How to get cited by ChatGPT, Claude and Perplexity
AI answer engines are becoming the new front page. Getting quoted by them — GEO — rewards the same clarity that good design does.
Kumar Gaurav
Founder
Generative engine optimisation (GEO), also called answer engine optimisation (AEO), is the practice of structuring content so AI assistants like ChatGPT, Claude and Perplexity quote and cite it in their answers. It overlaps with SEO but optimises for extraction and trust rather than just ranking — and it rewards clarity, structure and honesty over keyword stuffing.
What is GEO, and how is it different from SEO?
SEO earns a click from a ranked list of links. GEO earns a mention inside a generated answer, where the model has already summarised the web for the user. The signals overlap — good content still wins — but GEO leans harder on clear structure, named sources and extractable statements the model can lift with confidence.
How do you make a page easy for an AI to cite?
- Answer first — open each section with a direct, 40–60 word answer the model can lift verbatim.
- Use question-style headings that mirror how people actually ask.
- Write in short paragraphs and lists with clear definitions — the structure assistants parse best.
- Show your authority — dated, bylined pages, a consistent identity, and original points of view.
- Keep it fresh — update cornerstone pages, since some engines favour recent sources.
- Help the crawlers — Article and FAQ schema, an llms.txt, and access for GPTBot, ClaudeBot and PerplexityBot.
Does each assistant reward different things?
Broadly, yes. Encyclopedic, authoritative content tends to travel well with ChatGPT; Perplexity leans toward fresh, well-sourced pages; and clear structure, definitions and lists suit how Claude reads. Write for all three and the overlap is large — the habits that help one usually help the others.
What doesn't work?
The old tricks. Keyword-stuffed prose, thin pages, and invented statistics are easy for a model to distrust or ignore, and fabricated numbers are a reputational risk once they're repeated. GEO rewards being genuinely, verifiably useful — which is the same thing good design has always asked for.
How we use it
We write every insight answer-first, structured and honestly sourced, then back it with clean schema and crawler access. It's not a trick bolted onto content — it's the same clarity we design interfaces with, applied to words.
Let's build
Have something worth making well?
Tell us what you're working on. We'll come back with how we'd approach it — and whether we're the right studio for it.