What GEO means for brands that want to appear in AI answers

GEO is a marketing practice that aims to get a brand named, and its pages cited, in the answers that AI assistants such as ChatGPT and Google AI Overviews write. These assistants search the web as they write and link the pages that support the answer. GEO work shapes those pages, the crawler access to them and the other websites that mention the brand.

Where generative engine optimization comes from

The term GEO comes from a research paper by Pranjal Aggarwal and five co-authors, posted on arXiv on 16 November 2023 and accepted to the KDD 2024 conference. The authors defined generative engines as search tools that use large language models to gather information from several sources and summarize it into one answer.

Aggarwal, a researcher at the Indian Institute of Technology Delhi, and his co-authors called GEO “the first general optimization framework for website owners to optimize their websites for generative engines”.

The team tested nine ways of editing a web page on GEO-bench, a set of 10,000 queries answered by OpenAI’s GPT-3.5 Turbo model. Its word-count score measures how much of an answer comes from a source, weighted by how early the source is cited.

An unedited page scored 19.3. Adding quotations raised the score to 27.2, about 41%, adding statistics raised it to 25.2 and citing sources raised it to 24.6. On Perplexity, a live AI search engine, the methods raised visibility by up to 37%, and results varied by subject area.

How AI search engines choose sources

AI search engines fetch pages from a web index while they write, then link the pages that support the answer. Google’s guide to optimizing for generative AI features, last updated on 10 July 2026, says AI Overviews and AI Mode retrieve pages through Google’s core ranking systems. Query fan-out, a batch of related searches sent at once, adds results, so a question about a weedy lawn can trigger a search for lawn herbicides.

Under that guide, a page qualifies only if it is indexed and can appear in Google Search with a snippet. The site must also be included in Search generative AI features in Search Console, a condition missing from Google’s AI features page of 10 December 2025. Indexing is the first step in how Google AI Overviews choose pages.

OpenAI’s crawler documentation says, as of 29 October 2026, that its OAI-SearchBot crawler surfaces websites in ChatGPT’s search features. Sites that block it are left out of ChatGPT search answers, though they can still appear as navigational links. Crawler access is therefore the first condition in how ChatGPT search picks its sources.

OpenAI’s documentation for the web search tool in its API adds that a model often consults more pages than it cites.

What GEO work involves

GEO in digital marketing covers three jobs: sourced answers on the brand’s own pages, access for AI crawlers, and mentions on third-party pages.

On its own pages, a brand answers the buyer’s question in the opening lines. It backs the answer with statistics, short quotes and links to sources, the three edits the 2023 GEO paper named its top performers. Google’s guide to optimizing for generative AI features asks for first-hand experience beyond what others have published, and says pages need no special style for AI. More is in how to get a brand cited in ChatGPT answers.

For crawler access, Google’s AI features page recommends allowing crawling in robots.txt and in any CDN or hosting setup, plus internal links that make content easy to find. For ChatGPT, OpenAI recommends allowing OAI-SearchBot in robots.txt. The rest of the checklist is in how to get pages into Google AI Overviews.

Third-party presence means being named in the blogs, videos and forum discussions the engines draw on. Google says its AI features can show what such sources say about products, and warns that chasing inauthentic mentions helps less than it might seem. A brand’s own website is one source among many, as data on how often AI answers cite a brand’s own website shows.

How GEO relates to SEO

GEO, the work of getting cited in AI answers, builds on SEO, and Google counts it as SEO for its own results. Google’s guide to optimizing for generative AI features names GEO and answer engine optimization as terms for AI search work. It adds that its AI features rest on the same ranking and quality systems as Google Search. The other labels, LLMO among them, are compared in how GEO differs from similar terms.

The SEO work that still counts in AI search begins with crawling and indexing, which Google requires for its AI features. Outside Google, ChatGPT search depends on OpenAI’s own crawler, OAI-SearchBot, which a site can allow or block separately. GEO also tracks mentions and citations inside answers, where SEO tracks rankings and clicks.

Some SEO habits carry over badly. In the 2023 GEO paper, keyword stuffing, which the authors called a method traditionally used in SEO, scored 17.7 against 19.3 for an unedited page. Google’s guide warns that creating separate pages for fan-out queries mainly to sway AI answers breaks its scaled content abuse spam policy.

How GEO results are measured

GEO results are measured as AI search visibility, the share of answers to a fixed set of buyer questions that name the brand or cite its pages. For ChatGPT and Perplexity, the share comes from asking the same questions at regular intervals and recording each answer’s brands and pages. The method is set out in how AI visibility is measured.

Baselines are often low. Agenzy’s 2026 study of 187,810 AI answers found the median European SMB brand appears in just 5% of AI answers about its own product category (n=90 tracked market projects, 2026-04 to 2026-08).

For Google, its guide to optimizing for generative AI features points to the Generative AI performance report in Search Console, which covers AI features on Google Search and Discover. Google’s AI features page of December 2025 instead says AI Overviews and AI Mode traffic is counted in the Performance report under the Web search type. The guide also warns that no third-party tool can see Google’s internal ranking or AI systems.

What it means

Marketing leads who want their brand in AI answers should begin with access and end with tracking, in this order.

1. Check that robots.txt and your CDN or hosting setup let Google’s crawler and OAI-SearchBot reach your pages, and that the pages are indexed.
2. Confirm in Search Console that the site is included in Search generative AI features.
3. List the questions buyers ask AI assistants about your category, without your brand name, and note the pages the answers cite.
4. Open each page aimed at one of those questions with a direct answer, and back it with sourced statistics and quotes.
5. Earn genuine mentions on the third-party pages from that list.
6. Run the same questions every month and compare the results with the Generative AI performance report.

FAQ

What does GEO cost?

GEO costs what the work costs. The conditions Google and OpenAI publish for appearing in AI answers are technical: indexing and Search Console inclusion for Google, crawler access for ChatGPT. The budget goes to writing pages, fixing access, earning mentions and tracking answers, done in-house or by a GEO agency.

How long does GEO take to show results?

No engine promises a timeline. Google’s guide says meeting every requirement still does not guarantee crawling, indexing or serving. OpenAI’s crawler documentation says a robots.txt change takes about 24 hours to reach ChatGPT search results. Each page then waits for a first citation, a wait measured in how long GEO takes to show up in ChatGPT answers.

Do you need llms.txt or structured data for GEO?

Neither is needed for Google. Google’s guide says Google Search ignores llms.txt and similar AI text files, so one neither helps nor harms there. The same guide says structured data is not required for its generative AI features, though it still helps pages qualify for rich results. For ChatGPT search, OpenAI’s crawler documentation points to robots.txt rules for OAI-SearchBot.