Generative engine optimisation GEO

Why it matters

A growing share of buyers ask an AI tool for a shortlist and act on it. If the tool names three businesses and you are not one, you were never compared. GEO is the work of changing that.

The term comes from a 2023 academic paper on generative engines, and it is still young. The honest view is that GEO is mostly good SEO plus a few extras: Google says there are no special requirements for appearing in its AI features beyond the basics of being indexed and eligible for a snippet. What changes is where you put effort.

What GEO adds to SEO

• Passage-ready answers. AI systems pull passages, so each section should answer one question clearly under a plain heading.
• Third-party mentions. Engines lean on what other sites say about you: reviews, directories, press, community threads. Your own site is only one source.
• Specific, checkable facts with named sources, rather than vague claims.
• Crawl access. Allow the search bots of the assistants you care about in robots.txt, and keep key content in the page's HTML, not behind scripts.
• Measurement. Ask a fixed set of questions on a schedule and track how often you appear. This is answer engine optimisation in practice.

Common mistakes

• Promising rankings in ChatGPT. Answers vary by user, day and model; nobody controls them.
• Treating schema or llms.txt as a switch. Neither has shown a reliable lift.
• Neglecting the basics. If Google can't index you, an AI built on its results rarely finds you.
• Measuring once. Check monthly, with the same questions.

See also Google AI Overviews and query fan-out.