By Carson Cobb. Data collected November 4, 2026.
When a shopper asks ChatGPT for a recommendation, the answer usually comes as a ranked list. But a list of what? A brand name, or a specific product with a model name? The difference matters: one sends the shopper to your homepage, the other to a specific product page with a specific name they’ll search for.
I took the 48 buyer questions from my Who ChatGPT recommends research and sorted every ranked pick ChatGPT gave into brand or product.
The short version
- About 6 in 10 picks named a specific product. 107 of 181 ranked picks.
- Most lists were all products. 23 of 37 ranked lists named only products.
- Only 6 lists named only brands. Another 8 mixed the two.
- 11 answers had no numbered list at all. They described options in paragraphs or tables.
- The question decides it. Questions about brands or stores got brands. Questions about a need got products.
What ChatGPT named
| Count | |
|---|---|
| Answers with a ranked list | 37 of 48 |
| Ranked picks | 181 |
| Picks naming a specific product | 107 (59%) |
| Picks naming only a brand | 74 (41%) |
| Type of list | Answers |
|---|---|
| Only products | 23 |
| Only brands | 6 |
| Products and brands mixed | 8 |
| No ranked list | 11 |
When AI names brands and when it names products
The pattern followed the question. “What are the best sustainable activewear brands?” got a list of brands. “What are the best squat proof leggings?” got a list of specific leggings, with model names, sometimes with the color and size. Questions about where to buy (“best online stores for single origin spices”) got stores.
Most real shopper questions describe a need, not a brand. So most of the time, what AI recommends is a product, and the product’s name is what the shopper carries forward.
Why your product name now matters more
When ChatGPT names a specific product, the shopper searches for exactly that name, on Google, on Amazon or on your site. If your product page calls it “The Trailblazer”, or the name differs between your site, your product data and your retailer listings, the shopper may not find you, and AI may not connect the sources.
Several picks in the data used long names that mixed brand, model, color, size or pack count. The more parts a name has, the more ways it can fail to match the name on your own page.
What to do with this
- Find out which of your products AI names. Ask ChatGPT the questions your shoppers ask and note the exact product name it uses.
- Name products for what they are. A name that includes the product type (“Merino Hiking Sock”) gives AI and shoppers something to match. See naming products for Google and AI.
- Use the same name everywhere. Product page, product data, Merchant Center, Amazon and other retailers, and the name you give reviewers.
- Make each product page stand on its own. If AI sends a shopper straight to one product, that page has to answer their questions without the homepage. See do product pages answer AI shopper questions.
- Build brand pages for brand questions. For “best brands” questions, your About page, press and reviews of the brand carry the weight.
How AI names and describes your products is part of my AI Search and Conversion Audit. The AI Visibility Check is a quick first look.
How this was done
- Answers: the 48 short buyer questions from Who ChatGPT recommends, 8 in each of six product areas, asked in ChatGPT on November 4, 2026, each in a fresh session.
- Picks: the items in each numbered list in ChatGPT’s answer. A pick counted as a brand when its name matched a brand name in the answer and nothing more. Anything with a model or product name counted as a product. Matching was automated against a list of brand names and their variants, so a few picks may be misclassified.
- Limits: ChatGPT only, 48 questions. Gemini and Google AI Mode format answers differently and weren’t included.
No store or brand is named in this report.