Guides · HolicAudit

How to get your website cited by ChatGPT, Perplexity and Google AI Overviews

Published 2 November 2026

In short: nobody can guarantee a citation, and anyone who promises one is guessing. What you can control is whether AI search tools are allowed to fetch your pages, whether they can read them without running JavaScript, and whether your pages state clear, specific facts worth quoting. Most of that is ordinary technical SEO done properly.

How AI answers pick their sources

AI search tools answer a question by running a web search, reading a handful of the pages it returns and summarising them with links. Each has its own crawler and index:

ChatGPT search uses OpenAI's OAI-SearchBot. OpenAI says sites that opt out of OAI-SearchBot won't be shown in ChatGPT search answers. Pages a user asks ChatGPT to open are fetched as ChatGPT-User.
Perplexity builds its index with PerplexityBot.
Claude uses Claude-SearchBot for search results.
Google AI Overviews and AI Mode draw on the normal Google Search index, crawled by Googlebot. Google says there are no extra requirements: a page must be indexed and eligible to show a snippet.
Microsoft Copilot is grounded in Bing, so being well indexed in Bing matters.

Training crawlers such as GPTBot and ClaudeBot are separate. They feed future models, not today's search answers. See should you block GPTBot for that trade-off.

1. Let the search crawlers in

Check robots.txt for Disallow rules that catch OAI-SearchBot, PerplexityBot, Claude-SearchBot, Googlebot or Bingbot. Then check the layer people forget: CDN and firewall bot settings. Some bot-protection features block AI crawlers before robots.txt is ever read, and the only symptom is that AI tools say they can't open your site.

2. Put the content in the HTML your server sends

Googlebot renders JavaScript. Most AI crawlers don't: a late-2024 analysis by Vercel and MERJ found that none of the major AI crawlers it studied, including OpenAI's and Anthropic's, executed JavaScript. Some downloaded script files but never ran them. A React or Vue single-page app that ships an empty <div id="root"> looks blank to them. Use server-side rendering or static pre-rendering (Next.js, Nuxt, Astro, SvelteKit, Remix) for any page you want quoted. Our AI visibility check counts the words in your raw HTML so you can see what they see.

3. State facts that are easy to quote

AI answers quote specific, self-contained statements. Pages that win citations tend to say plainly what the product is, who it's for, what it costs, how it compares and what the limits are, in text rather than images. Useful patterns:

A one-sentence definition near the top of the page: "Acme is invoicing software for freelance designers in the EU."
Prices, plans and limits written out in HTML, not only in a pricing image or behind a calculator.
An FAQ answering the questions customers really ask, in their words.
Comparison and "best X for Y" pages that are honest about where competitors are better.
Visible dates on content that goes out of date, and updates when it does.

4. Describe yourself with structured data

JSON-LD structured data such as Organization, Product, SoftwareApplication, LocalBusiness and FAQPage gives search engines unambiguous facts: your name, logo, social profiles, prices and ratings. It isn't a ranking switch and Google only shows some types as rich results, but it removes guesswork about who you are. Mark up only what's visible on the page.

5. Be mentioned elsewhere

AI answers often cite third-party pages rather than your own: reviews, comparison articles, directories, forum threads and documentation. Listing your product in relevant directories, getting reviewed and answering questions where your customers ask them helps both classic SEO and AI answers. Don't fake it: spam and fake reviews get filtered and can get you removed.

6. Make sure you're indexed

Verify your site in Google Search Console and Bing Webmaster Tools, submit your XML sitemap to both, and fix pages marked as not indexed. Bing supports IndexNow for faster discovery of changed pages. Avoid nosnippet on pages you want cited: Google says it also keeps content out of AI Overviews.

Optional extras

An llms.txt file is a proposed convention for pointing AI tools at your key pages. It's cheap to add, but no major AI search engine has said it uses it for ranking or citations. If you have a public API, an MCP server lets assistants like Claude and Cursor use it directly; ours is free to try.

How to tell if it's working

In Google Analytics, look for referral visits from chatgpt.com, perplexity.ai, gemini.google.com, copilot.microsoft.com and claude.ai. ChatGPT adds utm_source=chatgpt.com to many of its links. Then ask the tools the questions your customers ask and note who gets cited. Results vary by wording, location and day, so look for trends rather than single answers.

Frequently asked questions

Can anyone guarantee my site gets cited by ChatGPT?

No. Which sources an AI answer cites depends on the question, the search results it gets and the model, and changes over time. You can remove the technical blocks and make your pages worth quoting, but no one controls the outcome.

Is GEO different from SEO?

Mostly not. Generative engine optimisation is largely technical SEO plus clear writing: crawlable pages, server-rendered content, good titles and descriptions, structured data, and being mentioned by others. The main AI-specific parts are the extra crawlers to allow and the fact that most of them don't run JavaScript.

Do I need llms.txt to appear in AI answers?

No. It's an optional proposed convention. Allowing the search crawlers and serving readable HTML matter far more.

Does blocking GPTBot remove me from ChatGPT search?

No. OpenAI treats GPTBot (training) and OAI-SearchBot (search) as independent settings. Blocking GPTBot while allowing OAI-SearchBot keeps you eligible for ChatGPT search.