SEO, AEO and GEO: what each one covers and where to start

The short answer

SEO, AEO and GEO are three layers of one job. SEO gets a page crawled, indexed and ranked. AEO shapes the page so a search system can lift a clean answer from it. GEO is about being retrieved and cited when an AI system writes the answer itself, in Google’s AI Overviews and AI Mode, ChatGPT, Gemini, Perplexity or Copilot.

Start with SEO. Google’s own documentation says a page must be indexed and eligible to show with a snippet before it can appear as a supporting link in AI Overviews or AI Mode. A page that fails the first layer cannot reach the other two.

What does SEO actually cover?

Google’s SEO Starter Guide describes SEO as helping search engines understand your content and helping people decide whether to visit. The same guide says there are no secrets that will rank a site first automatically.

Google describes Search in three stages: crawling, indexing and serving results. Its in-depth guide is explicit that not every page makes it through each stage and that Google does not guarantee it will crawl, index or serve any page. SEO is the work of getting through those stages:

• URLs Googlebot can reach, returning a successful status code
• content that renders and can be indexed
• pages that are worth indexing in the first place
• links and relevance that earn a position for a query

None of this is new, and none of it has been replaced. It has become the entry ticket for everything that follows.

What is AEO?

AEO (answer engine optimisation) is an industry term; Google does not use it. It predates generative AI: it grew out of featured snippets and voice assistants, where the engine reads out or displays one passage as the answer.

Google’s featured snippets page shows how little of this can be forced. Its systems decide whether a page would make a good featured snippet. There is no markup that requests one. The only controls are negative: nosnippet, data-nosnippet and a lower max-snippet value.

So AEO is information design. A page earns answer placements when a system can find a self-contained answer quickly:

• the question as a heading
• the direct answer in the first one or two sentences under it
• definitions that make sense when quoted alone
• tables for comparisons and numbered steps for procedures

Question-shaped content is also where AI answers appear most. Pew Research Center’s analysis of 68,879 Google searches by 900 US adults in March 2025 found that 18% of all searches produced an AI summary. Searches phrased as questions produced one 60% of the time, and searches of ten words or more did so 53% of the time, against 8% for one or two word searches.

What is GEO?

GEO (generative engine optimisation) was named in a research paper by Aggarwal and colleagues, first posted to arXiv in November 2023 and accepted at KDD 2024. It framed the problem as making content more visible inside responses written by generative engines, and reported visibility gains of up to 40% on the authors’ own benchmark, GEO-bench.

Later work has narrowed that claim. A critical survey of 45 GEO studies posted in July 2026 argues that the original gains hold only once a source is already in the model’s context. In other words, rewriting helped pages that had already been retrieved; it did not show that rewriting makes a page easier to retrieve. The same survey finds no reviewed technique with a stable, long-term causal effect across platforms.

What does move citations in controlled tests is narrower and more ordinary. The “What Gets Cited” study (May 2026) ran 252,000 trials across six language models, placing two competing sources in the model’s context and recording which one it cited first. Topical relevance and list position had the largest effects. Explicit prices and a recent timestamp helped consistently. Formatting-only edits had little effect.

GEO also reaches beyond Google. Each AI engine has its own crawler and retrieval stack. OpenAI’s crawler documentation recommends allowing OAI-SearchBot in robots.txt for a site to appear in ChatGPT search answers, and says sites that opt out will not be shown there except as navigational links. A site that blocks an engine’s search crawler has opted out of being cited by it.

How do the three layers depend on each other?

For Google, the dependency is documented. Google’s May 2025 explainer on AI Overviews and AI Mode says AI Overviews use a customised Gemini model that works with Google’s existing quality and ranking systems, and that they are built to show information backed up by top web results. Its AI features page adds that there are no additional requirements to appear in AI Overviews or AI Mode: no special markup, no new machine-readable files and no AI text files.

That page also explains why one page can be cited for questions it never targeted. Both features may use query fan-out, issuing several related searches across subtopics and then showing supporting pages from those searches. A page ranking well for a sub-question can be pulled into the answer to a broader one.

How do SEO, AEO and GEO compare?

	SEO	AEO	GEO
Question it answers	Can search find and rank this page?	Can a system lift a clean answer from it?	Is the page retrieved and cited when AI writes the answer?
Where it shows	Classic results	Featured snippets, voice answers, the text of AI answers	AI Overviews, AI Mode, ChatGPT, Gemini, Perplexity, Copilot
What it needs	Crawlable, indexable, useful pages; links	Question headings, direct answers, standalone definitions, tables	Crawler access per engine, topical match, specifics such as prices, dates and evidence
Who defines it	Search engine documentation	Industry practice	Research papers and industry practice
How to measure	Search Console indexing and performance reports	Which queries show your passage	Citation checks on a fixed prompt set; AI referral traffic
Evidence base	Decades of documentation	Featured snippet behaviour	A few controlled studies, many vendor claims

One measurement gap matters here. Google’s AI features page says traffic from AI features is counted inside the Web search type in the Search Console Performance report. There is no separate AI Overviews line, so GEO on Google has to be inferred from query patterns and from manual checks.

Does being cited matter if clicks are falling?

Yes, on the evidence available, though the effect is correlational. Seer Interactive’s April 2026 update covered 53 brands, 5.47 million queries and 2.43 billion organic impressions from January 2025 to February 2026. On informational queries, average organic CTR across the year was 2.07% when the brand was cited in the AI Overview, 0.94% when it was not cited and 3.35% when no Overview appeared.

Seer states plainly that it cannot claim causation, since higher-authority brands are also more likely to be cited. Pew’s panel points the same way on clicks overall: users clicked a traditional result on 8% of visits with an AI summary and 15% without one.

Being cited does not restore the clicks an AI answer takes away. It does appear to lose fewer of them.

What is still uncertain?

Three things. AEO and GEO are industry labels with no fixed definition, so vendors use them to mean different things. GEO studies disagree on which tactics work outside lab conditions. And most CTR studies come from agencies and tool vendors analysing their own client data, which is useful but not neutral.

What to do

1. Fix indexing first. Check the Search Console page indexing report and resolve anything stopping important pages from being crawled or indexed.
2. Check every engine’s search crawler in robots.txt and your CDN or firewall settings: Googlebot, Bingbot and OAI-SearchBot at minimum.
3. Rewrite your most important pages answer first: a question heading, a two-sentence answer, then the detail.
4. Add the specifics AI systems were shown to prefer in controlled tests: exact prices, dates, sources and evidence. Show an updated date only when the content has really changed.
5. Pick 20 to 50 real questions your buyers ask and record, monthly, which engines cite you for them. Treat that as your GEO baseline.
6. Ignore offers of special AI files or markup for Google. Google says none are needed.