By Steve Merrill, Founder of WRKNG Digital | September 23, 2026
An AI shopping assistant doesn't read your product page top to bottom like a person does. It reaches for a short list of facts first. If those facts are missing, it moves on.
I ran an ecommerce business to $10M a year. For years I obsessed over the words on the page. The machine that decides whether you get recommended never reads those words first. It reads the feed.
What does an AI assistant check before anything else?
Availability. Price. Title. Those three come first, in that order.
An assistant is trying to answer one question: can I recommend this product right now without being wrong? If it can't confirm the item is in stock and priced, it drops you rather than risk a bad answer. Google's Merchant Center spec treats those same fields as required for exactly that reason.
Why is availability the first field, not price?
Because a wrong recommendation costs the assistant trust. Recommending something out of stock is the worst outcome, so availability gets checked first.
Here's the bottom line: a product marked in stock in clean, machine-readable form beats a better product with blank availability. Every time.
Which fields decide the tie between two similar products?
GTIN or MPN. Brand. Shipping and returns. Those break the tie.
When two stores sell the same thing, the assistant picks the one whose data answered more questions. A missing GTIN means the assistant can't match your item to the wider catalog it already trusts. That gap hands the recommendation to your competitor.
What about the fields humans care about?
Long descriptions, lifestyle photos, review counts. They matter, but later. They shape the pitch, not the shortlist.
You get on the shortlist with clean structured facts. You win the shortlist with the human stuff. Skip the first step and the second never happens. Get your Product structured data right before you touch the copy.
How do you check your own feed this week?
Pull your ten best sellers. Run each URL through Google's Rich Results Test. Write down every field that comes back blank or wrong: availability, price, GTIN, brand.
Then fix the theme or feed app that generates those fields so they're complete and present in the raw HTML, not painted on by JavaScript after load. I ignored a platform shift once and it cost me years. This one is a cheaper lesson if you take it now.
Further reading
- Google Merchant Center product data spec
- Google product structured data guidelines
- Schema.org Offer reference
Frequently Asked Questions
Which Shopify feed fields do AI shopping assistants check first?
Availability, price, and title come first, in that order. An assistant needs to confirm a product is in stock and priced before it will recommend it, so those fields gate everything else. GTIN, brand, and shipping details are checked next to break ties between similar products.
Does a missing GTIN really hurt AI visibility?
Yes. The GTIN lets an assistant match your product to the broader catalog it already trusts. Without it, the assistant can't confirm what the item is, so when it's choosing between two similar products it favors the one with a complete identifier. Add GTIN or MPN to every product where one exists.
Why do my products show as out of stock to AI when they're available?
Usually because availability is rendered by JavaScript that loads after the page, and the assistant reads the raw HTML before that runs. The fix is making sure availability is present in the initial HTML and in your product feed, not added on the client side after load.
Want to know if AI assistants can actually find and recommend your store? Get a free AI-visibility read on your Shopify store at WRKNG Digital. We show you exactly what ChatGPT, Perplexity, and Google AI see when they look at your products.
