Why ChatGPT and Gemini name different stores for the same question

A customer types 'where can I buy a good robot vacuum cleaner' into ChatGPT. Then they ask Gemini the same thing. The two lists of stores barely overlap. If you run an online store, this matters, because both tools are increasingly the first place people look before they buy something.

They are built differently

ChatGPT and Gemini are both AI assistants, but they work in different ways. ChatGPT (made by OpenAI) was trained on a large dataset of text that has a fixed cutoff date. When you ask it a question, it draws on what it learned during training. It can also search the web in some versions, but the base model relies heavily on patterns from older text. Gemini (made by Google) is more tightly connected to Google Search. It pulls in current web results and processes them differently. This means the two tools are looking at different sources, weighted in different ways, when they decide which stores to mention.

Different sources, different stores

Because they use different underlying data, they pick up on different signals. ChatGPT tends to name stores that appeared frequently in reviews, articles, forum discussions and editorial content that existed before its training cutoff. Gemini tends to favor stores with a strong current presence in Google Search, including product pages, structured data and recent content. A store that is well covered in blogs and editorial pieces from a few years ago might show up in ChatGPT but not Gemini. A store with strong recent SEO and product listings might appear in Gemini but not ChatGPT. Neither tool is wrong. They are just reading different things.

The numbers show how much they differ

In the latest Krillo Index measurement, we tracked 4826 buying questions sent to ChatGPT and 4839 sent to Gemini, across 65 product categories in the Netherlands, Belgium and the United States. ChatGPT named at least one store in 99% of questions, with 5.6 stores per answer on average. Gemini named at least one store in 100% of questions, with 7.1 stores per answer on average. Gemini gives longer lists, which means more stores get a mention, but also more competition. Across the 3124 webshops we measured, 59% were not named by either tool in a single buying question. That is a large group of stores that are essentially invisible to AI shoppers right now.

What each tool responds to

If you want to appear in ChatGPT answers, the most useful thing is coverage. This means being mentioned in articles, buying guides, comparison pieces and reviews that exist on the open web. ChatGPT learned from text, so stores that appear in written content across many sources are more likely to show up. If you want to appear in Gemini answers, your website itself matters more. Clear product information, well-structured pages, good reviews on Google and a visible presence in current search results all help. Gemini reads the web the way Google does, because it is part of the same system.

What you can do this week

There are a few concrete things worth doing. First, search for your store name in both ChatGPT and Gemini, using the kind of questions your customers actually ask. Write down which buying questions you appear in, and which ones you do not. That tells you where the gap is. Second, look at whether your store is mentioned anywhere outside your own website. If the only place your name appears is your own pages, that is a problem for ChatGPT visibility. Reach out to blogs, comparison sites or editorial publications in your category and see if there is a natural way to get coverage. Third, check your own website for clarity. Do your product pages clearly state what you sell, who it is for and why someone should buy from you? Structured, specific information helps Gemini understand what your store does. Fourth, make sure your Google Business presence is up to date, and that you are collecting real customer reviews. These feed into the signals Gemini uses.

One thing to keep in mind

You are not optimizing for one tool. You are building a presence that is readable and trustworthy across multiple places on the internet. Stores that show up consistently in both ChatGPT and Gemini tend to have two things: editorial mentions from credible sources, and well-structured, informative websites. Neither of those things happens overnight, but both are things you can start working on now.

If you want to see how your store currently performs in AI recommendations, you can run a free check at krilloai.com. It shows you which buying questions your store appears in, and where you are missing.