Merchant Center

Merchant Center is Google's free platform where retailers and brands submit product data, store information, shipping and return policies, so their products can appear across Google: the Shopping tab, free listings and Shopping ads, Google Images and Lens, YouTube, Maps (for local inventory), and the product answers in AI Mode and Gemini. Data flows in through feeds, supplemental data sources, the Merchant API or platform integrations such as Shopify.

That data becomes part of Google's Shopping Graph, which Google said at I/O in May 2026 contains more than 60 billion product listings. When someone asks AI Mode for 'a quiet dishwasher under $800 that fits a 24-inch space', the answer is assembled from Shopping Graph attributes such as dimensions, price, availability, ratings and return policy. A missing or vague attribute can keep an otherwise strong product out of the answer.

Merchant Center gained several AI-era features in 2026:

• Conversational attributes: optional fields designed for AI Mode and other conversational surfaces. Google's help page lists question_and_answer, document_link, related_product, item_group_title, variant_option and popularity_rank. They can be added through a supplemental data source without affecting product approval.
• The UCP integration hub, which lets eligible US merchants choose capabilities of Google's Universal Commerce Protocol, its open standard for agentic commerce, such as native checkout on Google, cart transfer to their own site and identity linking. Google says Australia and Canada follow next year.
• AI performance insights, announced at Google Marketing Live on May 20, 2026, which compares a brand's share of voice on AI surfaces against similar brands. It is rolling out in Australia, Canada, India, New Zealand and the US.
• Business Agent, a branded shopping agent retailers can activate and customize in Merchant Center, and Ask Advisor, an assistant inside Merchant Center itself.

The fundamentals still decide most outcomes. Product identifiers (GTIN, brand, MPN) let Google match your offer to the right product. Price and availability must match the landing page, or products get disapproved. Titles and descriptions should name the attributes shoppers ask about. Images need to be clean and accurate. Shipping and return settings feed the delivery and returns information that AI answers mention. And on-page Product schema markup should agree with the feed, since Google reads both.

Merchant Center data also travels beyond Google. OpenAI's product feed specification accepts a Google-compatible feed format for merchants approved to share data with ChatGPT, and Microsoft says product feeds in Microsoft Merchant Center help inform organic Copilot results. A clean Google feed is therefore a reasonable base for product feeds for AI shopping in general, though each platform has its own fields and rules.

For AI shopping visibility, measure in three places: Merchant Center's own performance and diagnostics reports (including AI performance insights where available), the share of AI Mode and Gemini answers for your category prompts that show your products, and whether those answers state your current price, stock and return terms correctly. Errors in AI answers usually trace back to a feed attribute or a mismatch between feed and page.