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

Start every important section with its answer in the first sentence, then add the context. Write passages that stand on their own, with no reliance on 'as mentioned above', so a model can quote them out of context. Prefer plain definitions in the form 'X is...', and back claims with a number or a named example, because specific passages are quoted far more often than general ones.

AI crawlers read your raw HTML and most do not run JavaScript, so any content that only appears after a script runs is invisible to them. Server-render what matters. Confirm your robots.txt allows the AI crawlers you want, such as GPTBot, ClaudeBot, and PerplexityBot, and publish an llms.txt file at your root as a plain-text map to your best pages.

Give the model labels for what it is reading. A clean heading hierarchy maps sections to the questions people ask. Schema.org structured data marks up your organization, articles, and FAQs so a model knows what your page is and who wrote it. This is the difference between a model guessing at your content and a model being told.

This is the part you cannot fake with formatting. Google AI Overviews lean on pages that already rank, so traditional SEO and domain authority matter most there. ChatGPT leans on established, well-cited reference sources, so being mentioned across the web helps. Perplexity leans on community discussion and citations, so presence on the sites people reference, including Reddit and industry forums, moves the needle. Being cited by all three for the same query is rare, so treat each as its own target.

Ask each assistant the questions your customers ask and record who it cites. Watch analytics for referral traffic from ChatGPT, Perplexity, and Google AI surfaces. Track brand mentions, not only links, because a model can name you without sending a click. Then rewrite the pages that lost, and ask again.