ChatGPT shopping: what Peec’s regional tests show

Peec AI reports different product retrieval behaviour in US and European ChatGPT shopping tests. That is a useful reason to test your store’s visibility by market, but it is not an OpenAI announcement or proof that every shopper in a region sees the same system.

For Shopify merchants, there is a separate, confirmed starting point: OpenAI says Shopify Catalog already integrates product data into ChatGPT and individual merchants do not need additional work for that integration. Inclusion in the data supply should not be confused with a guaranteed recommendation.

What Peec observed

In its 16 October research article, Peec says it examined debug output from shopping requests. It reports US retrieval using an index called Labrador, while European examples drew from Google Shopping through a third party service. The article illustrates the US process with a running shoe query and the European process with a separate test prompt.

Peec also reports truncated product descriptions and a seller verification flag in the US examples. It explicitly says it has not identified what determines that flag. Those observations are not a merchant verification procedure, and they do not justify paying someone who promises to switch the flag on.

What the evidence does not establish

The published article gives concrete examples, but does not describe a representative sampling plan across countries, accounts, product categories and dates. We have not independently reproduced its debug analysis. A result from one observed configuration cannot establish a permanent rule for an entire region.

Peec sells AI visibility analytics, so readers should also distinguish its research observations from the product recommendations accompanying them. Its article is evidence of what its researchers report seeing, not official documentation of OpenAI’s production architecture.

OpenAI’s current shopping explanation describes first and third party metadata, user context and model responses among the inputs to product selection. Its merchant ranking explanation includes availability, price, quality and whether a merchant makes or primarily sells the item. It does not confirm the exclusive regional split described by Peec.

A practical regional visibility check

Our recommendation is to treat the report as a testing hypothesis. Choose a small, documented set of real customer questions covering your main products. Record the date, shopper country, account context, exact prompt, recommended product and linked merchant. Repeat the same checks over time; keep product inclusion separate from which retailer is offered for that product.

When an answer is wrong, compare the displayed price, availability, variant and description with your store and the feeds you maintain. Preserve screenshots and destination links before changing anything. Fix a demonstrable data mismatch first, then repeat the check. A missing recommendation alone does not identify the cause.

Do not delete useful product detail to meet an alleged universal word limit, multiply listings simply to occupy carousel positions, or abandon one acquisition channel on the basis of these examples. Make any proposed change a bounded experiment with an agreed success measure. Track accurate product presentation and qualified visits separately from sales; a visibility change does not itself demonstrate additional revenue.

Watch the original presentation

Peec’s article includes the BrightonSEO 2026 presentation below. The recording provides the researcher’s context; its inclusion is not independent validation of every claim.

Watch the presentation on YouTube.