Usually for one of three reasons. The sources ChatGPT trusts do not mention you, so it has nothing to draw on. Your content does not clearly answer the questions buyers ask, so you are never the obvious choice. Or you are testing prompts that no real buyer would use. Fix the sources and the answer content, and you become citable.
With 900 million people using ChatGPT every week, according to OpenAI in December 2025, being absent from its answers is a real commercial gap. The good news is that absence almost always traces back to a small number of fixable causes.
The sources do not mention you. ChatGPT builds answers from a model trained on the web plus, in many cases, live retrieval. Either way it leans on sources it considers credible. If the publications, directories, review sites and reference pages it trusts in your category never name you, the engine has nothing to work with. You are not penalised. You are simply not present in its raw material.
Your content does not answer the question. Engines cite the page that most clearly answers what was asked. If your site talks about your brand in vague, promotional language but never plainly answers the specific questions buyers ask, you give the engine no reason to pick you. The brand that wrote the clear answer wins the citation.
You are testing the wrong prompts. This is more common than people expect. Typing your own brand name proves nothing, because the engine will obviously discuss a brand you named. The real test is the unbranded buyer question, the kind someone asks when they do not yet know you exist. Many brands check a handful of vanity prompts, see themselves, and miss that they are invisible on the questions that actually matter.
You are looking at one engine in isolation. ChatGPT is one of three engines that matter, alongside Google AI Overviews and Perplexity. Being weak in one while strong in another is common, because they source answers differently. Judging your whole AI presence on a single engine gives you a misleading picture.
There are smaller causes too, such as a site that is hard to crawl or content that is too thin to be useful. But in most cases, the answer comes back to sources and answer quality.
Showing up in ChatGPT is not a trick you flip on. It is earned. The brands that appear are the ones that credible third-party sources talk about and whose own content clearly answers real questions. If you are missing, treat it as useful feedback rather than a glitch.
It also means the fix sits largely outside your own website. You can improve your pages, and you should, but a big part of becoming citable is earning mentions in the sources the engines already trust. That is slower and more deliberate than changing a meta tag, and it is also more durable.
We start by finding out whether you really are absent, and where. Our AI Visibility Strategy runs a realistic set of unbranded buyer prompts across ChatGPT, Google AI Overviews and Perplexity, then scores the result as your AI Visibility Score out of 100. That tells you if the problem is ChatGPT specifically or your wider AI presence.
From there we diagnose the cause. We show which sources the engines lean on, where your competitors are being cited and you are not, and which buyer questions you fail to answer. Source Quality and Narrative Fit, two of the three dimensions in our framework, are usually where the missing pieces sit.
A free Exec Brief gives you that first honest read on where you stand in ChatGPT and beyond. Our methodology explains exactly how we test and score it.
Book an AVS Exec Brief: a real, one-off measurement of how ChatGPT, Google AI Overviews and Perplexity talk about your business right now. Same methodology as the full AI Visibility Strategy, delivered manually, free of charge. Annual, Bi-Annual and Quarterly cadences are bespoke priced.
Book your AVS Exec Brief →