What is GEO (generative engine optimisation)?
GEO is the work of getting a business included and cited in answers written by generative AI. Where the term comes from, what the work involves, and its limits.
GEO, generative engine optimisation, is the work of getting a business included and cited in the answers generative AI writes, in ChatGPT, Google AI Overviews, Perplexity and Copilot. The term comes from a 2023 research paper that tested how changes to a source affect its visibility in those answers. In practice it overlaps almost entirely with AEO.
The term comes from a 2023 research paper
GEO: Generative Engine Optimization, by Pranjal Aggarwal, Ameet Deshpande and colleagues, was first posted in November 2023 and presented at KDD 2024. It describes generative engines as systems that answer a query by gathering information from several sources and summarising it with a large language model, and notes that content creators have little to no control over when and how their content is shown. The paper defines GEO as a way to help them improve their visibility in those responses.
The authors built a benchmark of queries across many domains, GEO-bench, and report that their methods can raise visibility by up to 40% in generative engine responses, with results that vary from one domain to another. That is a research finding on a benchmark. It says nothing about what any one client's answers will do.
A generative engine writes the answer, then cites a few sources
A generative engine does not hand the reader a list to choose from. It gathers sources, writes one answer and links a few of them. Google says its AI Overviews and AI Mode may run several related searches across subtopics, a query fan-out, to build a response. Perplexity says its crawler is designed to surface and link websites in its search results. So for the same question, a business can be named on one engine, read but not named on another, and not found on a third.
GEO, AEO and SEO compared
| Term | Emphasis | Measured by |
|---|---|---|
| SEO | A higher position in a list of search results | Rankings, clicks and impressions |
| AEO | Being the answer: named in the reply and cited as a source | The answers to the buyer's questions, engine by engine |
| GEO | Being included in answers that generative AI writes | The same answers, and the share of mentions against rivals |
Most agencies sell GEO and AEO as one service. The foundations still come from SEO: Google's guidance says its AI features need no special optimisations beyond a page being indexed and eligible to show with a snippet.
What GEO work involves for an agency
- A baseline. The client's buyer questions, asked on each engine, with every answer and its sources kept: who is named, who is named instead, and which pages the engines read.
- A fix list. The pages and changes the evidence points to, from questions no page answers to sources only rivals get.
- The work. Done by the agency's own team, or its AI coding tool, in the client's own site.
- Proof. An independent check that each page is live and answers its question, and the same question on each engine before and since.
Inciteful covers the measurement, the plan and the proof: AI visibility and share of voice for the baseline, the AEO Boost on Pro and Agency for the fix list your AI coding tool works from, and Proof afterwards. It does not write or publish pages.
What GEO cannot promise
No agency controls which sources an engine chooses, so no one can promise a client a place in an AI answer. What an agency can promise is the process: access for AI crawlers, pages that answer the buyer's questions, and a record of what was measured after the work. An answer can change for reasons that have nothing to do with the page that shipped, so report what was measured after the work, never that the work caused it.
What does GEO stand for?
Generative engine optimisation: the work of getting a business included and cited in answers written by generative AI, such as ChatGPT, Google AI Overviews and Perplexity. The term was introduced in a 2023 research paper, GEO: Generative Engine Optimization, presented at KDD 2024.
Is GEO the same as AEO?
In practice, largely. AEO, answer engine optimisation, stresses being the answer; GEO stresses visibility in answers that generative AI writes. Agencies sell them as the same service, and both are measured the same way: whether the client is named, read but not named, or not found.
Is GEO replacing SEO?
No. GEO builds on SEO. Google says a page needs to be indexed and eligible for a snippet to appear as a supporting link in its AI features, with no special optimisations beyond that. A site that search engines cannot read is hard for an AI assistant to cite.
How much can GEO improve visibility?
It depends on the market and the engine. The 2023 GEO paper reports gains of up to 40% in generative engine responses on its own benchmark, and notes that results vary by domain. For a real client, the number that counts is the one measured on its own buyer questions, before and after the work.
How do you measure GEO?
Ask each engine the questions a client's buyers ask, keep every answer with its sources, and sort each answer into named, read but not named, or not found. Compare the client's share of mentions with the rivals the answers name. Inciteful's free visibility report runs this once for any site, with no card.
Which engines matter for GEO?
The ones a client's buyers use. For most businesses that means ChatGPT, Google AI Overviews and Perplexity, then Copilot, Claude, Gemini and Google AI Mode. Inciteful tracks the first three on every plan, includes Copilot on Pro and Agency, and offers Claude, Gemini and Google AI Mode as add-ons per client.
