Google slipped something new into Merchant Center recently, and they did not even mention it at Google Marketing Live. It is called conversational attributes, and if you sell products online, this is the kind of quiet update that separates the brands who win on AI search from the ones who get skipped entirely.
Here is what you need to know, and what to do about it.
The Shift Nobody Is Talking About
For years, a "good" product feed meant clean titles, the right category, a GTIN, and decent images. That was enough to win on Google Shopping.
It is not anymore.
People are not typing "mens hiking boots" into search. They are typing "best waterproof hiking boots for wide feet with good ankle support for muddy trails under $150."
AI Mode, Gemini, AI Overviews, and the agentic commerce tools Google is building all need structured data to confidently match, compare, and recommend your products. If your feed does not hand the AI that data on a silver platter, it either guesses or skips you entirely.
Conversational attributes are how you feed the machine directly.
The 5 New Fields You Can Use Right Now
- question_and_answer (pre-answered buyer questions, up to 30 per product)
- document_link (PDF manuals or spec sheets Google can pull FAQs from)
- related_product (tag accessories, substitutes, often-bought-with items)
- item_group_title plus variant_option (much cleaner variant grouping)
- popularity_rank (a 0 to 100 score based on your actual internal sales data)
These are optional. They do not affect approval. There is zero risk.
Why This Is a Real Opportunity
Most brands ignore anything that is not actively on fire, which means there is a wide-open gap right now for whoever moves first.
This data powers every shopping surface (free listings, standard Shopping), but the biggest lift is in AI Max for Shopping and Performance Max. Those campaigns are built to use this kind of semantic depth. Google's early data on AI Max is showing roughly +5 percent conversions at similar CPA.
And here is the real prize: AI surfaces are brand new placements. The brands enriching their feeds now become the ones the AI trusts and recommends, while everyone else still treats Merchant Center like a dumb ad data dump.
Your Action Plan
Start with bestsellers. Popularity_rank tells Google which products to push on AI surfaces, so stack your richest data on the products most likely to surface anyway.
Use real buyer questions for Q&A. Pull from your support inbox, reviews, and sales calls. Specs, compatibility, sizing, usage. Do not recycle marketing copy, that is wasted space.
Build it into your store backend. If you are on Shopify or similar, bake these fields into your product setup so every new product carries this data automatically. Bonus: when this data lives on your website, Google can pull it for organic AI ranking too.
Tag relationships properly. For related_product, use the right type (accessory, substitute, often_bought_with) and reference by internal ID or GTIN.
The Window Is Open. It Will Not Stay That Way.
This is exactly the kind of move that defines AEO (Answer Engine Optimization) in 2026: building an AI-readable knowledge base around your products so the machine recommends you with confidence.
The brands setting this up this quarter will look like geniuses six months from now. The rest will be playing catch-up.
If you want help auditing your product feed or building this into your store backend the right way, that is what we do at BlueShore.AI. Start with our free AEO Readiness Score, learn more about our Answer Engine Optimization engagements, or contact BlueShore.AI and let us talk.