By Steve Merrill, Founder of WRKNG Digital | September 23, 2026
When ChatGPT recommends a product, it's not pulling a favorite out of a hat. It's assembling an answer from sources it can read and trust, and dropping everything it can't.
If your store isn't showing up, it's rarely because your product is worse. It's because ChatGPT couldn't confirm enough about it to feel safe recommending it.
Where does ChatGPT get its product information?
From structured product data, merchant feeds, and content it can read on the open web. OpenAI's own explanation of shopping results points to structured metadata like price, availability, and reviews as the backbone.
It stitches those signals into a recommendation. The cleaner and more complete your signals, the easier you are to include.
What makes a product easy for ChatGPT to recommend?
Facts it can confirm without interpreting. A clear price. In-stock status. A real brand. Specs written as text, not baked into a graphic.
Here's the bottom line: ChatGPT recommends what it can verify. Ambiguity is a reason to skip you, because a confident wrong answer is the thing it's built to avoid.
Why does it recommend a competitor instead of me?
Usually because the competitor made the same information machine-readable and you didn't. Their specs are text. Their price is in the HTML. Their availability is current.
Yours might be identical to a human eye and invisible to the model. A spec sheet saved as an image is a wall to ChatGPT. It reads the store that wrote it out.
Does having reviews help?
Yes, when they're structured. Review and rating markup gives ChatGPT a trust signal it can quote. A pile of reviews with no markup is far weaker.
Reviews also feed the language people use, which helps the model match your product to real questions. But the markup is what turns them into a citable fact instead of decoration.
How do you get recommended more often?
Make your best sellers fully readable. Price, availability, brand, and specs as structured text. Add review markup. Write a short, direct answer to the questions shoppers actually ask.
Then test it. Ask ChatGPT to recommend products in your category and watch where you land. I refused to look at a shifting platform once and paid for it in lost years. Looking is free. Do it this week.
Further reading
- OpenAI on shopping in ChatGPT
- Google product structured data guidelines
- Schema.org Product reference
Frequently Asked Questions
How does ChatGPT pick which products to recommend?
It assembles recommendations from structured product data, merchant feeds, and readable web content, favoring products whose price, availability, brand, and specs it can confirm. It skips products it can't verify because a confident wrong answer damages trust, so ambiguity works against you.
Why does ChatGPT recommend my competitor instead of me?
Usually because they made their product facts machine-readable and you didn't. If your specs are in an image or your price loads by JavaScript, ChatGPT reads a blank and cites the store that wrote the same information as plain text. The products can look identical to a human and only one is readable to the model.
Do product reviews help ChatGPT recommend me?
Yes, especially when they're marked up with review or rating schema, which turns them into a trust signal ChatGPT can quote. Reviews without structured markup are much weaker, working more as human decoration than a citable fact.
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.
