If ChatGPT, Perplexity, or other AI shopping tools keep skipping over your products (or describing them incorrectly when they do surface), the root cause is usually not a missing “AI strategy.” It’s incomplete or inconsistent product data. AI shopping systems can only recommend what they can accurately understand, and a vague or messy product catalog gives them very little to work with.
Vague Product Data
Titles like “The Essential One,” missing variant details, inconsistent naming, no clear availability or pricing signals.
Structured Product Data
Clear titles, accurate descriptions, current pricing and availability, distinct variants, and consistent naming an AI system can parse with confidence.
Why AI shopping tools care about your product feed
Traditional SEO rewarded content that ranked well for keywords. AI shopping assistants work differently: they need to confidently answer a specific question (“does this exist, in this size, in stock, at this price”) before they’ll recommend or transact on a product at all. When product information is ambiguous, the system has to guess, and guessing is not something AI shopping tools are built to do well, or reward.
OpenAI has published guidance for merchants on structuring product feed data (covering fields like pricing, availability, variants, and seller information) to support shopping features inside ChatGPT. Other platforms building AI shopping and agentic checkout experiences are moving in a similar direction.
Worth verifying before you act on this: the specific fields, formats, and requirements in AI shopping product feed specifications are changing quickly, and different platforms (OpenAI, Google, Perplexity, and others) are not all aligned on one standard. Before rebuilding your product feed around any single spec, check the platform’s current published documentation, since details published even a few months ago may already be out of date.
A practical audit: pick your top 10 products and check this
You don’t need a full catalog overhaul to start. Andreea’s approach with clients is always to start small and prove the model before scaling it, so pick your 10 best-selling or highest-margin products and run this checklist against each one.
1. Clear title
Does the product name describe what it actually is, not just your brand’s internal name for it?
2. Useful description
Does the description explain what the product is, who it’s for, and what problem it solves?
3. Correct availability
Is in-stock/out-of-stock status accurate and updated in something close to real time?
4. Current pricing
Does the listed price match what a customer would actually pay right now, including active sales?
5. Distinct variants
Are size, color, and other variants each clearly labeled, rather than bundled into one ambiguous listing?
6. Consistent naming
Does the same product use the same name and attributes everywhere it appears (site, feed, marketplace)?
Frequently asked questions
Is this the same thing as SEO?
It’s related but distinct. SEO is about being found in search results; AI shopping product data is about being understood and trusted enough for an AI system to recommend or facilitate a purchase. Clean, structured product information supports both.
Do I need special “AI schema” markup to fix this?
Not necessarily. The underlying issue this audit addresses is usually the accuracy and clarity of your core product data itself (titles, descriptions, pricing, availability, variants), not a missing technical markup layer. We’ve written more on the schema question specifically in our post on the AI schema myth.
Where should I start if I have hundreds or thousands of SKUs?
Start with your top 10 by revenue or margin, as outlined above. Fixing your highest-impact products first gives you a fast, low-risk way to see whether cleaner data actually changes how those products show up in AI shopping results, before you invest in a catalog-wide project. If you’d rather not audit by hand, our free Store Readability Audit runs a version of this same check against your public store automatically.
Related reading
This audit pairs well with two other pieces we’ve written on AI shopping visibility: how to check if your Shopify store is showing up in ChatGPT, and why your product pages might be invisible to AI shopping assistants.
Free through August 31, 2026. No card required.
Or talk to our team about implementing what you find.
