Why doesn't ChatGPT mention my brand? 7 reasons, and how to check each one

If AI assistants recommend your competitors but never you, one of these seven causes is usually why. How to check each one, and what to fix first.

Amreet · Founder, QuothPublished 6 min read
On this page
  1. First, confirm the problem
  2. 1. It can't reach your website
  3. 2. The pages it reads don't mention you
  4. 3. Your site doesn't clearly answer the question
  5. 4. It isn't sure what your name refers to
  6. 5. What it knows about you is wrong or out of date
  7. 6. Competitors own the comparison content
  8. 7. You're newer than its memory
  9. What to fix first
  10. Frequently asked questions

The short answer: ChatGPT usually skips a brand for one of seven reasons. It can't reach your site; the pages it reads don't mention you; your site doesn't clearly answer the question; it isn't sure what your name refers to; what it knows about you is wrong or out of date; competitors own the comparison content; or you're newer than the model's memory. Each one can be checked in a few minutes, and most can be fixed.

Before you diagnose anything, check that the problem is real. AI answers vary between runs, so one missing mention doesn't prove much.

First, confirm the problem

Pick five questions a buyer would really ask, such as "What's the best [category] for [type of company]?" Ask each one in a fresh chat in ChatGPT, with web search on, and repeat the most important ones in Claude, Perplexity, Gemini and Google. Then count how many answers name you.

If you're named in none, or far less often than competitors of a similar size, work through the reasons below. Keep the list of sources each answer cited: you'll need it for reasons 2 and 6.

1. It can't reach your website

What's happening: Your robots.txt, firewall or bot protection blocks the crawlers AI assistants use for search. Your site may work perfectly for people while being invisible to assistants.

How to check:

  • Open yourdomain.com/robots.txt and look for rules that block OAI-SearchBot (ChatGPT search), Claude-SearchBot, PerplexityBot, or everything (User-agent: * with Disallow: /).
  • Check your CDN or hosting security settings for AI-bot blocking. Some providers offer one-click options that block AI crawlers.
  • Ask ChatGPT: "What does [yourdomain.com] say about [your product]?" If it can't open the page, something is blocking it.

How to fix it: Allow the search crawlers. OpenAI says sites that block OAI-SearchBot won't be shown in ChatGPT search answers. You can still block the training crawlers (GPTBot, ClaudeBot) if you prefer; that's a separate decision and doesn't remove you from search answers.

2. The pages it reads don't mention you

What's happening: This is the most common reason. When an assistant searches, it reads a handful of pages, often "best [category]" articles, review sites, forums and videos, and recommends what they agree on. If you're not on those pages, you're not in the answer, however good your product is.

How to check: Look at the sources cited in the answers from your first check. Open each one and search it for your brand name. Count how many mention you and how many mention the competitors that were recommended.

How to fix it: Work through the cited pages one by one. Ask independent authors to consider including you, with a short, useful pitch. Complete your review-site profiles and ask real customers for reviews. Take part genuinely in the communities your buyers use. This is slow, but it's where most of the movement comes from.

3. Your site doesn't clearly answer the question

What's happening: The buyer asked about something specific, such as "expense software for a 50-person startup that works with NetSuite." Your site says "The modern platform for finance teams." An assistant can't confidently match the two, so it picks a competitor whose page says exactly that.

How to check: For each of your five questions, find the page on your site that answers it. If you can't point to one page whose opening lines directly answer the question, neither can an assistant.

How to fix it: Create or rewrite pages for the questions that matter: use-case pages, integration pages, "[you] vs [competitor]" and "alternatives to [leader]". Lead with a plain two- or three-sentence answer. Research on generative engine optimisation found that adding statistics and quotations from credible sources made content noticeably more visible in AI answers, while keyword stuffing didn't help.

4. It isn't sure what your name refers to

What's happening: If your brand shares its name with a common word, a place or another company, the assistant may not connect your product to the category. It may also mix you up with someone else.

How to check: Ask "What is [brand name]?" and "What does [brand name] do?" in a fresh chat. If the answer describes something else, or hedges, you have an identity problem.

How to fix it: Use one consistent description everywhere: "[Brand] is a [category] for [audience]." Put it in your homepage's opening lines, your About page, your LinkedIn page, review profiles and directories. Use your full brand name with its category ("Acme Expenses", not only "Acme") in titles and headings. Consistent, repeated signals help assistants treat you as one clear entity.

5. What it knows about you is wrong or out of date

What's happening: The assistant does mention you, but describes you badly: an old price, a missing feature, the wrong market. That can be worse than not being named, because buyers rule you out.

How to check: Ask "What are the pros and cons of [your product]?" and "How much does [your product] cost?" Compare the answers with the truth, and note the sources each answer cites.

How to fix it: Correct the facts at the source. Update your own pricing, feature and FAQ pages in plain text, not only in images or PDFs. Ask the authors of outdated articles to update them, and refresh your review-site profiles. Assistants with live search pick up changes once the pages they read change.

6. Competitors own the comparison content

What's happening: For "[competitor] alternatives" and "[A] vs [B]" questions, assistants often quote comparison pages, and in many categories the established players wrote most of them. Their pages frame the choice in their favour, and you're either missing or described on their terms.

How to check: Search for "[leader] alternatives" and "[leader] vs [you]". See who wrote the top results, and whether you appear and how you're described.

How to fix it: Publish your own fair, specific comparison pages, including where the other product is the better choice. Balanced pages are more credible to readers and more likely to be quoted. Then pursue independent comparisons on review sites and third-party blogs.

7. You're newer than its memory

What's happening: Without web search, an assistant answers only from what it learned in training. If your company launched, rebranded or changed category after that training data was collected, the model may not know you exist at all.

How to check: Ask the same question with web search turned off and then on (where the assistant lets you choose). If you only appear with search on, the model's memory simply predates you.

How to fix it: Focus on the live-search side, which you can influence now: reasons 1 to 6. The built-in memory only catches up with new model versions, which happen on the AI companies' schedules, not yours. The more consistently you're described across the web before then, the better your chances in future versions.

What to fix first

Start with the reasons that block everything else:

  1. Access (reason 1). It takes minutes and nothing else works until it's fixed.
  2. Identity (reason 4) and wrong facts (reason 5). These are quick, and they protect the mentions you already have.
  3. Answer pages (reason 3). A few days of work for your most important questions.
  4. Third-party mentions (reasons 2 and 6). The slowest, and the most powerful. Start now and keep going every week.

Re-run your five questions every week to see what's moving. If you want a head start, our free AI snapshot runs 10 buyer questions across five assistants and shows you which of these seven reasons apply to you.

Frequently asked questions

Does ChatGPT favour big brands?

In practice, often yes, but not because of size itself. Big brands appear in more of the articles, reviews and discussions assistants read. Challengers can win specific questions, especially about particular use cases or segments, where the big brands' content is generic.

Can I ask ChatGPT to remember my brand?

No. Telling ChatGPT about your product in a chat only affects that conversation (or your own account's memory). It doesn't change what it tells other people.

How long until changes show up?

For assistants that search the web, changes can appear within weeks once the pages they read are updated and indexed. Changes to a model's built-in knowledge only arrive with new model versions, typically months later. These are typical patterns rather than guarantees.

Should I block AI crawlers to protect my content?

That's a business decision, but separate the two kinds of crawler. Blocking training crawlers (such as GPTBot or ClaudeBot) keeps your content out of future model training. Blocking search crawlers (such as OAI-SearchBot) removes you from those assistants' search answers. Most companies that want to be recommended keep the search crawlers allowed.

Amreet, Founder, QuothAmreet founded Quoth to help B2B software teams understand and improve how AI assistants describe and recommend their brands.Questions or corrections? hello@tryquoth.com

Keep reading

How to get your SaaS recommended by ChatGPT, Claude, Perplexity and Gemini8 min readGEO agency vs GEO tool: Quoth vs Profound vs Peec AI5 min read