Press slash to focus this field, or Escape to return home.
Why doesn't ChatGPT know my brand?
AnswerIt is usually one of five things, and they have completely different fixes. The first question worth asking is whether it looked and found nothing, or never looked at all.
It is answering from training. A model answering from memory only knows what was in its training data, which has a cutoff and does not include most small businesses at any point. If the answer arrives without citations, that is usually what happened, and no amount of on-site work changes it this week.
The search crawler is blocked. ChatGPT search uses OAI-SearchBot. Perplexity uses PerplexityBot. Anthropic uses Claude-SearchBot. These are distinct from the training crawlers and are frequently blocked by a robots.txt written to keep AI out, by a security product, or by a firewall rule nobody remembers adding. This is the most common fixable cause and it takes minutes to check.
Your content only exists after JavaScript runs. No vendor documents whether their crawlers render. If yours do not, a client-rendered page is an empty page to them. Googlebot renders, so a site can rank perfectly well in Google, qualify for an AI Overview, and still be absent from ChatGPT. That split surprises people, and it is the part worth understanding properly.
Nothing outside your site says you exist. Models weight corroboration. If every claim about you appears only on your own domain, there is nothing to confirm it against. This is the slowest one to fix and the one most worth fixing, because it is also what makes the other four worth doing.
You are simply new. A domain that has existed for three weeks has not been crawled much, linked to much, or written about anywhere. There is just nothing there yet.
The diagnosis matters more than the fix, because four of these five have different remedies and the wrong one costs you months. Checking which applies is mostly reading your own server logs, and that is free.