Definition
GEO (generative engine optimization) is optimising your content and your brand so that generative AI systems — ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews and AI Mode — mention and cite you when someone asks a question you have the answer to. Where SEO is about a position in a list of links, GEO is about being part of one composed answer.
Generative engines almost all work the same way: for a question they retrieve a set of pages, cut them into passages, pick the passages that fit the question best and have a language model write an answer from them, with the passages used as sources. GEO acts on each of those steps. Your pages have to be reachable for AI crawlers, they have to sit in readable HTML without depending on JavaScript, and the passages have to be so direct and factual that they answer the question better than the competition.
Beyond pages, GEO is about your brand as an entity. A model that keeps encountering your name together with a category and a few facts is willing to recommend you for a question without sources. That is why an unambiguous brand description, a factual about page and mentions on other sites belong to GEO just as much as the text on your own pages.
GEO does not replace SEO. Engines find their sources largely through classic search indexes, so whoever is not crawlable and relevant does not end up in an AI answer either. GEO adds a layer: content written as an answer, a brand recognisable as an entity, and measurement per engine in mentions and citations instead of positions.
Measurement happens with prompt sampling: a fixed set of questions you ask each engine repeatedly, after which you count how often you are mentioned, which pages are cited and which competitors appear next to you. Because answers vary from run to run, one measurement says little; the trend over months says everything.