Why Your Business Doesn't Show Up in AI Answers
Someone asks ChatGPT for a supplier in your sector, in your region, doing exactly what you do. It names four companies. None of them is you. This piece is about working out which of the six likely causes applies to your site, because the fix for each one is different and most of them are not a content problem.
Why isn't my business showing up in ChatGPT or AI Overviews?
Because an AI assistant can only mention a business it can find, read and trust, and most sites fail at one of those three stages rather than all of them. The usual causes are a blocked crawler, content rendered client-side, pages that answer nothing directly, no third-party corroboration, thin or absent structured data, or a site that is simply too new to have accumulated any signal.
Work through them in that order. There is no point rewriting your service pages for answer extraction if a line in your robots.txt is stopping the crawler reading them in the first place.
Is my robots.txt blocking AI crawlers?
Very possibly, and it is the first thing to check because it makes every other effort pointless. AI assistants use their own named crawlers, separate from Googlebot, and a robots.txt written before those crawlers existed will not mention them. Some site builders and security plugins also block unrecognised user agents by default.
Open yourdomain.com/robots.txt and read it. A blanket disallow, or an allowlist that names only Googlebot and Bingbot, will keep you out of AI answers entirely no matter how good the content is. This is a five-minute check and the highest-leverage fix on the list.
Worth deciding deliberately rather than by accident: you may have legitimate reasons to block a particular crawler. Just make sure it is a decision you made, not a default you inherited.
Does my site render content that crawlers can actually read?
If your site is a React, Vue or similar single-page application without server-side rendering, quite possibly not. Crawlers read the HTML that comes back from the server. If your copy is assembled in the browser after load, the crawler may receive an almost empty shell with a script tag.
Test it by viewing source rather than using the browser inspector. The inspector shows you the page after JavaScript has run; view-source shows you what the crawler got. If your headings and body copy are missing from view-source, that is your answer.
The same failure catches verification tags and structured data injected at runtime. If it is not in the served HTML, assume it does not exist.
Why do AI answers cite competitors instead of us?
Usually because those competitors have published something that answers the question directly and you have published something that describes your services. An AI assistant assembling an answer needs a passage it can lift and attribute. A page that talks around a topic across six paragraphs gives it nothing to work with.
The second reason is corroboration. Assistants weight information that appears in more than one place, so a business with a Companies House record, consistent directory listings, and mentions on sites it does not own is easier to be confident about than one that exists only on its own domain.
Neither of these is fixed by writing more. They are fixed by writing differently, which is covered in how to get your business mentioned in AI answers.
Does having a blog help you appear in AI answers?
Only if the posts answer questions people actually ask, in a format that can be extracted. A blog of company news, product announcements and industry commentary contributes almost nothing, because none of it is the answer to a question anyone typed.
What works is narrower and less glamorous: a page per real question, the answer stated plainly in the first paragraph under the heading, and the rest of the page supporting that answer rather than delaying it. That structure is what makes a passage liftable.
It is also why volume is a poor strategy here. Forty vague posts perform worse than six that each definitively own a question.
How long does it take to start appearing in AI answers?
Longer than you would like, and it varies by cause. Unblocking a crawler or fixing server-side rendering can show up within weeks, because the barrier was mechanical. Building the corroboration and authority that makes an assistant confident enough to name you takes months, because you are waiting on other sites and on crawl cycles you do not control.
Be sceptical of anyone quoting you a specific timeline. The honest answer is that the mechanical fixes are fast and the trust-building is not, and which of those you are dealing with depends on which of the six causes above is actually yours.
If you would rather not work through the list on your own, our search recovery service starts with a diagnosis of which cause applies before anything gets changed. There is also a free audit if you want a second opinion first.
Related reading
Start with the audit, not the software.
Thirty minutes on a call. We map how work moves through your business, show you where the time is going, and tell you what is worth building. You keep the roadmap either way.