Glossary · GEO (Generative Engine Optimisation)

GEO (Generative Engine Optimisation)

The discipline of optimising content so that AI assistants (ChatGPT, Claude, Perplexity, Gemini) cite a subject favourably and accurately when answering relevant queries.

Full Definition

GEO is the AI-era successor to SEO. Where SEO optimises for ten blue links on Google, GEO optimises for the single AI-generated answer that an LLM produces — and for the two or three sources the LLM chooses to cite. GEO tactics include: shipping llms.txt and llms-full.txt files (per llmstxt.org), rich schema.org markup, canonical sources of truth (Wikipedia, Wikidata) that AI training runs absorb, structured Q&A blocks that match likely user queries, and direct relationships with publications whose content carries disproportionate weight in AI-training corpora. For reputation work, GEO is now critical: a hostile or misleading AI summary in the first sentence of an executive's profile can prejudice a deal, a board appointment, or a board's first impression. The fix involves correcting source content the AI trained on, and shipping AI-friendly canonical sources before the next training run.

In practice

  • A founder's company page is rewritten to be AI-friendly: clean Organization schema, FAQ schema, llms-full.txt with full company narrative — three months later, ChatGPT and Perplexity both cite the company correctly when asked about it.
  • A celebrity's Wikipedia page is updated with new career milestones; the next round of AI training runs absorbs these and the AI summary now reflects the new context.

Also known as

AI SEO · AEO · answer engine optimization