Does ChatGPT recommend your store? Here's how to check.
Shoppers increasingly ask ChatGPT, Perplexity and Google AI "where should I buy X" instead of googling it. There's no rankings page to check anymore — just one answer. Here's a free, manual way to see if that answer includes you, in about 10 minutes.
1. Ask the question a real shopper would ask
Not your brand name — the question someone types when they don't know your store exists yet. Open ChatGPT (or Perplexity, or Google's AI Overview) and ask something like:
"What's the best place to buy [your product category] online?"
"Where can I buy [specific product] that ships to [your country]?"
"Best [category] shop with good reviews?"
Use the phrasing your customers would actually use, not your marketing copy.
2. Try 5–10 variations, not just one
A single answer can be noisy — AI models don't return the same result every time. Vary the wording, ask about specific products you sell, and ask about your category in general. Keep a simple tally: mentioned or not, and who else showed up.
3. Write down who actually gets named
This is the useful part. You're not just checking if you appear — you're mapping who AI trusts instead of you. If a specific competitor keeps coming up, that tells you where to look next: their reviews, their content, where they're mentioned online.
What actually influences the answer
AI assistants aren't ranking pages — they're synthesizing an answer from what's been written about you elsewhere: reviews on Google/Trustpilot, mentions on Reddit and forums, comparison articles, structured product data, and how clearly your own site states what you sell and who it's for. Thin or vague product pages give the model nothing to cite.
Two different doors, and most guides only mention one
It matters which one you're trying to get through, because the work is not the same:
1. The answer written in prose. "Where should I buy X?" gets answered from pages the model can read — your product page included, via browsing, Perplexity, Google's AI Overview, Copilot. This is where your own markup counts: what your page states in machine-readable form is what can be compared and cited. We measured 168 stores and found price and stock near-universal, but rating 14%, returns 9%, shipping 9% — the fields that decide between two similar products are mostly just missing.
2. The product carousel inside ChatGPT Shopping. That one does not come from your page. OpenAI takes it from a merchant product feed — delimited files delivered over SFTP, with required fields like item_id, availability, price, target_countries and the is_eligible_search / is_eligible_checkout flags. The spec doesn't mention schema.org at all, and rules out JSON, XML and RSS sources. If you want to be in that carousel, no amount of page markup substitutes for submitting the feed.
So: submit the feed for the shop surface, and fix your page data for everything else — the citations, the comparisons, and every assistant that isn't ChatGPT Shopping. The second half is the part almost nobody has done.
The catch with doing this manually
Asking the questions yourself works, but it doesn't scale: answers shift over time, there are three-plus engines worth checking, and one run tells you little. Being straight about what we do and don't do — AI Product Check doesn't run those prompts for you. It takes the other half: it opens a real product page on your store, reports which of the six fields an assistant compares are actually stated there, and hands you the JSON-LD for the missing ones, filled in from your own page.
Fix the half you control
Paste your store URL and see in seconds what your product page tells an assistant — and get the markup for what's missing, ready to paste. No email, no signup.
Run the free instant check →