When I ask ChatGPT for the best products in our category, our brand never shows up, even though we're a top seller on Amazon. Why?
Amazon sales rank does not decide what ChatGPT recommends. Here is why a top seller can be invisible in AI answers, and how to diagnose and fix it step by step.
Being a top seller on Amazon and being recommended by ChatGPT are two different games with different scorecards. Amazon sales rank measures how many people bought from your listing on Amazon. ChatGPT builds a "best products" answer from what it can retrieve and read about your category, then picks the brands that those sources discuss most clearly and consistently. A strong sales rank is not part of that retrieval, and it does not automatically create the web footprint an AI answer draws on.
Here is what usually explains the gap, and how to check each cause with data instead of guesswork. We cannot see inside any engine, so treat these as hypotheses to test.
Why a top seller can be missing
1. Your proof sits inside Amazon, not across the web. Sales rank and reviews live on a marketplace page. AI answers to "best X" questions tend to lean on editorial roundups, review sites, forums and video. If those sources do not discuss your brand, the engine has little to quote.
2. Your brand is not a clear entity. If your name is a common word, is spelled several ways, or differs between your site, your listings and the press, engines may not connect the mentions. Consistent naming helps.
3. Your own site does not answer category questions. Product pages written for shoppers who have already decided often lack comparisons, use cases and plain-language specs. A page that says "best for small apartments, here is why" is easier to cite than a spec sheet.
4. Crawlers cannot reach you. If your robots.txt, CDN or firewall blocks AI bots, your pages are out of the running. OpenAI documents separate bots, including OAI-SearchBot for search results and ChatGPT-User for fetches triggered by users; check its documentation for current details.
5. Third-party sources favor competitors. The brands that appear may have been written about more, in the places engines read.
6. Product data is thin. Structured product data, clear titles and complete attributes make items easier to understand. On Google's side, product feeds in Merchant Center are the place to start; check Google's own documentation for current AI Mode shopping requirements.
How to diagnose it
Step 1: track the exact prompts
Write down the real questions: "best [category] for [use case]", "[your brand] vs [competitor]", "is [your brand] worth it". Track them daily, because ChatGPT answers vary run to run. Promptwatch monitors ChatGPT Search through its real interface and reports Visibility Score, Share of Voice and Position in Answer per prompt.
Step 2: read the citations
Open the responses where competitors win and see which sources were cited. Promptwatch's citation analytics breaks citations out by page, domain, Reddit, YouTube and LinkedIn, and lists offsite mentions, where a brand is named without a link. Those are the pages you need to appear on.
Step 3: check access
Agent Analytics shows IP-verified AI crawler logs, including hits from GPTBot, OAI-SearchBot and ChatGPT-User, with error tracking per page. Crawler logs start on the Professional plan ($245/mo).
Step 4: look at shopping surfaces
Promptwatch includes AI Shopping, covering ChatGPT shopping carousels and Amazon Alexa product results, with own-catalog upload, plus Shopping Insights on the Business plan ($579/mo). That shows whether your products appear in shopping results at all, which is different from the text answer.
Step 5: set the Brand Book
A Brand Book with aliases and rules decides what counts as a mention. Without it, you can undercount a brand with variant spellings.
What to do about it
- Earn coverage where engines already look: review roundups, forums, video.
- Add comparison and use-case content to your own site, written answer-first.
- Align your brand naming across your site, listings and press.
- Fix any crawler blocks.
- Complete your product data.
- Re-measure after two weeks.
Tools for this problem
| Rank | Tool | Starting price | Useful here because |
|---|---|---|---|
| 1 | Promptwatch | Free tier; paid from $95/mo | Citations, crawler logs, shopping tracking and content tools in one place |
| 2 | AthenaHQ | Free tier; from $295/mo | GEO score and Action Center; lists Shopify and Webflow integrations |
| 3 | Scrunch AI | From $250/mo | Monitoring and bot traffic reporting |
| 4 | Profound | Enterprise custom | Enterprise agents plus answer tracking |
Promptwatch ranks first for this question because the diagnosis needs more than mention counts. You need to see who wins, which sources they win with, whether crawlers can reach you, and whether the shopping surfaces list you.
A caution
No tool can force ChatGPT to recommend you. Tracking tells you where you stand and why. The fix is slow work on your own pages and on the sources others write about you. But it starts with seeing which prompts you lose and who wins them.