By Steve Merrill, Founder of WRKNG Digital | September 15, 2026
Before ChatGPT recommends your product, it checks whether it can understand you, verify you, and trust you enough to put its name behind the pick. Here are the eight things it looks at.
None of them are about having the best product. They're about being the clearest answer.
How does ChatGPT decide what to recommend?
It looks for the product it can match to the shopper's question and confirm fast. Clear data and verifiable offers beat vague listings, in line with how ChatGPT shopping surfaces results.
1. Does your title match the query?
The title is the first thing it reads to see if you're relevant. "Merino Base Layer, Men's, 200gsm" gets matched to a real search. "Item 4471" gets skipped. The model can only recommend what it understands.
2. Is your structured data valid?
Valid Product and Offer markup gives ChatGPT a confirmed record of what you sell and for how much. Broken or missing markup forces it to guess, and it prefers stores it doesn't have to guess about.
3. Can it verify your price?
A current, clearly stated price keeps you eligible. A missing or stale one reads as risk, and the model avoids recommending products it can't confirm are buyable at a real number.
4. Is the item actually in stock?
Availability that matches your live inventory matters as much as price. An out-of-stock flag or an unclear status drops you fast, because a bad recommendation costs the assistant credibility.
5. Does the page answer the question directly?
Pages that lead with a plain answer get quoted. When the first two sentences say what the product is and who it's for, the model has something clean to lift into its response.
6. Are your details specific?
Material, dimensions, compatibility, use case. Specific facts give the model reasons to choose you over a vaguer rival. Generic copy gives it nothing to work with.
7. Are there trust signals?
Reviews, ratings, and clear brand information act as confidence markers. A store that surfaces credible trust signals is a safer pick than one that hides them.
8. Can crawlers reach the page?
If AI crawlers can't reach your product pages, nothing above counts. Confirm your robots rules let the right bots in, or you're invisible before the check even runs.
Further reading
- OpenAI on shopping in ChatGPT
- Schema.org Product reference
- Schema.org Offer reference
- Google on robots.txt and crawlers
Frequently Asked Questions
What does ChatGPT check before recommending a product?
It checks whether it can match your title to the query, verify your structured data, confirm your price and stock, lift a direct answer from your page, find specific details, see trust signals, and reach the page at all. Miss several and you're filtered out.
Why does ChatGPT recommend a competitor over me?
Usually because the competitor is easier to understand and verify, not because their product is better. Clearer titles, valid structured data, and confirmable prices make a store a safer recommendation.
Can I pay ChatGPT to recommend my store?
Not for organic recommendations. The model surfaces stores it can read and trust from public product data. The lever you control is how readable and verifiable you make your store.
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.
