Learn
Article 01 / Foundations

What is GEO?

GEO, or Generative Engine Optimization, is the practice of making a business more likely to be cited, mentioned, or recommended in answers written by AI systems such as ChatGPT, Google AI Overviews, Google AI Mode, Perplexity, Gemini, and Microsoft Copilot. GEO builds on SEO, reviews, and Google Business Profile. It does not replace them.

Last updated: October 12, 2026

Who coined the term GEO?

The term "Generative Engine Optimization" comes from an academic paper titled "GEO: Generative Engine Optimization." Pranjal Aggarwal, Vishvak Murahari, Tanmay Rajpurohit, Ashwin Kalyan, Karthik Narasimhan, and Ameet Deshpande wrote it. The authors were affiliated with Princeton University and IIT Delhi, among others. The paper went up on arXiv on November 16, 2023, and was presented at the ACM KDD 2024 conference in Barcelona in August 2024.

The GEO paper defined a "generative engine" as a system that retrieves web sources and uses a large language model to write one answer with inline citations. The authors argued that ranking position, the core SEO metric, does not capture visibility inside a written answer. They proposed new "impression" metrics instead, such as how many words in an answer are attributed to a source and how early that source appears.

What did the GEO research actually find?

The GEO paper tested nine content-editing methods on GEO-bench, a benchmark of 10,000 queries. Adding quotations from credible sources, adding statistics, and citing sources gave the biggest gains. Quotation Addition raised the position-adjusted word-count score from a baseline of 19.3 to 27.2. Statistics Addition reached 25.2, and Cite Sources reached 24.6. The authors report visibility gains of up to 40%, and up to 37% in a test on Perplexity.ai.

The GEO paper also found that keyword stuffing, a classic SEO tactic, did not help. Keyword Stuffing scored 17.7 against the 19.3 baseline on the position-adjusted word-count metric. The authors also found that results varied by domain. No single tactic worked best for every topic.

The GEO paper is a controlled experiment, but its setup was simplified. The researchers used GPT-3.5-turbo to answer from the top five Google results. Commercial AI systems in 2026 work differently and change often. The paper shows which kinds of content edits can increase visibility. It does not show how any specific platform ranks sources today.

How does GEO relate to AEO, LLMO, and "AI search optimization"?

GEO, AEO (Answer Engine Optimization), LLMO (Large Language Model Optimization), and "AI search optimization" are overlapping labels for the same goal: visibility inside AI-generated answers. Google's guide "Optimizing your website for generative AI features on Google Search" (May 2026) defines AEO and GEO by name. It then states that, from Google Search's perspective, this work "is optimizing for the search experience, and thus still SEO."

AEO is an older industry term, linked to featured snippets and voice assistants. No single originator for AEO or LLMO could be verified. GEO is the only term in this group with a clear, citable origin: the 2023 Aggarwal et al. paper. Kiseo uses "GEO" and "AI visibility" as plain-language terms for this work. Kiseo did not invent either concept.

What does GEO involve in practice?

GEO starts with technical access. AI search systems can only cite pages they can reach. OpenAI says site owners should allow OAI-SearchBot for pages to appear in ChatGPT search. Perplexity says the same about PerplexityBot. Anthropic says blocking Claude-SearchBot "may reduce your site's visibility and accuracy in user search results." Google's AI Overviews and AI Mode rely on normal Googlebot crawling and indexing.

GEO depends on answer-first, specific content. Google's May 2026 guide asks for valuable, unique, "non-commodity" content. The GEO paper found that clear statistics, named sources, and quotations improve visibility. A page that answers a real customer question directly, with checkable facts, serves both people and AI systems.

GEO depends heavily on third-party presence. Ahrefs' study of 75,000 brands found that YouTube mentions (correlation about 0.737) and branded web mentions (about 0.66) tracked AI visibility far more closely than backlinks (about 0.218). For an independent business, that means reviews, directory listings, local press, and mentions on trusted sites carry as much weight as the business's own website, and often more.

GEO for local businesses runs through Google Business Profile and reviews. Google's generative-AI guide includes a section on optimizing local business details. Practitioners widely see review volume and rating as gates for local AI recommendations. No published study has set a universal threshold, so any review count should be treated as a direction, not a rule.

What does GEO not involve?

GEO does not require an llms.txt file. Google's May 2026 guide says site owners can ignore "unnecessary AI text files (like llms.txt)," and Ahrefs' 2026 study of 137,000 sites found that 97% of llms.txt files received zero requests. GEO also does not require special schema. Google says "there's no special schema.org markup you need to add" for generative AI search. In August 2023 the Google Search Central Blog limited FAQ rich results to "well-known, authoritative government and health websites," and Search Engine Journal reports that FAQ rich results stopped appearing in Google Search altogether on May 7, 2026.

How long does GEO take?

GEO results take weeks to months, not days. In Kiseo's practitioner experience, technical fixes show within about 2–4 weeks, first citations usually take 4–12 weeks, and visibility compounds over 3–6 months. These ranges come from practice, not from a published study. AI answers also change constantly. Ahrefs found that Google AI Overview content changes about every 2.15 days.

Key takeaways

  • GEO means earning citations and recommendations inside AI-generated answers.
  • Aggarwal et al. coined GEO in an arXiv paper (November 2023) presented at KDD 2024.
  • The GEO paper found that adding quotations, statistics, and cited sources boosted visibility by up to 40%. Keyword stuffing did not help.
  • Google treats AEO and GEO as "still SEO" for Google Search.
  • Third-party mentions, reviews, and Google Business Profile matter as much as owned content.
  • llms.txt and special schema are not meaningful GEO levers.

Questions and answers

Is GEO the same as SEO?

GEO overlaps heavily with SEO, and Google calls it "still SEO" for Google Search. GEO puts more weight on being cited inside answers, on third-party mentions, and on platforms beyond Google, such as ChatGPT and Perplexity.

Who invented GEO?

Pranjal Aggarwal, Vishvak Murahari, Tanmay Rajpurohit, Ashwin Kalyan, Karthik Narasimhan, and Ameet Deshpande introduced the term in "GEO: Generative Engine Optimization," posted to arXiv on November 16, 2023 and presented at KDD 2024.

Does a small business need GEO?

A small business benefits from GEO when customers research its category in AI tools before they buy. Informational and pre-purchase questions are the most realistic wins for independent businesses.

Do I need an llms.txt file for GEO?

An llms.txt file is not needed. Google says it does not use llms.txt for Search, and Ahrefs found that 97% of llms.txt files were never requested.

Can GEO guarantee that ChatGPT recommends my business?

GEO cannot guarantee any citation, ranking, or recommendation. AI answers are probabilistic and change often. GEO improves the odds by making a business easier to find, verify, and cite.

Sources

Kiseo helps independent businesses build AI visibility on these foundations.

See Solutions
Next in Learn

What is the difference between SEO and GEO?

Read article

Ahrefs' AI Benchmark Report

Read article