By Steve Merrill, Founder of WRKNG Digital — October 5, 2026
This is the Shopify product data checklist for AI search, updated for 2026. Work through these eight checks and your store gives ChatGPT, Perplexity, and Google AI Overviews clean data to recommend. It builds on our breakdown of the product feed fields that determine if AI recommends your store and turns it into an audit you can run this week.
Here's the bottom line. You can't fix what you don't check. Run the list, mark each item pass or fail, then fix the fails in order.
1. Every product has a unique identifier
Confirm a GTIN or MPN on every product. Google's product data spec treats these as core. No identifier, no confident match.
2. Titles are written for shoppers
Scan your titles. If they read like internal SKUs, rewrite them with the feature and use case first. This one change lifts matches more than any other.
3. Descriptions are 40 to 200 words
Check length on your top sellers. Thin descriptions give the assistant nothing to quote or match. Expand them with specific, plain detail.
4. Attribute fields are filled
Color, size, material, fit, use case. Confirm these are populated, not blank. They are the filters inside real shopping queries.
5. Categories map correctly
Verify each product sits in the right Google product category. The wrong category hides you from the right comparison set.
6. Inventory and price are synced
Check that feed availability and price match your live store. Drift gets listings suppressed and breaks trust with the assistant.
7. Product schema is on every page
Confirm live pages carry Product schema that matches the feed. Run a sample through the Rich Results Test to verify it parses.
8. Reviews are marked up
If you have reviews, confirm they use AggregateRating schema so AI can read them as a trust signal, not just show them to humans.
How to run this checklist
Pull your 20 top-revenue products and score each against the eight checks. Count the fails. Most stores find the same three or four gaps repeated across the catalog, which means one fix pattern clears dozens of products at once. Start with identifiers and titles, then work down.
Frequently Asked Questions
How long does this product data checklist take to run?
Score your top 20 products against the eight checks in about an hour. Most stores find the same few gaps repeated, so one fix pattern clears many products at once.
What should I fix first?
Unique identifiers and shopper-style titles. Those two earn the most product matches for the least effort.
Does product schema need to match my feed?
Yes. The feed and the live page should carry the same clean data so an assistant gets consistent signals whichever one it reads.
Ready to see how AI shopping assistants view your store?
Most Shopify stores are invisible to AI right now. I can show you exactly where yours stands. See how your store shows up in agentic commerce.
