Knowledge base v2.13.0 · Community edition · data through 2 October 2026
Day 3 · Friday 2 October 2026 · 13:40
Speaker matched from his self-introduction to the official speaker list. No slides were photographed.
Google's Merchant Center feed specification is how millions of merchants send Google billions of product offers, which are used in Shopping ads, Search ads and organic search results.
Speaker Alex JansenEvidence transcript
Used byglossary term Merchant Center
Schema.org product markup on websites is another source of billions of product offers for Google Shopping, next to merchant feeds.
Speaker Alex JansenEvidence transcript
Product snippets in Google Search are powered by both shopping feeds and schema.org structured data.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-01
Traditional shopping results use a small set of well-defined attributes because they take little screen space: title, image, price, perhaps star ratings, shipping cost and time, and return information; in the past that was sufficient.
Speaker Alex JansenEvidence transcript
Behind the visible attributes, small pieces of product data such as brand and UPC or other barcodes are very important for Google Shopping's query understanding, retrieval and ranking.
Speaker Alex JansenEvidence transcript
In AI Mode, shoppers keep asking deeper, more nuanced follow-up questions about brands and products, so AI Mode must return much more detailed product information than classic shopping results needed.
Speaker Alex JansenEvidence transcript
Product data quality matters more than ever for AI ('garbage in, garbage out'), and even more for agents acting on behalf of consumers, because an agent with bad data cannot do the right thing.
“If the agent has bad data, it cannot do the right things.”
Speaker Alex JansenEvidence transcript
In 2026, a few months before the October event, Google added new Merchant Center feed attributes to help AI systems understand product nuances, prompted by feedback since 2025 from merchants worried whether AI experiences represented them correctly.
Speaker Alex JansenEvidence transcript
Google announced the new conversational feed attributes at Google Marketing Live (GML) 2026; the recording also appears to name Google I/O.
Speaker Alex JansenEvidence transcript
Google decided against adding hundreds of thousands of highly specific feed attributes, such as heel height for shoes or lens details for cameras, and instead added a few flexible ones that leave merchants in control.
Speaker Alex JansenEvidence transcript
Google Shopping said highly prescriptive product data is a lot of hard work for merchants and is no longer needed for LLMs.
Speaker Alex JansenEvidence transcript
Google added six Merchant Center feed attributes for AI shopping experiences: question and answer, documents, related products, item group title and variant option for variants, and popularity rank.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-03glossary term Conversational attributes
The question and answer feed attribute lets a merchant or brand upload questions and answers relevant to a product, whether they come from the manufacturer or from consumers.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-03
Merchants can now provide a document describing a product, such as a PDF manual, a spec sheet or the package contents, so that AI can work out what the product is about.
Speaker Alex JansenEvidence transcript
The related products feed attribute describes accessories, parts and replacement products, answering frequent questions such as what else to buy with a printer.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-03
Variant data in Google's feed specification used to be limited to basics such as colour and size; the new item group title and variant option attributes make it much more flexible.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-03
Google added a popularity rank feed attribute because consumers want to know how well a product sells on a marketplace.
Speaker Alex JansenEvidence transcript
Product Q&A data lets AI answer very detailed questions, such as a kettle's temperature settings, capacity or keep-warm safety, that separate feed attributes could never represent; Google's example came from a real merchant's product.
Speaker Alex JansenEvidence transcript
Google Shopping called rich data with light structure the AI sweet spot: a little structure that helps the AI system understand the data, not deep, highly nested, complex structures.
Speaker Alex JansenEvidence transcript
Some structure is much better than none for AI systems, because without structure the model does not really know what is going on, Google Shopping said.
Speaker Alex JansenEvidence transcript
Most of the new conversational feed attributes were already available in schema.org, so Google ties them back to structured data and can use them from product markup as well as from feeds.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-04
Basic product markup has two parts: product data that is the same for every merchant (name, description, brand, images) and offer data specific to one merchant's offer (URL, condition, availability, price).
Speaker Alex JansenEvidence transcript
In product markup, an offer's shipping and return information can point through a JSON-LD identifier (@id) to shipping and return data defined elsewhere.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SDA-09
Google Shopping uses its own crawler, Storebot (Storebot-Google), instead of Googlebot because it needs fresh product information, prices, availability and shipping details and therefore crawls much more often.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-02glossary term Storebot-Google
Storebot-Google is used to validate merchant feeds: when a feed gives a price, Google double-checks it on the merchant's site.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-02
Storebot-Google also does deep crawls, for example of checkout pages, which do not belong in the search index; this is another reason it is kept separate from Googlebot.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-02
Product Q&A can be marked up in schema.org by linking Question items, each with an acceptedAnswer of type Answer, to the product through the subjectOf property, because the product is the subject of the Q&A.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-04
Google now emphasises two existing Merchant Center attributes for AI, product highlights (a short bulleted list) and product detail (detailed specifications), and updated its Help Center guidance to say they help AI systems.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-03glossary term Product highlights and product details
Instead of adding feed attributes for individual specifications such as connectivity, memory, camera settings or box contents, Google asks merchants to send them as key-value pairs in product detail, because merchants know best what matters.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-03
Google works to make sure that what merchants can express in its shopping feeds can also be expressed in schema.org, adding vocabulary where schema.org lacks it, for example for product details.
Speaker Alex JansenEvidence transcript
Google named the Universal Commerce Protocol (UCP), next to the Merchant Center feed, as an industry standard whose capabilities it also wants schema.org to cover.
Speaker Alex JansenEvidence transcript
Used byglossary term Universal Commerce Protocol (UCP)
A recent schema.org release added two properties, specification and valueGroup, so that the Merchant Center product detail attribute can be expressed in schema.org, with product specs nested under specification.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-04
Rich, lightly structured information is critical to AI systems: product data with many levels of nesting is not needed, but some structure is very important.
“rich, lightly structured information is critical to AI systems.”
Speaker Alex JansenEvidence transcript
Basic, well-specified product attributes such as price, brand and availability remain as important as ever for Google Shopping.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-01
Rich, less structured product data that is not prescribed by Google, schema.org or any feed or data model is increasingly important, because the merchant or brand knows best what to provide.
Speaker Alex JansenEvidence transcript
The bar for product data quality is higher than ever because agents and users need correct information: with wrong data an agent might buy the wrong product for a user.
Speaker Alex JansenEvidence transcript
Web markup is an efficient and unambiguous way for sites to share product data with Google, Google Shopping said, repeating the Day 2 structured data talk.
Speaker Alex JansenEvidence transcript
Google Shopping named the downside of adding Q&As and documentation to product markup: the page's markup becomes somewhat bigger.
Speaker Alex JansenEvidence transcript
Used byrequirement DEV-SHP-04
Google Shopping pointed to a Google holiday shopping blog post with early results from real merchants using the conversational attributes, and to the Merchant Center Help Center and schema.org for the attributes and properties.
Speaker Alex JansenEvidence transcript
Google Shopping described product data for AI as early days in a fast-moving area, with many more changes to come.
Speaker Alex JansenEvidence transcript
On 11 January 2026 Google announced 'dozens of new data attributes' in Merchant Center for conversational commerce on AI Mode, Gemini and Business Agent, to roll out first with a small group of retailers.
Publisher Google blog (11 January 2026)
At Google Marketing Live on 20 May 2026 Google said retailers globally can now use conversational attributes in Merchant Center to reflect the more conversational way people search.
Publisher Google blog (20 May 2026)
Used byglossary term Conversational attributes
Google's Merchant Center help marks the question_and_answer, document_link, related_product and popularity_rank attributes as primarily intended for conversational experiences such as AI Mode in Google Search.
Publisher Google Merchant Center Help
Used byrequirement DEV-SHP-03glossary term Conversational attributes
Merchant Center's question_and_answer attribute takes up to 30 question-and-answer pairs per product, each question and answer up to 1,000 characters and 10,000 characters in total, and must not contain prices, shipping, dates or the company name.
Publisher Google Merchant Center Help
Used byrequirement DEV-SHP-03
Merchant Center's document_link attribute takes up to five URLs of PDF documents about a product, such as manuals, user guides or assembly instructions, and the merchant must own the content licensing rights.
Publisher Google Merchant Center Help
Used byrequirement DEV-SHP-03
Merchant Center's popularity_rank is a 0-100 value the merchant assigns to rank a product's popularity, based on its recent sales, against the rest of its own inventory; it does not reflect user ratings.
Publisher Google Merchant Center Help
Used byrequirement DEV-SHP-03
Google's merchant listing documentation recommends defining shipping and return policies once under Organization markup and referencing them from an Offer, even on another page, using only the @id keyword (through hasShippingService for shipping, hasMerchantReturnPolicy for returns).
Publisher Google Search Central
Used byrequirement DEV-SDA-09
Google's Merchant Center help says the StoreBot crawler goes through product detail, cart and checkout pages, can fill in checkout forms, and records price, shipping, availability, coupons and payment methods to verify the data merchants share in Merchant Center.
Publisher Google Merchant Center Help
Used byrequirement DEV-SHP-02glossary term Storebot-Google
Google's crawler list says robots.txt rules addressed to the Storebot-Google user agent affect all surfaces of Google Shopping, such as the Shopping tab in Google Search.
Publisher Google
Used byrequirement DEV-SHP-02glossary term Storebot-Google
Merchant Center's product_highlight attribute takes 2 to 100 highlights of up to 150 characters each that describe only the product itself, without promotional text, keywords or search terms.
Publisher Google Merchant Center Help
Used byrequirement DEV-SHP-03glossary term Product highlights and product details
Merchant Center's product_detail attribute takes up to 100 specifications, each a section name, attribute name and attribute value, and Google's help says clean key-value specifications improve product details on Google Shopping and AI-driven surfaces.
Publisher Google Merchant Center Help
Used byrequirement DEV-SHP-03glossary term Product highlights and product details
Google launched the Universal Commerce Protocol (UCP) on 11 January 2026 as an open standard for agentic commerce across discovery, buying and post-purchase support, co-developed with retailers and platforms and compatible with A2A, AP2 and MCP.
Publisher Google blog (11 January 2026)
Used byglossary term Universal Commerce Protocol (UCP)
Google's 16 September 2026 holiday shopping post says that in testing with lululemon, conversational attributes submitted by the brand were used 50% of the time in relevant product recommendations in AI Mode.
Publisher Google blog (16 September 2026)
Used byrequirement DEV-SHP-03
Google's January 2026 announcement spoke of 'dozens' of new Merchant Center attributes, while the Day 3 talk counted six; all six named on stage (question_and_answer, document_link, related_product, item_group_title, variant_option, popularity_rank) are in the Merchant Center product data specification as of 3 October 2026.
Author Ibrahim Anjro
On stage the popularity rank was framed as how well a product sells on a marketplace, but Google's documentation defines it as the merchant's own 0-100 ranking against the rest of its inventory: a self-reported, relative value, not a sales figure across shops.
Author Ibrahim Anjro
As of 3 October 2026 Google's Merchant Center help lists no schema.org property for question_and_answer, document_link, related_product, popularity_rank, variant_option or product_detail (only item_group_title maps, to ProductGroup.name), and Search Central does not document such markup: reading these attributes from schema.org is a stage statement, the feed is the documented route.
Author Ibrahim Anjro
Used byrequirement DEV-SHP-04
Treat Storebot-Google separately from Googlebot in robots.txt and bot protection: a shop that blocks it on product, cart or checkout paths can undermine Shopping price and availability checks, even when Googlebot is allowed.
Author Ibrahim Anjro
Used byrequirement DEV-SHP-02
Schema.org's release notes show version 30.1 (16 September 2026) added specification, isOftenBoughtWith and consumerNotice for Product, valueGroup for PropertyValue and itemPopularity for Offer as vocabulary for retail feed data; Google's Search Central docs did not yet document these properties on 3 October 2026.
Author Ibrahim Anjro
Used byrequirement DEV-SHP-04
Shop developers should first keep core feed and markup values (price, availability, brand, GTIN) accurate, then add Q&As, product highlights, specs as key-value product details and related products in the Merchant Center feed, and mirror them in schema.org Product markup only as Google documents support for it.
Author Ibrahim Anjro
Content teams can feed the new Q&A attribute from real customer questions (support tickets, reviews, sales chats), answered factually per product; this is the detailed, merchant-owned data Google said AI shopping answers need.
Author Ibrahim Anjro
Google added six Merchant Center feed attributes for AI shopping experiences: question and answer, documents, related products, item group title and variant option for variants, and popularity rank.
More shopping structured data news was left for a Day 3 talk by a Google colleague, Alex.
Google Shopping called rich data with light structure the AI sweet spot: a little structure that helps the AI system understand the data, not deep, highly nested, complex structures.
Describing every semantic detail of a page in markup is probably not worth the effort; focus on the structured data that Google or other consumers actually use.
In product markup, an offer's shipping and return information can point through a JSON-LD identifier (@id) to shipping and return data defined elsewhere.
Google's shopping structured data launches of the previous year (2025) added support for merchant loyalty programs and shipping policies, letting merchants define a policy at organisation level and specify details at product level.
Google Shopping uses its own crawler, Storebot (Storebot-Google), instead of Googlebot because it needs fresh product information, prices, availability and shipping details and therefore crawls much more often.
The scheduler is shared infrastructure that decides what to fetch and when and sends URLs to the crawler. Each team decides the scheduling parameters for its own user agents.
Google works to make sure that what merchants can express in its shopping feeds can also be expressed in schema.org, adding vocabulary where schema.org lacks it, for example for product details.
Google's structured data speaker said that in the months before the event Google added support for validity dates on sale prices in product structured data, so merchants no longer need to rush to remove a sale price when the sale ends for fear it shows wrongly in snippets.
Web markup is an efficient and unambiguous way for sites to share product data with Google, Google Shopping said, repeating the Day 2 structured data talk.
Parsing structured data is significantly cheaper and more efficient than relying on LLMs for every complex extraction task.