Key points
- Generative engine optimisation (GEO) is the practice of making content more likely to be selected, quoted and cited by AI systems that write answers, such as ChatGPT, Perplexity, Gemini and Google’s AI Overviews.
- The term comes from a 2023 research paper which found that adding citations, quotations and statistics could raise a source’s visibility in AI answers by up to 40%.
- GEO builds on SEO rather than replacing it: pages still need to be crawlable and indexed. The difference is the goal, being cited inside an answer rather than ranked in a list.
Search engine optimisation (SEO) is about ranking a page in a list of results. Generative engine optimisation (GEO) is about getting a page used and cited when an AI system writes an answer. The two overlap heavily, but they reward slightly different things. This article is part of our AI search visibility guide.
Where the term comes from
The phrase was introduced in a paper titled “GEO: Generative Engine Optimization”, first published in November 2023 by researchers including Pranjal Aggarwal, Karthik Narasimhan and Ameet Deshpande (arXiv). The authors built a benchmark of queries across many subject areas, called GEO-bench, and tested how changes to a source’s text affected how prominently a generative engine used it. They reported that the right changes could increase visibility in AI responses by up to 40%, and that what works varies by subject.
The methods that helped were the ones that make a source more credible and quotable: citing sources, including quotations from relevant authorities and adding specific statistics. Stuffing keywords, a classic SEO tactic, did little.
GEO and SEO compared
| SEO | GEO | |
|---|---|---|
| Goal | Rank high in a list of links | Be selected, quoted and cited in an AI-written answer |
| Unit of success | A page ranking for a query | A passage or fact being used, with a link |
| Main signals | Relevance, links, page experience | Clarity, specific facts, sources, author and publisher credibility |
| Measurement | Rankings, clicks, impressions | Citations in AI answers, AI crawler visits, referrals from assistants |
| Foundations | Both need crawlable, indexable pages and useful, original content | |
Google says there are no additional requirements to appear in its AI Overviews or AI Mode beyond being eligible to appear in Search (Google Search Central). So GEO is not a separate technical system to set up. It is a way of writing and structuring content so that, once a page is found, an AI system can lift a clear, accurate, attributable answer from it.
What GEO looks like in practice
- Answer first. Put a direct, one- or two-sentence answer near the top, then explain. AI systems often take the clearest self-contained passage.
- Be specific. Dates, numbers, names and places are easier to cite than general statements.
- Cite your sources inline. Link each important claim to its primary source. This is the change the GEO research found most consistently useful.
- Name a real author. Show who wrote the piece and why they are qualified, with a profile that can be checked.
- Structure the page. Use descriptive headings, short paragraphs, lists and tables, and an FAQ for common questions.
- Keep facts consistent. If your site, your structured data and other sites describe your organisation differently, AI systems have to guess which is right.
- Do not neglect access. None of this matters if AI crawlers cannot reach the page. See our guide to AI crawlers.
What GEO is not
GEO is not a way to trick AI systems. Hidden instructions aimed at AI models, invented statistics or fake reviews may work briefly but risk being treated as manipulation, and they damage trust with readers. Several so-called GEO tactics, such as publishing an llms.txt file, are also unproven; our llms.txt explainer looks at the evidence.
A worked example
NI Geovanta, the WordPress plugin Newsinterpretation uses, turns these practices into a score out of 100 built from eight signals: answer readiness, extractability, entity clarity, authority attribution, semantic structure, hallucination risk, content depth and citation quality. Scores like this are a checklist, not a guarantee of being cited, but they make the gaps visible. On our own site, only 27 of about 1,670 posts were rated compliant when we first ran it, which tells us where to start.
Frequently asked questions
What is generative engine optimisation (GEO)?
GEO is the practice of shaping content so that AI systems which write answers, such as ChatGPT, Perplexity and Google’s AI Overviews, are more likely to select, quote and cite it.
Is GEO replacing SEO?
No. GEO builds on SEO. Pages still need to be crawlable and indexed; GEO focuses on making them clear, specific and credible enough to be cited inside AI answers.
What does the research say works for GEO?
The 2023 GEO paper found that adding citations, quotations from relevant sources and statistics increased a source’s visibility in generative engine answers by up to 40%, while keyword stuffing did little.
Do I need special code for GEO?
No. Google says no special structured data or optimisation is needed for its AI features. Accurate structured data, clear writing and crawl access are what matter.
Last reviewed: 22 October 2026. Sources are linked inline.
Disclosure: NI Geovanta is developed by Newsinterpretation, the publisher of this article. It is used here as a worked example; other tools can perform similar checks.
