invisible. ## The vision: structured commerce for a machine-readable world CommerceClarity is building the agent layer for modern commerce — a network of specialized AI agents that automate and optimize catalog creation, pricing, and advertising. Our platform transforms raw product data into structured, high-quality content that algorithms and AI systems can understand. What once took weeks now takes minutes, helping brands reduce operational costs by up to 90% and increase sales by 30%. We started with the most painful problem: the product catalog. Our Catalog Agent is already used by leading brands like Nestlé Purina, Arcaplanet, Cisalfa, and 1000Farmacie. Each new agent builds on shared intelligence, multiplying the value of the network. We’re not replacing existing systems — we’re augmenting them with always-on AI agents that make teams faster, more consistent, and more resilient. ## The opportunity: a $100B+ automation frontier The market for e-commerce automation is massive but fragmented. Traditional PIM systems, feed managers, and SEO tools solve parts of the problem but don’t deliver an integrated approach. CommerceClarity connects these layers into one intelligent system that helps teams scale product data globally with speed and precision. With this round, we’ll continue investing in AI, data, and product engineering, expanding our network of agents across the entire commerce stack. ## The challenge ahead Commerce is entering the agentic era. The winners will be those who can operate at scale with consistency, not just speed. That’s the challenge we’ve set for ourselves — and the opportunity we’re building for. If you want to help shape the next generation of AI commerce, we’re hiring. → Join us in building the future of commerce. Ale, Dani, Fede, Mike ## The facts, in one place - The round · €2.7M Pre-seed - Announced · 13 November 2025 - Led by · Italian Founders Fund (Koinos Capital SGR) and Entourage - Founded · Milan, 2024 **Investors label:** Investors in the round ## Investors - Italian Founders Fund (Koinos Capital SGR) - Entourage - Euregio+ (Alpine VC) - Redstone - Vento - Vesper Holding - Ithaca Investments **Investors note:** Plus business angels from the tech, retail and AI sector. **Contact:** press@commerceclarity.com ## People - [Alessandro Angelini](/author/alessandro-angelini) · Co-founder and CRO, CommerceClarity - [Daniele Vella](/author/daniele-vella) · Co-founder, CommerceClarity - [Federico Sargenti](/author/federico-sargenti) · Co-founder and CEO, CommerceClarity - [Michele Sampieri](/author/michele-sampieri) · Co-founder and Product Engineer, CommerceClarity ## Questions ### How much did CommerceClarity raise? €2.7 million ($3.1 million), in an oversubscribed pre-seed round announced on 13 November 2025. ### Who led the round? Italian Founders Fund (Koinos Capital SGR) and Entourage co-led, with participation from Euregio+ (Alpine VC), Redstone, Vento (Exor), Ithaca Investments, Vesper Holding, and business angels from the tech and AI sector. ### What will the funding be used for? Investment in AI, data and product engineering, expanding the network of agents across the commerce stack, and international expansion beyond Italy. ### What does CommerceClarity do? It runs AI agents for retail catalog operations: ingesting, structuring, enriching and validating product data against each company’s own rules. Customers reduce catalog management costs by up to 95%, bring products online in hours instead of weeks, and record around +30% in traffic and sales. ### Why does the product catalog matter in the agentic era? Because AI agents are starting to mediate what consumers see and buy: by 2030, over $5 trillion in global retail sales could be mediated by AI agents, according to McKinsey. Those systems decide from structured product data, and a product whose data a machine cannot read risks becoming invisible. ## Entities - [CommerceClarity](https://commerceclarity.com) · Organization, about - [Italian Founders Fund](https://www.italianfoundersfund.com) · Organization, mentions - [Entourage](https://entourage.io) · Organization, mentions - [Vento](https://www.vento.ventures) · Organization, mentions - [Redstone](https://redstone.vc) · Organization, mentions - [McKinsey & Company](https://www.mckinsey.com) · Organization, mentions ## Sources - [E-commerce, la rete di agenti di intelligenza artificiale (su cui investono Elkann e Berlusconi)](https://www.corriere.it/economia/finanza/25_novembre_13/e-commerce-la-rete-di-agenti-di-intelligenza-artificiale-su-cui-investono-elkann-e-berlusconi-37686617-81a2-48dc-a88d-c7324ae5bxlk.shtml) · Corriere della Sera, L’Economia - [CommerceClarity completes €2.7M funding to power the agentic era of e-commerce](https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/) · Tech.eu - [CommerceClarity raccoglie 2,7 milioni per potenziare l’e-commerce con l’IA](https://forbes.it/2025/11/13/commerceclarity-raccoglie-27-milioni-per-potenziare-le-commerce-con-lia) · Forbes Italia - [CommerceClarity chiude un round da 2,7 milioni: operazione guidata dai fondi Iff e Entourage](https://www.milanofinanza.it/news/commerceclarity-chiude-un-round-da-2-7-milioni-operazione-guidata-dai-fondi-iff-e-entourage-202511131501462937) · Milano Finanza - [CommerceClarity, round investimento per potenziare piattaforma IA](https://www.ansa.it/sito/notizie/economia/2025/11/13/commerceclarityround-investimento-per-potenziare-piattaforma-ia_0a81cc7d-37d5-42d6-b745-aef02d827e65.html) · ANSA --- # Agent Product Manager **Team:** Product ## Locations - Milan · Italy - Barcelona · Spain **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about As Agent Product Manager you own the AI agents we build and deploy for enterprise customers, from discovery to production, from first value to the expansion of the account. You partner directly with engineering and with the customer to ship agents that handle thousands of product data operations a day. You will know more about how our agents behave in the real world than anyone else here, and you will use that to shape what we build next. ## What you would do ### Build and ship agents with customers Partner with engineering and with the customer to configure, test and deploy agents across catalog enrichment, assortment operations and data publishing. You own the agent, from requirements to go-live and past it. ### Engage deeply with customer stakeholders - Run discovery sessions to map business processes, data flows and the decision logic the agents have to carry. - Present demos and working prototypes to business and technical stakeholders, up to C-level. - Diagnose the hard edge cases hands-on, through prompt engineering, configuration and workflow logic. - Iterate on deployed agents against performance data and what the customer tells you. - Become the partner each account calls when it wants to think bigger. ### Drive commercial growth Find the expansion, articulate the incremental value and lead the conversation about it. Outcomes over features. ### Shape the roadmap Bring the patterns you see in the field to product and engineering. What shows up across deployments becomes what we build next. ## Who you are ### Essential - Five or more years in product, solutions engineering or technical account management, on complex SaaS or AI products. - Product management chops. You take a requirement from fuzzy to shipped. - A technical communicator: data engineer, category manager and CPO, in the same day. - Hands-on AI fluency. You have built or configured something real with LLMs, prompt engineering or agent workflows. - Commercial instinct. Value, objections and deals forward, without being salesy. - Fluent English. Italian, Spanish or another European language is a strong plus. - Comfortable with ambiguity and autonomy. Small team, fast moves. ### Preferred - A background in ecommerce, retail technology or supply chain. - Direct experience with AI agents or LLMs in production. - A record of growing accounts, or of moving product from what you learned in the field. - Early-stage startup experience. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## Why join us Agents are going to run a meaningful part of enterprise retail operations within five years. We are building that infrastructure now, for the companies that will lead European retail. This seat sits at the centre of it. ### What we offer - Influence over every part of the product, working directly with the founding team. - Relationships with the most forward-thinking commerce teams in Europe. - An AI-native company. Agents are your primary interface, not a legacy config panel. - Flexible hybrid. The Milan and Barcelona hubs, three days a week minimum, plus four remote weeks a year. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # AI Engineer **Team:** Engineering ## Locations - Milan · Italy - Barcelona · Spain **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about You build and ship the agents that run product data infrastructure for enterprise retailers across Europe. These are not prototypes. They are production systems handling thousands of catalog operations a day, where reliability, cost and accuracy land on somebody’s revenue. You own agents end to end: prompt design, orchestration, evaluation, deployment, monitoring and the iteration after that. You work with the other engineers, the CPO and the customer-facing teams to ship agents that hold up in messy real conditions. ## What you would do - Design, build and ship production-grade agents across catalog enrichment, classification, assortment and publishing. - Own the full lifecycle, from the first prompt to a deployed system with evaluation running against it. - Build the orchestration, the retrieval pipelines and the evaluation frameworks that keep agents reliable at scale. - Integrate Anthropic, OpenAI and other foundation models, and make the calls on model, cost and latency. - Optimize token usage, caching and observability across the agent stack. - Turn a messy customer problem into an agent that solves it, with the Agent Product Managers and the delivery team. - Feed what production teaches you back into the platform roadmap. - Raise the bar for how agents are built, tested and operated here. ## Who you are ### Essential - Three or more years shipping production systems, with hands-on LLM work in the last twelve months. - You have put an LLM API into a real product: Anthropic, OpenAI or similar. - Strong prompt engineering with systematic evaluation behind it. You do not ship on vibes. - Comfortable building and running retrieval pipelines, agent orchestration and vector search. - Solid Python, and fluent in the tooling that exists this year. - A real grasp of the tradeoffs: cost, latency, quality, failure modes. - Comfortable with ambiguity, fast iteration and incomplete information. - Clear thinker, low ego, high standards, open to feedback. - Fluent English, and strong writing. - Genuine curiosity about where this is going next. ### Preferred - You have built or led an agent product in production. - Eval frameworks, LLM observability, ML monitoring. - B2B SaaS, or an AI-native company. - Traditional ML and NLP as well: transformers, scikit-learn. - Orchestration frameworks: LangChain, LlamaIndex or similar. - Ecommerce, retail or catalog data. - Open source contributions, or technical writing on agents. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## Why join us You build agents that actually run in production at enterprise scale. Not demos, not proofs of concept. Real systems, real customers, real money on the line. And you do it on a team where AI engineering is the core of what we ship, not a feature beside it. ### What we offer - Meaningful early-stage equity. - Competitive salary. - An AI-native platform built for retail, the largest and fastest-moving market there is. - Founders who have built architectures serving more than 25 million users. - Agents that go live with named enterprise retailers, on tight feedback loops. - Flexible hybrid. The Milan and Barcelona hubs, three days a week minimum, plus four remote weeks a year. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # AI Engineer, Internship **Team:** Engineering ## Locations - Milan · Italy **Employment type:** Internship **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about Six months on the team that builds the agents. You take a real piece of the agent stack, with an AI Engineer next to you, and you ship it. Not a side project kept away from production: interns here work on the agents that run on customer catalogs, starting where a mistake is cheap to undo. The internship is designed to end in an offer if the work says so, and you will hear where you stand at every step rather than at the end. ## What you would do - Write and test prompts against real catalog data, and measure the output instead of eyeballing it. - Extend an agent workflow: a new step, a new tool, a new source of product data. - Write the evaluation cases that tell us whether an agent got better or worse. - Dig into failures. Read the traces, find the pattern, propose the fix. - Take one improvement all the way to production, reviewed, inside the first two months. ## Who you are ### Essential - You write Python on your own, without a tutorial open next to you. - You have built something with an LLM API, however small, and you can explain why it broke. - You are finishing or have just finished a degree in computer science, engineering, maths, physics or something equally quantitative. We read applications from people who taught themselves too. - You ask early when you are stuck, rather than losing two days to pride. - Fluent English, written and spoken. - In Milan, or moving there. Three days a week in the office, like everyone else. ### Preferred - Some exposure to retrieval pipelines, agent frameworks or evals. - A repository, a write-up or a project we can look at. - Italian, for the customer data you would be reading. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## Why join us The fastest way to learn how agents behave in production is to stand next to the people running them. That is what this seat is. ### What we offer - A paid six-month internship, with a mentor accountable for your growth and not only your tickets. - Work that reaches customers, and the credit for it. - A conversation about a permanent seat before the six months are up. - The Milan hub, three days a week minimum. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Business Development Representative, Italy **Team:** Sales & Partnerships ## Locations - Milan · Italy **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about Retailers and brands still spend a huge amount of time fixing product information by hand. CommerceClarity builds AI agents that take on that work, from organizing raw supplier data to producing complete, channel-ready product content. Our customers are already using the product. Now we need to start more of the right conversations. As our Business Development Representative for Italy, you will decide which companies to approach, find the right people inside them, and earn the first meeting. You will focus on Italy and also work on opportunities in the UK. We are looking for someone who is excited to use AI as part of the job and who can represent CommerceClarity with confidence, whether that is in an email, on the phone, or face to face at an industry event. ## Why this role is different This is not a role where you receive a list, follow a script, and hit a daily call target. We use our own agents for prospecting and outreach. Your job is to create opportunities while making that system better every week. You will review the accounts it finds, improve the messages it writes, and work out why some approaches get a reply while others do not. Good judgement matters more than activity for its own sake. You need to understand which companies are worth pursuing, who inside them is likely to care, and what will make the conversation relevant. You will speak with ecommerce leaders, catalog and PIM owners, CIOs, and operations teams at large retailers and brands. You will often be the first person they meet from CommerceClarity, so how you show up matters. You will work closely with the CRO and Account Executives while we continue to build our sales motion. What you learn from the market will influence our target accounts, our outreach, and sometimes the product itself. For the right person, this role can grow into an Account Executive position and later into leading part of the function. Progress here depends on the work you do and the ownership you take. ## What you would do - Research and build your own target lists, focusing on companies that fit our ICP and the people most likely to feel the problem. - Run outbound across email, LinkedIn, and phone in Italian and English. - Book qualified first meetings and prepare the Account Executive with useful context on the company and the people involved. - Work with our prospecting and outreach agents, review their output, and improve the lists, prompts, and sequences. - Build simple automations when the current stack is missing something, then share what works with the team. - Learn how retail catalog operations work, including PIMs, supplier data, product feeds, marketplaces, and the manual work that still sits between them. - Represent CommerceClarity at industry events and trade fairs, then turn those conversations into real opportunities. - Keep HubSpot accurate and up to date on the same day. - Share what you are learning from the market, including the messages that get replies, the people who engage, and the objections that keep coming back. - Qualify inbound opportunities quickly as that channel grows. ## What we are looking for ### Essential - One to three years of experience opening new business conversations, whether in B2B outbound, consulting, business development, or something you built yourself. - Examples of cold emails or messages that received replies, and a clear view on why they worked. - Hands-on experience using AI to build or improve something. We are looking for more than regular use of ChatGPT. - Confidence speaking with senior people, both on the phone and in person. - Curiosity about how technology works and the willingness to learn enough to hold a useful conversation with a technical buyer. - The ability to bring structure to an unclear problem and start without waiting for a perfect brief. - A healthy relationship with rejection. You learn from it, adjust, and keep going. - The ambition to take on more responsibility over time, starting by becoming excellent at this job. - Fluent Italian and English, written and spoken. Your English must be strong enough to open conversations with UK accounts. - Based in Milan or willing to relocate. We work from the office three days a week. ### Preferred - You have built something using an LLM API, an agent framework, or a tool such as Clay, and can explain what went wrong along the way. - You have prospected into retail, ecommerce, or consumer brands. - You have worked in a large organization or sold into one, and understand how decisions move through it. - You have represented a company or spoken at an industry event. - You have a quantitative degree or can show us equivalent skills you developed yourself. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## What we offer - Ownership. The accounts you open and the outreach you write shape the go-to-market playbook directly. - Growth. A high-impact team, real commercial complexity, tight feedback loops. - An ambitious roadmap. Help define the commercial infrastructure for agentic commerce in Europe. - Flexible hybrid. The Milan hub, three days a week minimum, plus four remote weeks a year. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Business Development Representative, UK **Team:** Sales & Partnerships ## Locations - London · United Kingdom **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about Retailers and brands still spend a huge amount of time fixing product information by hand. CommerceClarity builds AI agents that take on that work, from organizing raw supplier data to producing complete, channel-ready product content. Our customers are already using the product. Now we need to start more of the right conversations. As our Business Development Representative for the UK, you will decide which companies to approach, find the right people inside them, and earn the first meeting. You will focus on the UK and also work on selected opportunities across Europe. We are looking for someone who is excited to use AI as part of the job and who can represent CommerceClarity with confidence, whether that is in an email, on the phone, or face to face at an industry event. ## Why this role is different This is not a role where you receive a list, follow a script, and hit a daily call target. We use our own agents for prospecting and outreach. Your job is to create opportunities while making that system better every week. You will review the accounts it finds, improve the messages it writes, and work out why some approaches get a reply while others do not. Good judgement matters more than activity for its own sake. You need to understand which companies are worth pursuing, who inside them is likely to care, and what will make the conversation relevant. You will speak with ecommerce leaders, catalog and PIM owners, CIOs, and operations teams at large retailers and brands. You will often be the first person they meet from CommerceClarity, so how you show up matters. You will work closely with the CRO and Account Executives while we continue to build our sales motion. What you learn from the market will influence our target accounts, our outreach, and sometimes the product itself. For the right person, this role can grow into an Account Executive position and later into leading part of the function. Progress here depends on the work you do and the ownership you take. ## What you would do - Research and build your own target lists, focusing on companies that fit our ICP and the people most likely to feel the problem. - Run outbound across email, LinkedIn, and phone, primarily in English. - Book qualified first meetings and prepare the Account Executive with useful context on the company and the people involved. - Work with our prospecting and outreach agents, review their output, and improve the lists, prompts, and sequences. - Build simple automations when the current stack is missing something, then share what works with the team. - Learn how retail catalog operations work, including PIMs, supplier data, product feeds, marketplaces, and the manual work that still sits between them. - Represent CommerceClarity at industry events and trade fairs, then turn those conversations into real opportunities. - Keep HubSpot accurate and up to date on the same day. - Share what you are learning from the market, including the messages that get replies, the people who engage, and the objections that keep coming back. - Qualify inbound opportunities quickly as that channel grows. ## What we are looking for ### Essential - One to three years of experience opening new business conversations, whether in B2B outbound, consulting, business development, or something you built yourself. - Examples of cold emails or messages that received replies, and a clear view on why they worked. - Hands-on experience using AI to build or improve something. We are looking for more than regular use of ChatGPT. - Confidence speaking with senior people, both on the phone and in person. - Curiosity about how technology works and the willingness to learn enough to hold a useful conversation with a technical buyer. - The ability to bring structure to an unclear problem and start without waiting for a perfect brief. - A healthy relationship with rejection. You learn from it, adjust, and keep going. - The ambition to take on more responsibility over time, starting by becoming excellent at this job. - Fluent English, written and spoken. - Based in London or willing to relocate. We work from the office three days a week. ### Preferred - You have built something using an LLM API, an agent framework, or a tool such as Clay, and can explain what went wrong along the way. - You have prospected into retail, ecommerce, or consumer brands. - You have worked in a large organization or sold into one, and understand how decisions move through it. - You have represented a company or spoken at an industry event. - You speak another European language. - You have a quantitative degree or can show us equivalent skills you developed yourself. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## What we offer - Ownership. The accounts you open and the outreach you write shape the go-to-market playbook directly. - Growth. A high-impact team, real commercial complexity, tight feedback loops. - An ambitious roadmap. Help define the commercial infrastructure for agentic commerce in Europe. - Flexible hybrid. The London hub, three days a week minimum, plus four remote weeks a year. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Enterprise Account Executive, Germany **Team:** Sales & Partnerships ## Locations - Berlin · Germany **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about Retail is one of the largest industries in the world. It is also one of the least equipped to operate at the speed the market now demands. Product data fragmented across systems. Catalog operations still run on spreadsheets. Entire teams whose job is to fix what should never have broken. The companies that will define the next decade of commerce are the ones that solve this at the infrastructure level. Not with more people, but with agents that take the work off the table. CommerceClarity is building that infrastructure. The product works. The first customers are live and expanding. Now we need someone to take it to the market it deserves. As Enterprise Account Executive you own the full sales cycle with enterprise retailers and brands in Germany and the DACH region. You run deep discovery, you sell a different way of operating, and you turn what you learn in the field into the playbook that scales the motion. This is one of the first seats we are opening in Germany. You would work out of Berlin on DACH accounts, with the platform and the delivery method already built behind you in Milan and Barcelona. ## Why this role is different You will not inherit a territory, a script, or a CRM full of somebody else’s leads. The buyers are real decision-makers: heads of ecommerce, CIOs and operations leaders at enterprise retailers and brands. The problem is visceral and well understood. The answer is genuinely new. This is consultative selling in the truest sense, because you are not selling features, you are selling a different way of operating. What you learn in the field reshapes the product. What you close funds the roadmap. ## What you would do - Own the full sales cycle with enterprise retailers and brands. - Work with the founders to pressure-test and adapt the playbook, sharpening ICP fit, discovery quality and deal structure as the motion scales. - Manage and grow a network of technology partners and resellers. - Run deep discovery to surface the pain around catalog operations, time to market and channel complexity. - Bring market signal, competitive intelligence and the feature gaps that close deals back to product. - Work with Customer Success so a deal is set up for retention and expansion, not only for signature. - Represent CommerceClarity at industry events, trade fairs and with strategic partners. - Optimize for the company’s long-term success, not only for your own target. ## What we are looking for ### Essential - Five to seven years of B2B SaaS sales, ideally in retail technology, ecommerce or adjacent. - A record of closing enterprise deals on your own. - Experience running both direct and partner-led motions. - You build structure where there is none, without needing a perfect brief to start. - Sharp on the phone, tight in writing, adaptable in the room with different buyer personas. - You think in outcomes rather than activities, and you push back when the process is slowing the result. - The rigor of somebody who has sold into complex organizations, and the ambition of somebody who wants to build rather than execute. - German to native standard and fluent English, and you are at home in the DACH market. ### Preferred - You have sold AI or automation to retail, brand or ecommerce buyers. - You have been an early or founding sales hire, and shaped a go-to-market function. - Familiarity with enterprise buying that involves IT, operations and C-suite sign-off. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## What we offer - Ownership. Your deals reach production fast, and shape the go-to-market playbook directly. - Growth. A high-impact team, real commercial complexity, tight feedback loops. - An ambitious roadmap. Help define the commercial infrastructure for agentic commerce in Europe. - Flexible hybrid. Based in Berlin, three days a week minimum in person, plus four remote weeks a year. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Enterprise Account Executive, Italy **Team:** Sales & Partnerships ## Locations - Milan · Italy **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about Retail is one of the largest industries in the world. It is also one of the least equipped to operate at the speed the market now demands. Product data fragmented across systems. Catalog operations still run on spreadsheets. Entire teams whose job is to fix what should never have broken. The companies that will define the next decade of commerce are the ones that solve this at the infrastructure level. Not with more people, but with agents that take the work off the table. CommerceClarity is building that infrastructure. The product works. The first customers are live and expanding. Now we need someone to take it to the market it deserves. As Enterprise Account Executive you own the full sales cycle with enterprise retailers and brands in Italy. You run deep discovery, you sell a different way of operating, and you turn what you learn in the field into the playbook that scales the motion. ## Why this role is different You will not inherit a territory, a script, or a CRM full of somebody else’s leads. The buyers are real decision-makers: heads of ecommerce, CIOs and operations leaders at enterprise retailers and brands. The problem is visceral and well understood. The answer is genuinely new. This is consultative selling in the truest sense, because you are not selling features, you are selling a different way of operating. What you learn in the field reshapes the product. What you close funds the roadmap. ## What you would do - Own the full sales cycle with enterprise retailers and brands. - Work with the founders to pressure-test and adapt the playbook, sharpening ICP fit, discovery quality and deal structure as the motion scales. - Manage and grow a network of technology partners and resellers. - Run deep discovery to surface the pain around catalog operations, time to market and channel complexity. - Bring market signal, competitive intelligence and the feature gaps that close deals back to product. - Work with Customer Success so a deal is set up for retention and expansion, not only for signature. - Represent CommerceClarity at industry events, trade fairs and with strategic partners. - Optimize for the company’s long-term success, not only for your own target. ## What we are looking for ### Essential - Five to seven years of B2B SaaS sales, ideally in retail technology, ecommerce or adjacent. - A record of closing enterprise deals on your own. - Experience running both direct and partner-led motions. - You build structure where there is none, without needing a perfect brief to start. - Sharp on the phone, tight in writing, adaptable in the room with different buyer personas. - You think in outcomes rather than activities, and you push back when the process is slowing the result. - The rigour of somebody who has sold into complex organizations, and the ambition of somebody who wants to build rather than execute. - Fluent Italian and English. ### Preferred - You have sold AI or automation to retail, brand or ecommerce buyers. - You have been an early or founding sales hire, and shaped a go-to-market function. - Familiarity with enterprise buying that involves IT, operations and C-suite sign-off. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## What we offer - Ownership. Your deals reach production fast, and shape the go-to-market playbook directly. - Growth. A high-impact team, real commercial complexity, tight feedback loops. - An ambitious roadmap. Help define the commercial infrastructure for agentic commerce in Europe. - Flexible hybrid. The Milan hub, three days a week minimum, plus four remote weeks a year. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Enterprise Account Executive, UK **Team:** Sales & Partnerships ## Locations - London · United Kingdom **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about Retail is one of the largest industries in the world. It is also one of the least equipped to operate at the speed the market now demands. Product data fragmented across systems. Catalog operations still run on spreadsheets. Entire teams whose job is to fix what should never have broken. The companies that will define the next decade of commerce are the ones that solve this at the infrastructure level. Not with more people, but with agents that take the work off the table. CommerceClarity is building that infrastructure. The product works. The first customers are live and expanding. Now we need someone to take it to the market it deserves. As Enterprise Account Executive you own the full sales cycle with enterprise retailers and brands in the UK. You run deep discovery, you sell a different way of operating, and you turn what you learn in the field into the playbook that scales the motion. This is one of the first seats we are opening in the UK. You would work out of the London hub on UK accounts, with the platform and the delivery method already built behind you in Milan and Barcelona. ## Why this role is different You will not inherit a territory, a script, or a CRM full of somebody else’s leads. The buyers are real decision-makers: heads of ecommerce, CIOs and operations leaders at enterprise retailers and brands. The problem is visceral and well understood. The answer is genuinely new. This is consultative selling in the truest sense, because you are not selling features, you are selling a different way of operating. What you learn in the field reshapes the product. What you close funds the roadmap. ## What you would do - Own the full sales cycle with enterprise retailers and brands. - Work with the founders to pressure-test and adapt the playbook, sharpening ICP fit, discovery quality and deal structure as the motion scales. - Manage and grow a network of technology partners and resellers. - Run deep discovery to surface the pain around catalog operations, time to market and channel complexity. - Bring market signal, competitive intelligence and the feature gaps that close deals back to product. - Work with Customer Success so a deal is set up for retention and expansion, not only for signature. - Represent CommerceClarity at industry events, trade fairs and with strategic partners. - Optimize for the company’s long-term success, not only for your own target. ## What we are looking for ### Essential - Five to seven years of B2B SaaS sales, ideally in retail technology, ecommerce or adjacent. - A record of closing enterprise deals on your own. - Experience running both direct and partner-led motions. - You build structure where there is none, without needing a perfect brief to start. - Sharp on the phone, tight in writing, adaptable in the room with different buyer personas. - You think in outcomes rather than activities, and you push back when the process is slowing the result. - The rigour of somebody who has sold into complex organizations, and the ambition of somebody who wants to build rather than execute. - English to native standard, and you are at home in the UK market. Italian is a plus. ### Preferred - You have sold AI or automation to retail, brand or ecommerce buyers. - You have been an early or founding sales hire, and shaped a go-to-market function. - Familiarity with enterprise buying that involves IT, operations and C-suite sign-off. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## What we offer - Ownership. Your deals reach production fast, and shape the go-to-market playbook directly. - Growth. A high-impact team, real commercial complexity, tight feedback loops. - An ambitious roadmap. Help define the commercial infrastructure for agentic commerce in Europe. - Flexible hybrid. The London hub, three days a week minimum, plus four remote weeks a year. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Forward Deployed Engineer, Italy **Team:** Service & Delivery ## Locations - Milan · Italy **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about A Forward Deployed Engineer works inside the customer’s stack. You go where the product data actually lives, the PIM nobody has migrated, the ERP export that arrives as a fixed-width file, the marketplace feed whose rules exist in an email from 2019, and you make our agents work against it. This is the engineering half of a deployment. The Retail Engineer owns the account and the method. You own the code at the edge: the integrations, the transformations, the pipelines, the parts that decide whether an agent reads clean input or garbage. Deployed forward means the customer’s problem is your problem, and that you are in the room while it is being described rather than reading a ticket about it a week later. You work on accounts in Italy. ## What you would do - Build the integrations that connect our agents to customer systems: PIM, ERP, ecommerce platforms, marketplaces, data warehouses, and whatever else holds the catalog. - Write the transformations that turn a customer’s export into something an agent can reason about, and the agent’s output into something their systems accept. - Stand up pipelines that run at volume, with retries, with observability, and with somebody woken up when they fail. - Debug in production, on their data, with them watching. Then make that class of bug impossible. - Ship the change that unblocks the deployment today, then decide with the platform team whether it belongs in the product. - Leave a runbook behind, so the account does not depend on your memory. ## Who you are ### Essential - Three or more years writing production code, and you have owned something on call. - Strong Python or TypeScript, and willing to work in whatever the customer’s stack forces on you. - You are good with other people’s data: broken encodings, silent truncation, three sources of truth that disagree. - SQL you reach for without thinking, and comfort with the shape of a data warehouse. - APIs, queues, batch jobs, retries, idempotency. You know why the second run is the hard one. - You are calm in front of a customer when something is broken, and you say what you know and what you do not. - You use AI coding assistants as a matter of course, and you can tell where they help and where they do not. - Willing to travel to customer sites. - Fluent Italian and English. ### Preferred - You have been a forward deployed engineer, a solutions engineer or an implementation engineer at an enterprise software company. - Retail or ecommerce data: catalogs, feeds, attributes, marketplace rules. - Workflow orchestration: n8n, Airflow, Temporal or similar. - LLM work in production, and an opinion about evals. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## Why join us The hardest part of putting agents inside an enterprise is never the model. It is everything the model has to touch. That is the part you own, and it is the part that decides whether any of it works. ### What we offer - Meaningful early-stage equity. - Competitive salary. - The shortest distance there is between a line of code and a retailer’s catalog. - A new craft. You learn to build and ship with agents and AI coding assistants, and it compounds. - Flexible hybrid. The Milan hub, three days a week minimum, plus four remote weeks a year. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Forward Deployed Engineer, UK **Team:** Service & Delivery ## Locations - London · United Kingdom **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about A Forward Deployed Engineer works inside the customer’s stack. You go where the product data actually lives, the PIM nobody has migrated, the ERP export that arrives as a fixed-width file, the marketplace feed whose rules exist in an email from 2019, and you make our agents work against it. This is the engineering half of a deployment. The Retail Engineer owns the account and the method. You own the code at the edge: the integrations, the transformations, the pipelines, the parts that decide whether an agent reads clean input or garbage. Deployed forward means the customer’s problem is your problem, and that you are in the room while it is being described rather than reading a ticket about it a week later. You work on accounts in the UK. This is one of the first seats we are opening in the UK. You would work out of the London hub on UK accounts, with the platform and the delivery method already built behind you in Milan and Barcelona. ## What you would do - Build the integrations that connect our agents to customer systems: PIM, ERP, ecommerce platforms, marketplaces, data warehouses, and whatever else holds the catalog. - Write the transformations that turn a customer’s export into something an agent can reason about, and the agent’s output into something their systems accept. - Stand up pipelines that run at volume, with retries, with observability, and with somebody woken up when they fail. - Debug in production, on their data, with them watching. Then make that class of bug impossible. - Ship the change that unblocks the deployment today, then decide with the platform team whether it belongs in the product. - Leave a runbook behind, so the account does not depend on your memory. ## Who you are ### Essential - Three or more years writing production code, and you have owned something on call. - Strong Python or TypeScript, and willing to work in whatever the customer’s stack forces on you. - You are good with other people’s data: broken encodings, silent truncation, three sources of truth that disagree. - SQL you reach for without thinking, and comfort with the shape of a data warehouse. - APIs, queues, batch jobs, retries, idempotency. You know why the second run is the hard one. - You are calm in front of a customer when something is broken, and you say what you know and what you do not. - You use AI coding assistants as a matter of course, and you can tell where they help and where they do not. - Willing to travel to customer sites. - English to native standard, and you are at home in the UK market. Italian is a plus. ### Preferred - You have been a forward deployed engineer, a solutions engineer or an implementation engineer at an enterprise software company. - Retail or ecommerce data: catalogs, feeds, attributes, marketplace rules. - Workflow orchestration: n8n, Airflow, Temporal or similar. - LLM work in production, and an opinion about evals. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## Why join us The hardest part of putting agents inside an enterprise is never the model. It is everything the model has to touch. That is the part you own, and it is the part that decides whether any of it works. ### What we offer - Meaningful early-stage equity. - Competitive salary. - The shortest distance there is between a line of code and a retailer’s catalog. - A new craft. You learn to build and ship with agents and AI coding assistants, and it compounds. - Flexible hybrid. The London hub, three days a week minimum, plus four remote weeks a year. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Founding Head of Engineering **Team:** Engineering ## Locations - Milan · Italy - Barcelona · Spain **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about We are looking for a Founding Head of Engineering, the person who owns engineering at CommerceClarity. You own the function: the architecture, the technical bar, delivery, and the four engineers on the team. The seat is hybrid. Base yourself anywhere in the EU and travel to Milan and Barcelona once a month. ## What you would own - The core platform architecture and the technology choices behind it. You stay a product builder. - The engineering standards: code quality, review, testing, and what counts as done. - The team. You manage it and you grow it. - The technical roadmap with the CPTO, and what engineering commits to. - Data architecture for high-volume catalog work: millions of SKUs, many languages, many channels. - Infrastructure and the operational bar: infrastructure as code, CI/CD, monitoring, security, high availability, incident response. - The definition of what the platform can do with AI. ## Who you are ### Essential - You have managed engineers directly. - Around ten years building platforms or SaaS, with architecture you actually owned. - You have built inside an early-stage startup. - You are fluent at using LLMs to orchestrate several coding agents. - You have shipped AI agents: prompting, tool use, orchestration, evals. - Deep data architecture knowledge under real traffic. - Comfortable deciding on incomplete information, and iterating fast. - Clear with technical and business stakeholders alike, and you hold your position when it matters. - Based in the EU, able to travel to Milan and Barcelona once a month. - Fluent English. ### Preferred - Retail and ecommerce experience. - You have built software for large enterprises. - You have been a technical co-founder. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## Why join us You architect the engineering foundation of a platform changing how enterprise retailers run their commerce operations, next to founders who have operated retail at the largest scale in Europe. ### What we offer - A founding-level equity package. - Competitive salary. - Hybrid. Anywhere in the EU, with monthly travel to Milan and Barcelona. - The engineering function to build and lead, from the ground up. - Founders who have built architectures serving more than 25 million users. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Platform Engineer **Team:** Engineering ## Locations - Milan · Italy - Barcelona · Spain **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about CommerceClarity builds AI agents that run catalog operations for enterprise retailers. The agents are the product. This role builds what they run on. You own the infrastructure under the platform: the execution and orchestration layer where workflows run, the data layer under millions of SKUs, the APIs, the deploys, and the observability that tells us when an agent is slow, wrong or expensive. You need to have worked around AI systems, because what you are running is agents and the failure modes are different from a normal backend. But the depth we are hiring for is infrastructure and code, not prompting. You would be the engineer who makes the platform hold as volume grows, working directly with the CPO and the tech lead. The seat is hybrid. Base yourself in Milan, in Barcelona, or anywhere in the EU and travel to one of the two offices once a month. ## What you would do - Own the orchestration and execution layer where agent workflows run, and make it reliable: retries, queues, idempotency, backpressure, isolation between customers. - Build the data layer for high-volume catalog work: millions of SKUs, many languages, many channels, and the throughput to process them without falling over. - Push the platform towards API-first, so every capability is callable and nothing lives only in a screen. - Build the observability across the agent stack: traces, LLM cost and token usage, latency, quality signals, alerting that fires before the customer notices. - Own infrastructure as code, CI/CD, environments, secrets, and the security bar. - Build and own the integration layer with retailer systems: PIM, ERP, ecommerce platforms, marketplaces, data warehouses. - Take a prototype the team built in a week and turn it into something that holds under enterprise load. - Be on the hook for production. You debug it, you fix it, you make sure it does not come back. ## Who you are ### Essential - Five or more years building backend and infrastructure in production, with systems you actually owned. - Strong distributed systems fundamentals: queues, async processing, concurrency, failure handling, data consistency. - You have run data pipelines at real volume and know where they break. - Infrastructure as code, CI/CD, containers, monitoring and alerting are daily tools, not something you read about. - Hands-on with LLM systems in production, even if you did not write the prompts. You understand cost, latency, rate limits, non-determinism and how to instrument them. - Strong in at least one backend language, and willing to work in the stack we have rather than the one you prefer. - You work fast with AI coding tools and you have opinions about where they help and where they lie. - Comfortable with ambiguity and with a spec that arrives as a conversation. - Low ego, high standards, direct in writing. - Based in the EU, able to travel to Milan or Barcelona once a month. - Fluent English. ### Preferred - Workflow orchestration engines, Windmill, Temporal, Airflow, n8n or similar. - Laravel or PHP, React and MongoDB, which is what the platform runs on today. - LLM observability and eval plumbing: Langfuse, LangSmith or equivalent. - You have built multi-tenant B2B infrastructure for enterprise customers. - Ecommerce, retail, PIM or catalog data. - You have been early at a startup and remember what that costs. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## Why join us The agents only work if what runs them works. You own that layer, with production traffic from named enterprise retailers on it from the first week, on a team small enough that what you decide is in the product within days. ### What we offer - Meaningful early-stage equity. - Competitive salary. - Hybrid. Anywhere in the EU, with monthly travel to Milan or Barcelona. - The infrastructure layer of an AI-native platform to own outright, at the point where it is being built. - Founders who have built architectures serving more than 25 million users. - Agents that go live with named enterprise retailers, on tight feedback loops. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Retail Engineer (Deployment Strategist), Germany **Team:** Service & Delivery ## Locations - Berlin · Germany **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about Retailers still run too much catalog work through spreadsheets and copy-paste: filling missing attributes, adapting products to each marketplace, and keeping thousands of SKUs accurate. CommerceClarity replaces that manual work with AI agents. As a [Retail Engineer](/blog/meet-the-retail-engineer) you put those agents to work inside each customer’s operation, from discovery and workflow design to integration and production. Part retail consultant, part engineer, working with our customers across Germany and the DACH region. This is a high-ownership seat. You own our most strategic accounts end to end and you set the delivery method the rest of the team runs on. We weigh attitude and the combination over years on paper: the sweet spot is three to six years, and the right builder energy beats seniority. This is one of the first seats we are opening in Germany. You would work out of Berlin on DACH accounts, with the platform and the delivery method already built behind you in Milan and Barcelona. ## A job that did not exist three years ago New tools tend to produce job titles that were not there a few years earlier. Growth engineer and marketing engineer both appeared once the people who understood the funnel could write the code that ran it, and forward deployed engineer and deployment strategist appeared when the same thing happened to enterprise delivery. None of them was a rename. They exist because splitting the person who understands the process from the person who builds it stopped paying for itself. Retail Engineer is that seat for catalog operations. Outside retail the same job is called deployment strategist, and most of the people who fit it have spent five or six years in solution delivery or solution consulting at a retail software vendor or a consultancy. They read business processes for a living and they own the customer, and almost all of them have a technical side they were never given room to use: the configuration screen ended exactly where the interesting part started, and everything past it went to a development team on another continent. What changed is the [platform](/platform). The agents are built there, changed there and run there, so the person who designs the workflow is the person who builds it, in front of the customer, in the same week. You write the prompts and you wire up their systems, and when a number moves you go and change whatever produced it. AI coding assistants work alongside you, so a change that used to take a sprint becomes a prompt and a test. The half of the job you were already good at stays. Discovery, a room full of stakeholders, an account that trusts you: here that is what keeps the building honest, because you are the one who heard what they actually meant. What you gain is the other half. You spend your days on agent design, evals and prompt engineering, inside a company that runs its own operations on agents, which is a faster education in this than anything you could sign up for. ## Why this role is different It is not pure delivery and it is not pure engineering. You sit at the intersection: you understand the retailer’s catalog workflow well enough to redesign it around agents, and you have the depth to build the agent flows yourself. You cover the whole lifecycle. The preparation, mapping their processes, designing the workflow, writing and iterating the prompts, defining the KPIs. Then the run, deployment, integration with their systems, monitoring, debugging, and optimization against those KPIs. And you do not only deliver. You shape how we deliver and you grow the engineers around you, so one senior hire lifts the team instead of adding a pair of hands. You work closely with the Agent Product Managers, who own the product direction, and with the AI Engineers on delivery. ## What you would do - Lead agent implementations end to end for enterprise retailers, from kickoff to production, on our DIPE method: design, implement, prove, evolve. - Own our most strategic accounts and the heaviest proofs of concept directly. - Define the success KPIs with the customer, and build the measurement loop that proves the agents are working. - Design and build the agent workflows, and write the prompt layer that drives them. - Integrate agents with customer systems: PIM, ERP, ecommerce platforms, marketplaces, data warehouses. - Own the run. Monitor, debug, iterate on prompts and flows, and move the numbers over time. - Structure how we deliver, and grow the engineers around you, so the team scales on method instead of drowning. - Feed what you learn back into the product and the agent roadmap. ## What we are looking for ### Essential - Three or more years of experience, most of it in consulting or solution delivery. Many of our best fits are management engineers who read a business process and enjoy working with data. - You sit between business and technology. You can redesign a retailer’s catalog workflow and build the agent flow yourself. Not a pure PM, not a pure data engineer. - Prompt engineering sense. You iterate against real data, you measure output quality, and you know when to fix something with a prompt and when to fix it with code. - Standing with enterprise customers, and a record of bringing complex delivery home. - You have managed projects and people, and you still want to build rather than only coordinate. - Customer-facing and high agency. You handle ambiguity, you lead the conversation with the retailer, and you do not wait for a perfect brief. - German to native standard and fluent English, and you are at home in the DACH market. ### Preferred - Hands-on with workflow builders such as n8n, and with AI-native coding tools such as Claude Code or Codex. - Comfortable coding with AI assistance, in Python or TypeScript, to glue systems together and extend what an agent does. - You have worked with a workflow orchestration engine. - Deep retail workflow knowledge: catalog management, PIM, assortment, attribute governance, channel publishing, marketplace operations. - The major PIM platforms, Akeneo, Salsify, inRiver, Pimcore, and the feed and channel managers. - A background as Forward Deployed Engineer, Solutions Engineer, Implementation Lead or technical CSM at an enterprise software company. - Retail data quality at scale: millions of SKUs, many languages, many marketplaces. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## What we offer - Ownership. Your work reaches production and customers fast, and shapes the delivery method directly. - Growth. A high-leverage team, real commercial complexity, tight feedback loops. - An ambitious roadmap. Help define the operating infrastructure for agentic commerce in Europe. - A new craft. You learn to build and ship with agents and AI coding assistants, and it compounds. - Flexible hybrid. Based in Berlin, three days a week minimum in person, plus four remote weeks a year. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Retail Engineer (Deployment Strategist), Italy **Team:** Service & Delivery ## Locations - Milan · Italy **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about Retailers still run too much catalog work through spreadsheets and copy-paste: filling missing attributes, adapting products to each marketplace, and keeping thousands of SKUs accurate. CommerceClarity replaces that manual work with AI agents. As a [Retail Engineer](/blog/meet-the-retail-engineer), you stay with the customer from discovery through production. You redesign their catalog workflow around agents, build the flows yourself and improve them against the agreed KPIs, working with customers across Italy. This is a high-ownership seat. You own our most strategic accounts end to end and you set the delivery method the rest of the team runs on. We weigh attitude and the combination over years on paper: the sweet spot is three to six years, and the right builder energy beats seniority. Outside retail the same seat is called deployment strategist. If you come from solution delivery, solution consulting or implementation, you already have half of it: you read the operation, you hold the customer, you bring complex delivery home. The other half is that here you build the agent yourself instead of writing the spec and handing it to a development team. ## What you would do - Lead agent implementations end to end for enterprise retailers, from kickoff to production, on our DIPE method: design, implement, prove, evolve. - Own our most strategic accounts and the heaviest proofs of concept directly. - Define the success KPIs with the customer, and build the measurement loop that proves the agents are working. - Design and build the agent workflows, and write the prompt layer that drives them. - Integrate agents with customer systems: PIM, ERP, ecommerce platforms, marketplaces, data warehouses. - Own the run. Monitor, debug, iterate on prompts and flows, and move the numbers over time. - Structure how we deliver, and grow the engineers around you, so the team scales on method instead of drowning. - Feed what you learn back into the product and the agent roadmap. ## What we are looking for ### Essential - Three or more years of experience, most of it in consulting or solution delivery. Many of our best fits are management engineers who read a business process and enjoy working with data. - You sit between business and technology. You can redesign a retailer’s catalog workflow and build the agent flow yourself. Not a pure PM, not a pure data engineer. - Prompt engineering sense. You iterate against real data, you measure output quality, and you know when to fix something with a prompt and when to fix it with code. - Standing with enterprise customers, and a record of bringing complex delivery home. - You have managed projects and people, and you still want to build rather than only coordinate. - Customer-facing and high agency. You handle ambiguity, you lead the conversation with the retailer, and you do not wait for a perfect brief. - Fluent Italian and English. ### Preferred - Hands-on with workflow builders such as n8n, and with AI-native coding tools such as Claude Code or Codex. - Comfortable coding with AI assistance, in Python or TypeScript, to glue systems together and extend what an agent does. - You have worked with a workflow orchestration engine. - Deep retail workflow knowledge: catalog management, PIM, assortment, attribute governance, channel publishing, marketplace operations. - The major PIM platforms, Akeneo, Salsify, inRiver, Pimcore, and the feed and channel managers. - A background as Forward Deployed Engineer, Solutions Engineer, Implementation Lead or technical CSM at an enterprise software company. - Retail data quality at scale: millions of SKUs, many languages, many marketplaces. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## What we offer - Ownership. Your work reaches production and customers fast, and shapes the delivery method directly. - Growth. A high-leverage team, real commercial complexity, tight feedback loops. - An ambitious roadmap. Help define the operating infrastructure for agentic commerce in Europe. - A new craft. You learn to build and ship with agents and AI coding assistants, and it compounds. - Flexible hybrid. The Milan hub, three days a week minimum, plus four remote weeks a year. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Retail Engineer (Deployment Strategist), UK **Team:** Service & Delivery ## Locations - London · United Kingdom **Employment type:** Full-time **Office addresses** **Hq** - **Street address:** Via Cerva 20 - **Address locality:** Milan - **Address region:** Lombardia - **Postal code:** 20122 - **Address country:** IT ### Offices - Norrsken House, Passeig del Mare Nostrum 15 · Barcelona · Catalunya · 08039 · ES - 167-169 Great Portland Street, 5th Floor · London · England · W1W 5PF · GB ## What the role is about Retailers still run too much catalog work through spreadsheets and copy-paste: filling missing attributes, adapting products to each marketplace, and keeping thousands of SKUs accurate. CommerceClarity replaces that manual work with AI agents. As a [Retail Engineer](/blog/meet-the-retail-engineer) you put those agents to work inside each customer’s operation, from discovery and workflow design to integration and production. Part retail consultant, part engineer, working with our customers across the UK. This is a high-ownership seat. You own our most strategic accounts end to end and you set the delivery method the rest of the team runs on. We weigh attitude and the combination over years on paper: the sweet spot is three to six years, and the right builder energy beats seniority. This is one of the first seats we are opening in the UK. You would work out of the London hub on UK accounts, with the platform and the delivery method already built behind you in Milan and Barcelona. ## A job that did not exist three years ago New tools tend to produce job titles that were not there a few years earlier. Growth engineer and marketing engineer both appeared once the people who understood the funnel could write the code that ran it, and forward deployed engineer and deployment strategist appeared when the same thing happened to enterprise delivery. None of them was a rename. They exist because splitting the person who understands the process from the person who builds it stopped paying for itself. Retail Engineer is that seat for catalog operations. Outside retail the same job is called deployment strategist, and most of the people who fit it have spent five or six years in solution delivery or solution consulting at a retail software vendor or a consultancy. They read business processes for a living and they own the customer, and almost all of them have a technical side they were never given room to use: the configuration screen ended exactly where the interesting part started, and everything past it went to a development team on another continent. What changed is the [platform](/platform). The agents are built there, changed there and run there, so the person who designs the workflow is the person who builds it, in front of the customer, in the same week. You write the prompts and you wire up their systems, and when a number moves you go and change whatever produced it. AI coding assistants work alongside you, so a change that used to take a sprint becomes a prompt and a test. The half of the job you were already good at stays. Discovery, a room full of stakeholders, an account that trusts you: here that is what keeps the building honest, because you are the one who heard what they actually meant. What you gain is the other half. You spend your days on agent design, evals and prompt engineering, inside a company that runs its own operations on agents, which is a faster education in this than anything you could sign up for. ## Why this role is different It is not pure delivery and it is not pure engineering. You sit at the intersection: you understand the retailer’s catalog workflow well enough to redesign it around agents, and you have the depth to build the agent flows yourself. You cover the whole lifecycle. The preparation, mapping their processes, designing the workflow, writing and iterating the prompts, defining the KPIs. Then the run, deployment, integration with their systems, monitoring, debugging, and optimization against those KPIs. And you do not only deliver. You shape how we deliver and you grow the engineers around you, so one senior hire lifts the team instead of adding a pair of hands. You work closely with the Agent Product Managers, who own the product direction, and with the AI Engineers on delivery. ## What you would do - Lead agent implementations end to end for enterprise retailers, from kickoff to production, on our DIPE method: design, implement, prove, evolve. - Own our most strategic accounts and the heaviest proofs of concept directly. - Define the success KPIs with the customer, and build the measurement loop that proves the agents are working. - Design and build the agent workflows, and write the prompt layer that drives them. - Integrate agents with customer systems: PIM, ERP, ecommerce platforms, marketplaces, data warehouses. - Own the run. Monitor, debug, iterate on prompts and flows, and move the numbers over time. - Structure how we deliver, and grow the engineers around you, so the team scales on method instead of drowning. - Feed what you learn back into the product and the agent roadmap. ## What we are looking for ### Essential - Three or more years of experience, most of it in consulting or solution delivery. Many of our best fits are management engineers who read a business process and enjoy working with data. - You sit between business and technology. You can redesign a retailer’s catalog workflow and build the agent flow yourself. Not a pure PM, not a pure data engineer. - Prompt engineering sense. You iterate against real data, you measure output quality, and you know when to fix something with a prompt and when to fix it with code. - Standing with enterprise customers, and a record of bringing complex delivery home. - You have managed projects and people, and you still want to build rather than only coordinate. - Customer-facing and high agency. You handle ambiguity, you lead the conversation with the retailer, and you do not wait for a perfect brief. - English to native standard, and you are at home in the UK market. Italian is a plus. ### Preferred - Hands-on with workflow builders such as n8n, and with AI-native coding tools such as Claude Code or Codex. - Comfortable coding with AI assistance, in Python or TypeScript, to glue systems together and extend what an agent does. - You have worked with a workflow orchestration engine. - Deep retail workflow knowledge: catalog management, PIM, assortment, attribute governance, channel publishing, marketplace operations. - The major PIM platforms, Akeneo, Salsify, inRiver, Pimcore, and the feed and channel managers. - A background as Forward Deployed Engineer, Solutions Engineer, Implementation Lead or technical CSM at an enterprise software company. - Retail data quality at scale: millions of SKUs, many languages, many marketplaces. ## The team you would join Our founders have spent their careers building and operating retail at scale. They have founded retail companies generating tens of millions in revenue, launched and scaled Amazon’s operations in Italy and Spain, led operations at Everli across more than 2,000 employees, and shaped product direction at Shopify and Zalando. Operator, founder, product leader. Every seat that defined modern European commerce. ## What we offer - Ownership. Your work reaches production and customers fast, and shapes the delivery method directly. - Growth. A high-leverage team, real commercial complexity, tight feedback loops. - An ambitious roadmap. Help define the operating infrastructure for agentic commerce in Europe. - A new craft. You learn to build and ship with agents and AI coding assistants, and it compounds. - Flexible hybrid. The London hub, three days a week minimum, plus four remote weeks a year. **Is open:** yes **Footnote:** CommerceClarity is an equal opportunity employer. We hire on what you can do and how you work with people. --- # Alessandro Angelini > Alessandro Angelini is co-founder and chief revenue officer of CommerceClarity, where he owns go to market, sales and marketing. He spent seven years as chief operating and financial officer of Everli, joining at one million in revenue and leaving at more than a hundred million, with the team grown from ten people to over three hundred across four European countries. He was a consultant at Bain and a strategy manager at EDF before it, and holds an MBA from SDA Bocconi. Alessandro is co-founder and CRO of CommerceClarity, and owns go to market, sales and marketing both. He spent seven years as chief operating and financial officer of Everli, which he joined at one million in revenue and left at more than a hundred million, with the team grown from ten people to over three hundred across four European countries and more than a hundred million dollars raised to fund it. He ran finance and operations there, and at different points product, international expansion and the commercial organisation, and he sourced and closed the company's acquisition in Poland. He also negotiated the first national agreement in the European Union regulating self-employed workers on gig platforms. He was interim chief strategy and financial officer at Unobravo after that, and before Everli a consultant at Bain and a strategy manager at EDF in the United Kingdom. He has an MBA from SDA Bocconi. **Role:** Co-founder and CRO, CommerceClarity **Linkedin:** [https://www.linkedin.com/in/alessandro-angelini-33b19220/](https://www.linkedin.com/in/alessandro-angelini-33b19220/) **Given name:** Alessandro **Family name:** Angelini ## Expertise - Ecommerce operations - Operations management - Corporate finance - Mergers and acquisitions - International expansion - Organizational design - Fundraising - Go to market strategy - Enterprise sales - P&L management ## Languages - Italiano - English ## Work history - CommerceClarity · Co-founder and CRO · [https://commerceclarity.com](https://commerceclarity.com) · 2024 - Unobravo · Interim Chief Strategy and Financial Officer · [https://www.unobravo.com](https://www.unobravo.com) · 2024 · 2025 - Everli · Chief Operating and Financial Officer · [https://everli.com](https://everli.com) · 2016 · 2024 - Bain & Company · Consultant · [https://www.bain.com](https://www.bain.com) · 2016 · 2016 - EDF · Strategy Manager · [https://www.edfenergy.com](https://www.edfenergy.com) · 2013 · 2015 - Edison · Financial Planning and Control Analyst · [https://www.edison.it](https://www.edison.it) · 2010 · 2013 - Ernst & Young · Consultant · [https://www.ey.com](https://www.ey.com) · 2009 · 2010 ## Education - SDA Bocconi · MBA · [https://www.sdabocconi.it/](https://www.sdabocconi.it/) · 2016 - Stanford University Graduate School of Business · Innovation and Growth Program · [https://www.gsb.stanford.edu/](https://www.gsb.stanford.edu/) ## Featured in - Firmato l'accordo tra Assogrocery e le rappresentanze sindacali degli shopper · [https://www.mark-up.it/firmato-laccordo-tra-assogrocery-e-le-rappresentanze-sindacali-degli-shopper/](https://www.mark-up.it/firmato-laccordo-tra-assogrocery-e-le-rappresentanze-sindacali-degli-shopper/) · Mark Up · 2024-02-21 ## In press - [CommerceClarity raises €2.7M ($3.1M) pre-seed to power the AI agentic era of e-commerce](/press/commerceclarity-raises-3m-ai-commerce) · 2025-11-13 ## Entities - [Everli](https://www.everli.com) · Organization, mentions - [Agentic commerce](https://en.wikipedia.org/wiki/Agentic_commerce) · Thing, mentions ## Sources - [Why we invested in CommerceClarity](https://www.italianfoundersfund.com/posts/why-we-invested-in-commerceclarity) · Italian Founders Fund - [Everli, the European marketplace for online grocery shopping, bags $100M Series C](https://techcrunch.com/2021/03/29/everli/) · TechCrunch - [L'accordo integrativo Everli per la regolamentazione dello Shopper](https://farecontrattazione.adapt.it/laccordo-integrativo-everli-per-la-regolamentazione-dello-shopper/) · ADAPT - [CommerceClarity](https://redstone.vc/portfolio/commerceclarity) · Redstone --- # Daniele Vella > Daniele Vella is a co-founder of CommerceClarity and a serial ecommerce founder before it, with more than five companies behind him. He built an electronics ecommerce from zero to 30 million in revenue, founded an agency that helped businesses sell on Amazon, founded CiaoBazar, a marketplace with a conversational shopping assistant, and co-founded Nowstock, which sold to an Italian electronics retailer. Daniele is a co-founder of CommerceClarity, and a serial founder before it: more than five companies, all of them in ecommerce. He built an electronics ecommerce from zero to 30 million in revenue, then founded an agency that helped businesses sell on Amazon, and a services company after it. He also founded CiaoBazar, a marketplace with a conversational shopping assistant, and co-founded Nowstock, which sold to an Italian electronics retailer. He has been building e-commerce businesses and apps since he was a teenager. **Role:** Co-founder, CommerceClarity **Linkedin:** [https://www.linkedin.com/in/daniele-vella-52525396/](https://www.linkedin.com/in/daniele-vella-52525396/) **Given name:** Daniele **Family name:** Vella ## Profiles - https://twitter.com/danivellax ## Expertise - Agentic commerce - AI agents - Catalog operations - Product data management - Ecommerce operations - Pricing optimization - Marketplace development - Conversational commerce - Technical architecture - SEO - Paid media - Amazon marketplace ## Languages - Italian - English ## Work history - CommerceClarity · Co-founder · [https://commerceclarity.com](https://commerceclarity.com) · 2024 - Nowstock · Co-founder · 2022 · 2023 - CiaoBazar · Founder and CEO · [https://ciaobazar.it](https://ciaobazar.it) · 2020 · 2023 - Sunex · Founder and CEO · 2020 · 2023 - DV Consulting · Founder and CEO · 2020 · 2023 ## Featured in ### E-commerce, la rete di agenti di intelligenza artificiale (su cui investono Elkann e Berlusconi) [https://www.corriere.it/tecnologia/cards/e-commerce-rete-agenti-intelligenza-artificiale-su-cui-investono-elkann-berlusconi/commerceclarity_principale.shtml](https://www.corriere.it/tecnologia/cards/e-commerce-rete-agenti-intelligenza-artificiale-su-cui-investono-elkann-berlusconi/commerceclarity_principale.shtml) Corriere della Sera 2025-11-13 ### CommerceClarity raccoglie 2,7 milioni per potenziare e-commerce con IA [https://forbes.it/2024/11/13/commerceclarity-raccoglie-2-7-milioni-per-potenziare-e-commerce-con-ia/](https://forbes.it/2024/11/13/commerceclarity-raccoglie-2-7-milioni-per-potenziare-e-commerce-con-ia/) Forbes Italia 2025-11-13 ### CommerceClarity, round investimento per potenziare piattaforma Ia [https://www.ansa.it/sito/notizie/economia/2025/11/13/commerceclarityround-investimento-per-potenziare-piattaforma-ia_0a81cc7d-37d5-42d6-b745-aef02d827e65.html](https://www.ansa.it/sito/notizie/economia/2025/11/13/commerceclarityround-investimento-per-potenziare-piattaforma-ia_0a81cc7d-37d5-42d6-b745-aef02d827e65.html) ANSA 2025-11-13 ### CommerceClarity chiude un round da 2,7 milioni: operazione guidata dai fondi Iff e Entourage [https://www.milanofinanza.it/news/commerceclarity-chiude-un-round-da-2-7-milioni-operazione-guidata-dai-fondi-iff-e-entourage-202411131629156634](https://www.milanofinanza.it/news/commerceclarity-chiude-un-round-da-2-7-milioni-operazione-guidata-dai-fondi-iff-e-entourage-202411131629156634) Milano Finanza 2025-11-13 ### L'Intelligenza artificiale per e-commerce di CommerceClarity porta a casa round da 2,7 milioni [https://startupitalia.eu/startup/commerceclarity-round-da-2-7-milioni-euro/](https://startupitalia.eu/startup/commerceclarity-round-da-2-7-milioni-euro/) Startup Italia 2025-11-13 ### Italian Founders Fund ed Entourage guidano un round da 2,7 mln in favore della tech company CommerceClarity [https://www.financecommunity.it/italian-founders-fund-ed-entourage-guidano-un-round-da-27-mln-in-favore-della-tech-company-commerceclarity/](https://www.financecommunity.it/italian-founders-fund-ed-entourage-guidano-un-round-da-27-mln-in-favore-della-tech-company-commerceclarity/) Finance Community 2025-11-13 ### Rivoluzione e-commerce: nasce Bazar, il primo commesso virtuale [https://www.liberoquotidiano.it/news/tv-news/28978777/rivoluzione-e-commerce-nasce-bazar-il-primo-commesso-virtuale/](https://www.liberoquotidiano.it/news/tv-news/28978777/rivoluzione-e-commerce-nasce-bazar-il-primo-commesso-virtuale/) Libero Quotidiano 2021-10-08 ### Rivoluzione e-commerce: nasce Bazar, il primo commesso virtuale [https://stream24.ilsole24ore.com/video/economia/rivoluzione-e-commerce-nasce-bazar-primo-commesso-virtuale/AE6adbo](https://stream24.ilsole24ore.com/video/economia/rivoluzione-e-commerce-nasce-bazar-primo-commesso-virtuale/AE6adbo) Il Sole 24 Ore 2021-10-08 ### Rivoluzione e-commerce, nasce il commesso virtuale: sara empatico e dara consigli al cliente [https://www.ilmessaggero.it/tecnologia/news/ecommerce_commesso_virtuale_bazar_novita_tecnologia-6395040.html](https://www.ilmessaggero.it/tecnologia/news/ecommerce_commesso_virtuale_bazar_novita_tecnologia-6395040.html) Il Messaggero 2021-12-20 ### Bazar e il primo commesso virtuale per acquisti on-line umanizzati [https://www.today.it/tech/bazar-commesso-virtuale-e-commerce.html](https://www.today.it/tech/bazar-commesso-virtuale-e-commerce.html) Today ### Un commesso virtuale dentro i negozi sostituisce quello umano [https://tech.meteoweek.com/2021/12/29/un-commesso-virtuale-dentro-i-negozi-sostituisce-quello-umano/](https://tech.meteoweek.com/2021/12/29/un-commesso-virtuale-dentro-i-negozi-sostituisce-quello-umano/) Tech Meteoweek 2021-12-29 ### Rivoluzione e-commerce, nasce il commesso virtuale: sarà empatico e darà consigli al cliente [https://www.ilmattino.it/tecnologia/hitech/ecommerce_commesso_virtuale_bazar_novita_tecnologia-6395061.html](https://www.ilmattino.it/tecnologia/hitech/ecommerce_commesso_virtuale_bazar_novita_tecnologia-6395061.html) Il Mattino 2021-12-20 ### CommerceClarity completes €2.7M funding to power the agentic era of e-commerce [https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/](https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/) Tech.eu 2025-11-13 ### Italy's e-commerce tech startup CommerceClarity raises €2.7M pre-seed round [https://www.vestbee.com/insights/articles/commerce-clarity-raises-2-7-m](https://www.vestbee.com/insights/articles/commerce-clarity-raises-2-7-m) Vestbee 2025-11-13 ### Why we invested in CommerceClarity [https://www.italianfoundersfund.com/posts/why-we-invested-in-commerceclarity](https://www.italianfoundersfund.com/posts/why-we-invested-in-commerceclarity) Italian Founders Fund ### CommerceClarity in the Koinos Capital portfolio [https://www.koinoscapital.com/it/portfolio/commerceclarity](https://www.koinoscapital.com/it/portfolio/commerceclarity) Koinos Capital ### CommerceClarity in the Redstone portfolio [https://redstone.vc/portfolio/commerceclarity](https://redstone.vc/portfolio/commerceclarity) Redstone ## Posts - [ChatGPT product feed: nine required fields, and the optional ones that decide what gets surfaced](/blog/chatgpt-product-feed-required-fields) · The feed fields · 2026-09-18T16:10:25Z - [Your ecommerce has an AI agent score.](/blog/llms-txt-webmcp-ecommerce) · The agent score · 2026-09-14T12:00:00Z - [Claude Commerce Agents: the agent got easy to build. The catalog decides what it can enforce.](/blog/claude-commerce-agents-catalog-constraint) · The catalog constraint · 2026-09-03T08:00:00Z - [Meet the Retail Engineer](/blog/meet-the-retail-engineer) · Retail Engineer · 2026-09-01T09:00:00Z ## In press - [CommerceClarity raises €2.7M ($3.1M) pre-seed to power the AI agentic era of e-commerce](/press/commerceclarity-raises-3m-ai-commerce) · 2025-11-13 ## Entities - [Amazon](https://www.amazon.com) · Organization, mentions - [Agentic commerce](https://en.wikipedia.org/wiki/Agentic_commerce) · Thing, mentions - [AI agent](https://en.wikipedia.org/wiki/AI_agent) · Thing, mentions ## Sources - [Why we invested in CommerceClarity](https://www.italianfoundersfund.com/posts/why-we-invested-in-commerceclarity) · Italian Founders Fund - [CommerceClarity](https://redstone.vc/portfolio/commerceclarity) · Redstone --- # Federico Sargenti > Federico Sargenti is co-founder and chief executive of CommerceClarity. He was chief executive of Everli from 2016 to 2023, and before that ran the FMCG categories at Amazon for Italy and Spain. He co-founded Witailer in 2017 and has been an Endeavor Entrepreneur since the same year. Federico is co-founder and CEO of CommerceClarity. He was CEO of Everli from 2016 to 2023, and before that ran the FMCG categories at Amazon for Italy and Spain. He co-founded Witailer in 2017 and has been an Endeavor Entrepreneur since the same year. He started out as a supply chain analyst. **Role:** Co-founder and CEO, CommerceClarity **Linkedin:** [https://www.linkedin.com/in/sargenti/](https://www.linkedin.com/in/sargenti/) **Given name:** Federico **Family name:** Sargenti **Location:** Milan, Italy ## Profiles - https://x.com/Fede_Sargenti - https://www.crunchbase.com/person/federico-sargenti ## Expertise - Agentic commerce - Catalog operations - Product data management - Marketplace management - Category management - Retail technology - Ecommerce operations - Supply chain management - P&L management - International expansion - Fundraising ## Languages - Italiano - English ## Awards - Endeavor Entrepreneur - FT 1000: Europe’s Fastest-Growing Companies ## Work history - CommerceClarity · Co-founder and CEO · [https://commerceclarity.com](https://commerceclarity.com) · 2024 - Witailer · Co-founder and investor · [https://witailer.com](https://witailer.com) · 2017 · 2024 - Everli · CEO · [https://everli.com](https://everli.com) · 2016 · 2023 - Amazon · Category Leader, FMCG, Italy and Spain · 2014 · 2016 - Amazon · Head of Toys and Baby · 2011 · 2014 - Ariston Thermo Group · Central Europe Logistics Manager · 2009 · 2011 - Ariston Thermo Group · Supply Chain Analyst · 2006 · 2009 - Prysmian · Logistics Analyst · 2006 · 2006 - HwManiac · Founder and CEO · 2000 · 2001 ## Education - Università Politecnica delle Marche · [https://www.univpm.it/](https://www.univpm.it/) - Stanford University Graduate School of Business · [https://www.gsb.stanford.edu/](https://www.gsb.stanford.edu/) ## Featured in ### CommerceClarity completes 2.7M funding to power the agentic era of e-commerce [https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/](https://tech.eu/2025/11/13/commerceclarity-completes-eur27m-funding-to-power-the-agentic-era-of-e-commerce/) Tech.eu 2025-11-13 ### CommerceClarity incassa round da 2,7 mln euro guidato da Italian Founders Fund e il fondo belga Entourage [https://bebeez.it/venture-capital/commerceclarity-incassa-round-da-27-mln-euro-guidato-da-italian-founders-fund-e-il-fondo-belga-entourage/](https://bebeez.it/venture-capital/commerceclarity-incassa-round-da-27-mln-euro-guidato-da-italian-founders-fund-e-il-fondo-belga-entourage/) BeBeez 2025-11-13 ### Startup in spotlight: CommerceClarity bags 2.7M to help retailers cut costs and boost sales with AI [https://techfundingnews.com/commerceclarity-raises-2-7m-ai-driven-ecommerce-catalog-management/](https://techfundingnews.com/commerceclarity-raises-2-7m-ai-driven-ecommerce-catalog-management/) Tech Funding News 2025-11-13 ### Italy's e-commerce tech startup CommerceClarity raises 2.7M pre-seed round [https://www.vestbee.com/insights/articles/commerce-clarity-raises-2-7-m](https://www.vestbee.com/insights/articles/commerce-clarity-raises-2-7-m) Vestbee 2025-11-13 ### Round da 2,7 milioni di euro per CommerceClarity [https://www.startupbusiness.it/round-da-27-milioni-di-euro-per-commerceclarity/148345/](https://www.startupbusiness.it/round-da-27-milioni-di-euro-per-commerceclarity/148345/) Startup Business 2025-11-14 ### CommerceClarity porta l'e-commerce nell'era dell'AI [https://www.efanews.eu/item/55264-commerceclarity-porta-l-e-commerce-nell-era-dell-ai.html](https://www.efanews.eu/item/55264-commerceclarity-porta-l-e-commerce-nell-era-dell-ai.html) EFA News 2025-11-13 ### CommerceClarity is Building the Intelligent Infrastructure for the Agentic Era of Commerce [https://www.menlotimes.com/post/commerceclarity-is-building-the-intelligent-infrastructure-for-the-agentic-era-of-commerce](https://www.menlotimes.com/post/commerceclarity-is-building-the-intelligent-infrastructure-for-the-agentic-era-of-commerce) Menlo Times 2025-11-13 ### Everli, cambio al vertice, Andrea Zocchi nuovo Ceo [https://www.mark-up.it/everli-cambio-al-vertice-andrea-zocchi-nuovo-ceo/](https://www.mark-up.it/everli-cambio-al-vertice-andrea-zocchi-nuovo-ceo/) Mark Up 2023-05-04 ### Italian grocery 'soonicorn' Everli being sold for 1 euro, according to leaked document [https://sifted.eu/articles/everli-sale-e1-news](https://sifted.eu/articles/everli-sale-e1-news) Sifted 2024-02-16 ### Supermercato24 diventa Everli, Federico Sargenti: 'Renderemo piu serena la vita delle persone' [https://www.think.it/supermercato24-diventa-everli-federico-sargenti-intervista/](https://www.think.it/supermercato24-diventa-everli-federico-sargenti-intervista/) Think Magazine 2020-07-07 ### Building the AI Agent Layer for E-Commerce: Interview with Federico Sargenti, Co-CEO of CommerceClarity [https://eqvista.com/interview-with-federico-sargenti-commerceclarity/](https://eqvista.com/interview-with-federico-sargenti-commerceclarity/) Eqvista 2026-06-04 ## In press - [We have acquired Katalogo.ai and welcome Luca Cozzolino as Founding Chief Product Officer](/press/commerceclarity-acquires-katalogo-ai) · 2026-06-04 - [CommerceClarity raises €2.7M ($3.1M) pre-seed to power the AI agentic era of e-commerce](/press/commerceclarity-raises-3m-ai-commerce) · 2025-11-13 ## Entities - [Amazon](https://www.amazon.com) · Organization, mentions - [Everli](https://www.everli.com) · Organization, mentions - [Agentic commerce](https://en.wikipedia.org/wiki/Agentic_commerce) · Thing, mentions ## Sources - [Why we invested in CommerceClarity](https://www.italianfoundersfund.com/posts/why-we-invested-in-commerceclarity) · Italian Founders Fund - [Everli, the European marketplace for online grocery shopping, bags $100M Series C](https://techcrunch.com/2021/03/29/everli/) · TechCrunch - [CommerceClarity](https://redstone.vc/portfolio/commerceclarity) · Redstone --- # Lara Bonalume > Lara Bonalume runs marketing at CommerceClarity: brand and communication, campaigns, the copy on the company's surfaces, SEO and press. Lara runs marketing at CommerceClarity: brand and communication, campaigns, the copy on the company’s surfaces, SEO and press. **Role:** Marketing, CommerceClarity **Linkedin:** [https://www.linkedin.com/in/larabonalume](https://www.linkedin.com/in/larabonalume) **Given name:** Lara **Family name:** Bonalume ## Expertise - Brand and communication - Marketing campaigns - Copywriting - SEO - Public relations ## Work history - CommerceClarity · Marketing · [https://commerceclarity.com](https://commerceclarity.com) --- # Lorenzo Riccò > Lorenzo Riccò is a Go to Market Engineer at CommerceClarity. He runs go to market execution, mostly outbound, and owns the sales process from discovery through to trial. The weekly pipeline review is his. Lorenzo is a Go to Market Engineer at CommerceClarity. He runs go to market execution, mostly outbound, and owns the sales process from discovery through to trial. The weekly pipeline review is his. **Role:** Go to Market Engineer, CommerceClarity **Linkedin:** [https://www.linkedin.com/in/lorenzo-ricco-cestudent/](https://www.linkedin.com/in/lorenzo-ricco-cestudent/) **Given name:** Lorenzo **Family name:** Riccò ## Expertise - Go to market strategy - Outbound - Sales process - Pipeline analytics ## Work history - CommerceClarity · Go to Market Engineer · [https://commerceclarity.com](https://commerceclarity.com) --- # Luca Cozzolino > Luca Cozzolino is co-founder and chief product officer of CommerceClarity. He joined in 2026, when CommerceClarity acquired Katalogo.ai, the Barcelona company he founded to build agents that read a fashion catalog and propose looks that sell together. He spent the three years before it at Shopify on the omnichannel surface inside Shopify Core, and at Zalando he built the order management system behind Connected Retail. Luca is co-founder and CPO of CommerceClarity, and owns product: what gets built, in what order, and when to stop. He joined in 2026, when CommerceClarity acquired Katalogo.ai, the Barcelona company he founded to build agents that read a fashion catalog and propose looks that sell together. He spent the three years before it at Shopify, where he shipped the omnichannel surface inside Shopify Core: click and collect, ship to store, pickup points, local delivery. At Zalando he built the order management system behind Connected Retail, which more than ten thousand European merchants use and which carries close to a tenth of Zalando's GMV. He was one of the first twenty people at Perk, where as founding product manager he started the ground transportation business and took it from zero to more than a hundred million in GMV. He was a merchant before any of it: Michele Bazar, the scarf ecommerce he co-founded in Pompei and ran as CEO, was bootstrapped past two million in sales and twenty thousand customers. **Role:** Co-founder and CPO, CommerceClarity **Linkedin:** [https://www.linkedin.com/in/lucacozzolino/](https://www.linkedin.com/in/lucacozzolino/) **Given name:** Luca **Family name:** Cozzolino **Location:** Barcelona, Spain ## Expertise - Agentic commerce - Catalog operations - Product strategy - Product discovery - Roadmap prioritisation - Omnichannel retail - Order management systems - Ecommerce platforms - Fashion merchandising - Ecommerce operations ## Languages - Italiano - English - Español ## Work history - CommerceClarity · Co-founder and CPO · [https://commerceclarity.com](https://commerceclarity.com) · 2026 - Katalogo.ai · Founder and CEO · 2025 · 2026 - Shopify · Senior Product Manager · [https://www.shopify.com](https://www.shopify.com) · 2022 · 2025 - Zalando · Senior Product Manager, Connected Retail · [https://www.zalando.com](https://www.zalando.com) · 2020 · 2022 - Michele Bazar · Co-founder and CEO · 2018 · 2022 - Perk · Founding Product Manager · 2017 · 2020 - Eodyne Systems · Researcher and Software Engineer · 2016 · 2017 ## Education - Universitat Pompeu Fabra · Master's degree, Cognitive Systems and Interactive Media · [https://www.upf.edu/](https://www.upf.edu/) · 2017 - Università degli Studi di Salerno · [https://www.unisa.it/](https://www.unisa.it/) ## Featured in ### CommerceClarity acquires Katalogo.ai [https://tech.eu/2026/06/04/commerceclarity-acquires-katalogoai/](https://tech.eu/2026/06/04/commerceclarity-acquires-katalogoai/) Tech.eu 2026-06-04 ### CommerceClarity acquires Katalogo.ai, creating synergy in AI-powered e-commerce [https://www.startupbusiness.it/en/commerceclarity-acquires-katalogo-ai-creating-synergy-in-ai-powered-e-commerce/174054/](https://www.startupbusiness.it/en/commerceclarity-acquires-katalogo-ai-creating-synergy-in-ai-powered-e-commerce/174054/) Startup Business 2026-06-04 ## In press - [We have acquired Katalogo.ai and welcome Luca Cozzolino as Founding Chief Product Officer](/press/commerceclarity-acquires-katalogo-ai) · 2026-06-04 ## Entities - [Katalogo.ai](https://katalogo.ai) · Organization, mentions - [Shopify](https://www.shopify.com) · Organization, mentions - [Zalando](https://corporate.zalando.com) · Organization, mentions - [Agentic commerce](https://en.wikipedia.org/wiki/Agentic_commerce) · Thing, mentions ## Sources - [Why we invested in CommerceClarity](https://www.italianfoundersfund.com/posts/why-we-invested-in-commerceclarity) · Italian Founders Fund --- # Matteo Codutti > Matteo Codutti is a Retail Engineer at CommerceClarity, on the delivery side. He works on moving product data in and out of customers' ERPs, and built the tooling the team uses for the feed conversions and the customer report. Matteo is a Retail Engineer at CommerceClarity, on the delivery side. He works on moving product data in and out of customers' ERPs, and he built the tooling the team uses for the feed conversions and the customer report. **Role:** Retail Engineer, CommerceClarity **Linkedin:** [https://www.linkedin.com/in/matteocodutti/](https://www.linkedin.com/in/matteocodutti/) **Given name:** Matteo **Family name:** Codutti ## Expertise - Retail engineering - ERP product feeds - Product data mapping - Customer reporting ## Work history - CommerceClarity · Retail Engineer · [https://commerceclarity.com](https://commerceclarity.com) ## Entities - [Product information management](https://en.wikipedia.org/wiki/Product_information_management) · Thing, mentions --- # Matteo Giuzzi > Matteo Giuzzi is the Founding Product Designer at CommerceClarity. He owns the design of the platform, from the data grid and workflow configuration to the review and evals surfaces, and the CommerceClarity design system, and writes about AI product design and how people review AI output. Matteo Giuzzi is the Founding Product Designer at CommerceClarity, where he leads the product experience and design system across data-heavy workflows, AI evaluation, review, and configuration. Previously a senior product designer at Meta, he has worked across complex products and design systems at scale. He writes about AI product design, human judgment, and how people understand, review, and act on AI output. **Role:** Founding Product Designer, CommerceClarity **Linkedin:** [https://www.linkedin.com/in/mgiuzzi/](https://www.linkedin.com/in/mgiuzzi/) **Given name:** Matteo **Family name:** Giuzzi ## Expertise - AI product design - Product design - Design systems - AI evals and review UX ## Work history - CommerceClarity · Founding Product Designer · [https://commerceclarity.com](https://commerceclarity.com) · 2026 --- # Michele Sampieri > Michele Sampieri is a co-founder of CommerceClarity and its product engineer, working on both what the product needs and what the platform has to do. He joined Wanderio as a founding engineer when the team was under six people and was head of mobile by the time he left almost six years later, through the pivots, the growth past twenty five million users and the exit. He is a full-stack engineer by background, with a master's in computer engineering from Roma Tre. Michele is a co-founder of CommerceClarity and its product engineer: he connects what the product needs to what the platform has to do, because he can work on both sides himself. He joined Wanderio as a founding engineer when the team was under six people, and was head of mobile by the time he left almost six years later. He was there for the pivots, for the growth from zero to more than twenty five million users, and for the exit. He co-founded Commodia before CommerceClarity. He is a full-stack engineer by background, with a master's in computer engineering from Roma Tre. **Role:** Co-founder and Product Engineer, CommerceClarity **Linkedin:** [https://www.linkedin.com/in/michele-sampieri-63332a124/](https://www.linkedin.com/in/michele-sampieri-63332a124/) **Given name:** Michele **Family name:** Sampieri **Location:** Milan, Italy ## Expertise - AI agents - Agentic commerce - Software engineering - Technical architecture - Full-stack development - Product engineering - Mobile engineering - Ecommerce technology - Automation ## Languages - Italian - English ## Awards - Winner, Bridgestone Open Innovation Hackathon ## Work history - CommerceClarity · Co-founder and Product Engineer · [https://commerceclarity.com](https://commerceclarity.com) · 2024 - Wanderio · Founding engineer, then Head of Mobile · [https://www.wanderio.com](https://www.wanderio.com) · 2015 · 2021 ## Education - Università degli Studi Roma Tre · Master’s degree, Computer Engineering · [https://www.uniroma3.it/](https://www.uniroma3.it/) - Università degli Studi Roma Tre · Bachelor’s degree, Computer Engineering · [https://www.uniroma3.it/](https://www.uniroma3.it/) ## In press - [CommerceClarity raises €2.7M ($3.1M) pre-seed to power the AI agentic era of e-commerce](/press/commerceclarity-raises-3m-ai-commerce) · 2025-11-13 ## Entities - [AI agent](https://en.wikipedia.org/wiki/AI_agent) · Thing, mentions - [Agentic commerce](https://en.wikipedia.org/wiki/Agentic_commerce) · Thing, mentions ## Sources - [Why we invested in CommerceClarity](https://www.italianfoundersfund.com/posts/why-we-invested-in-commerceclarity) · Italian Founders Fund - [Europcar investe in Wanderio, startup italiana del travel booking](https://www.economyup.it/fintech/europecar-investe-in-wanderio-startup-italiana-del-travel-booking/) · EconomyUp - [Europcar Group investe nella startup italiana Wanderio per sperimentare nuove modalita di trasporto multimodale](https://www.brand-news.it/brand/servizi/trasporti/europcar-group-investe-nella-startup-italiana-wanderio-sperimentare-nuove-modalita-trasporto-multimodale/) · Brand News - [CommerceClarity](https://redstone.vc/portfolio/commerceclarity) · Redstone --- # Nathan Salter > Nathan Salter is Sales Director at CommerceClarity. Nathan is Sales Director at CommerceClarity. **Role:** Sales Director, CommerceClarity **Linkedin:** [https://www.linkedin.com/in/nathan-salter-a2b63011/](https://www.linkedin.com/in/nathan-salter-a2b63011/) **Given name:** Nathan **Family name:** Salter ## Expertise - Enterprise sales - Deal qualification - Negotiation - Pipeline management ## Work history - CommerceClarity · Sales Director · [https://commerceclarity.com](https://commerceclarity.com) --- # Vittorio Rossi > Vittorio Rossi is an AI Engineer at CommerceClarity, where he builds the technology behind the AI agents that run on the Platform. He writes about what changes when agents own the product data pipeline. Vittorio is an AI Engineer at CommerceClarity, where he builds the technology behind the AI agents that run on the Platform. He writes about what changes when agents own the product data pipeline. **Role:** AI Engineer, CommerceClarity **Linkedin:** [https://www.linkedin.com/in/vitrossi/](https://www.linkedin.com/in/vitrossi/) **Given name:** Vittorio **Family name:** Rossi ## Expertise - AI agents - Agent evaluation - Workflow design - Product data pipelines ## Work history - CommerceClarity · AI Engineer · [https://commerceclarity.com](https://commerceclarity.com) ## Entities - [AI agent](https://en.wikipedia.org/wiki/AI_agent) · Thing, mentions - [Agentic commerce](https://en.wikipedia.org/wiki/Agentic_commerce) · Thing, mentions - [Product information management](https://en.wikipedia.org/wiki/Product_information_management) · Thing, mentions