What is generative engine optimization (GEO)? A plain-English guide

When someone asks ChatGPT, Perplexity, Gemini or Claude "which tool should I use for X", the assistant writes an answer and names a few brands. GEO is everything you do so that yours is one of them, described accurately.

Some people call it AI SEO, LLM SEO or answer engine optimization (AEO). They mean roughly the same thing.

• SEO aims for a position in a list of links. GEO aims for a mention inside one answer, often with only three to five brands named.
• With SEO, your own pages do most of the work. With GEO, other people's pages matter as much: the articles, lists and reviews the assistant reads.
• SEO has rank trackers. GEO is measured by asking the same questions repeatedly and recording who gets mentioned and which sources are cited.

Good SEO still helps. Assistants that search the web often start from pages that rank well.

• Being named on the pages assistants cite for your category: reviews, comparisons, "best X" lists and news coverage.
• Saying the same clear facts about yourself everywhere, so the model has something consistent to repeat.
• A website that states plainly what you do, who it's for and what it costs, and that loads without a login.
• Time. New coverage has to be published, indexed and picked up before it shows in answers.

Pick ten questions your customers would ask, without your brand name in them. Ask them in two or three assistants every week or two and record whether you're mentioned, how you're described, and which sites are cited.

The cited sites are the useful part: they tell you where coverage would make a difference.

• Write your one-sentence description and three facts you can back up, and use them on your site and profiles.
• Run your ten questions once to get a baseline.
• List the sites cited most often, and plan how to be on two or three of them: a pitch, a review, or a clearly labelled sponsored piece.