Answer engine optimization (AEO): what it is and what gets a page cited

Answer engine optimization (AEO) is the work of getting a web page quoted as a source when an AI system such as ChatGPT, Google's AI Overviews or Perplexity answers a question. Search engine optimization competes for a place in a list of links. Answer engine optimization competes to be one of the handful of sources the answer actually names.

(If you arrived looking for AEO in customs and trade, that's the Authorized Economic Operator program, a different thing entirely.)

The reason anyone cares: the click has moved. About two-thirds of US Google searches ended without a click to any website in early 2026, according to SparkToro's June 2026 analysis (the zero-click article covers the numbers and their caveats). And a page can be cited by an AI engine without ranking anywhere near the top of Google, which means ranking well no longer guarantees you're part of the answer.

Which engines does AEO cover?

Any system that composes an answer and cites sources. In practice, for a content site in the US:

ChatGPT, which cites sources when its search tool runs. How ChatGPT picks sources.
Google AI Overviews, the generated summary above the results. Since late August 2026 Google expands some Overviews into full AI Mode-style answers by default. The AI Overviews guide.
Google AI Mode, the conversational search tab, which Search Console now reports alongside AI Overviews.
Perplexity, which cites on nearly every answer. The Perplexity guide.
Gemini, Google's standalone assistant. The Gemini guide.
Claude, when web search is on. The Claude guide.
Microsoft Copilot, which draws on Bing.

Ask any of them "what's the best AI video editor for a solo founder?" and you don't get ten links. The engine searches, reads a set of candidate pages, and writes an answer that credits a few of them. AEO is the work of being one of the few.

Being cited is not the same as ranking

Two studies make the gap concrete.

AirOps analyzed 548,534 pages that ChatGPT retrieved across 15,000 prompts (March 2026). 55.8% of the pages it cited ranked in Google's top 20 for the prompt or one of its fan-out queries, so roughly 44% did not. Only 15% of retrieved pages were cited at all: being fetched is not being quoted.

Ahrefs looked at 863,000 keywords and about 4 million AI Overview URLs and found 37.9% of cited pages ranked in Google's top 10, down from 76% seven months earlier. Ahrefs attributes part of that drop to better citation detection on its side, not only to Google changing behavior, so read it as "the overlap is smaller than people assumed" rather than as a precise trend.

If your rankings are healthy and your citations are not, the problem is usually how the page is built, not how much authority it has. That's the part AEO addresses.

How a page becomes a citation

Every engine runs some version of the same three steps. Knowing which one you're failing tells you more than any single tactic.

Rank well but never get cited? You're being dropped at filtering. Not ranking at all? No amount of filtering work will help until retrieval is fixed.

Where on the page citations come from

Kevin Indig's Growth Memo analysis of 1.2 million ChatGPT answers and 18,012 verified citations (reported by Search Engine Land, February 2026) found citations are front-loaded: 44.2% came from the first 30% of a page's text, 31.1% from the middle and 24.7% from the final 30%. The study covers ChatGPT only, but the shape is intuitive for any engine that reads until it has an answer.

The practical consequence is cheap: put the answer near the top, and do it per section, not just per article. A reader who scrolls will find your depth anyway. An engine that finds an answer in someone else's opening paragraph may never get to yours.

Weak opening: "AEO is a complex topic that has evolved rapidly. In this article, we'll dive deep into what it means and why it matters for modern marketers."

Strong opening: "Answer engine optimization is the practice of structuring content so AI engines cite it as a source. It differs from SEO by prioritizing extractability over ranking."

How AEO differs from SEO

They share a foundation (crawlable pages, real quality, technical health) and diverge above it.

Factor	SEO	AEO
Backlinks	A core authority signal	Weaker than brand mentions (see below)
Answer in the opening	Optional; helps featured snippets	Central: the opening is what gets extracted
Structured data	Produces rich results for supported types	Helps machines parse; not a documented citation input
Q&A structure	Helps with People Also Ask	The cleanest citable unit there is
Named author	Part of E-E-A-T	Anonymous content appears to be filtered harder
Length	Comprehensive pages tend to rank	Self-contained chunks matter more than total length
Freshness	Matters for some queries	Matters broadly (see below)

In short, SEO rewards comprehensiveness and AEO rewards extractability. A 4,000-word guide can rank first and never be cited if nothing in it can be quoted without the paragraphs around it.

If you're deciding where a small site should spend its first hours, AEO vs SEO: what to do first on a small site works through that.

Free tool · No signup

Free AEO Readiness Score

Paste content or a URL → a 0-100 score across ten AEO signals, with the fixes worth doing first.

Score your content →

Which factors have evidence behind them?

Ordered by the strength of the evidence, which is not the order of how confidently they're usually asserted.

Cite your own sources

The strongest family of tactics in the original research. The GEO paper (Aggarwal et al., KDD 2024) tested nine content changes on a 10,000-query benchmark; Cite Sources, Quotation Addition and Statistics Addition came out on top, with visibility gains of up to 40%.

The direction still makes sense: a page that points to checkable sources is easier to trust and to quote. It only works with real attribution. A made-up statistic lends nothing and costs everything the first time someone checks.

Keep content fresh, by actually updating it

Ahrefs analyzed about 17 million cited URLs across ChatGPT, Perplexity, Gemini, Copilot and AI Overviews: AI-cited content was 25.7% fresher than organic results for the same queries (an average of 1,064 days old against 1,432).

Seer Interactive's 2026 study of 47,097 citations across ChatGPT, Gemini and Perplexity found 75% of cited pages had been updated within the past year and 88% within two. Its sharper finding is where that freshness comes from: 72% of cited pages were fresh by last-update date, but only 42% by publish date. In Seer's words, "the freshness LLMs reward is being manufactured by updates, not by new publishing."

So the practice is to change the visible content (new figures, a rewritten opening, a section that's no longer true removed), not just the date in the markup.

Build brand signals, not only backlinks

Ahrefs studied 75,000 brands and measured which signals correlate with visibility in AI Overviews:

Mentions correlate about three times more strongly than backlinks. For a small site, getting talked about in your niche may be worth more than getting linked from it, which is part of why build-in-public suits AEO.

Write in extractable units

A question with its answer is the cleanest thing an engine can lift: self-contained, labeled, unambiguous. So is a table row, a one-sentence definition, a short numbered list. A practical target is a block of roughly 40 to 120 words that makes sense with nothing above it.

FAQ rich results no longer exist, though. Google stopped showing them on May 7, 2026, dropped the report and Rich Results Test support in June, and removed Search Console API support in August. A visible FAQ section is still useful for extraction; the search decoration is gone.

Name your author

Anonymous content appears to be filtered harder, particularly on money and health topics. This is practitioner consensus, not a measured result; nobody has published a clean experiment. What to add: a named author with a bio page and Person schema pointing to it, experience in the prose where it's real, and data you gathered yourself, however small. Details in Article schema and E-E-A-T.

What does schema do for AEO?

Less than it's sold as. Google says there are no additional requirements to appear in AI Overviews or AI Mode, and no special markup or "AI text files" are needed. Schema helps machines understand a page; it isn't a documented citation input.

Markup that still produces rich results includes Article, Organization, Person, BreadcrumbList, Product, Review, Recipe, Video and LocalBusiness. Retired since 2023: HowTo, Sitelinks search box, seven types in June 2025 (including Course Info, Claim Review and Estimated Salary), Practice problems, and FAQ. The rule that hasn't changed: markup must match what's visible on the page. More in schema markup for AI Overviews.

Can AI crawlers reach your site?

This is the most common self-inflicted wound in AEO, and it takes two minutes to check. AI crawlers come in two classes that do very different things.

A few specifics that guides often get wrong:

Blocking GPTBot does not remove you from ChatGPT's answers. Blocking OAI-SearchBot does.
Google-Extended is a robots.txt token, not a separate crawler. Google says it does not affect inclusion or ranking in Google Search, so it has no effect on AI Overviews. It does cover grounding in the Gemini app, so blocking it can cost you citations there.
The user-triggered fetchers behave differently. Anthropic says blocking Claude-User may reduce your visibility in user-directed search; OpenAI and Perplexity say robots.txt may not apply to theirs.
Since August 31, 2026, Search Console has a Search generative AI control that removes a site from AI Overviews, AI Mode and Discover's AI features without affecting regular ranking. It's the only Google-side way to leave those features while staying in Search, which also means it's one more setting to check isn't switched on by accident.

Most accidental blocks come from "block AI crawlers" templates written in 2024, before the two classes were widely understood, or from CDN bot settings nobody remembers enabling.

Free tool · No signup

Free AI Crawler Checker

Paste a domain → see which AI crawlers your robots.txt allows and which it blocks, with training and retrieval bots told apart.

Check your crawlers →

Where to start

Confirm you're retrievable. robots.txt, the CDN's bot settings and the Search Console AI control, for every engine you care about.
Rewrite five openings. Take five pages that already get search traffic and make their first 50 to 80 words answer the page's question completely.
Attribute every number. Each statistic gets a named, linked source. Delete the ones you can't source.
Measure your own citation rate. Ten to twenty prompts, run across engines every week, recorded rather than remembered. Published findings replicate unevenly; your own pages are the only benchmark that's definitely relevant.

Pages that already rank have cleared retrieval, so filtering work has something to act on. The full pre-publish list is in the AEO content checklist.

FAQ

Is AEO the same thing as GEO?

For day-to-day work, yes. GEO (generative engine optimization) is the name from a 2023 research paper; AEO (answer engine optimization) is older and came out of featured-snippet and voice-search work. Both mean getting content cited in AI-generated answers, and the checklist is the same. The GEO name carries a research record that is worth reading critically.

How long until AEO changes show up?

There is no reliable published figure. A change has to be re-crawled before any engine can see it, which takes days to weeks depending on the site, and citation sets vary from run to run, so you need several weeks of repeated sampling before a difference is distinguishable from noise. Anyone promising results within days is describing a re-crawl, not a change in citation behavior.

Which AI engine should I measure first?

Perplexity is the easiest to start with because it cites on nearly every answer, so a citation rate becomes readable quickly. Google AI Overviews reach more people but vary by session and location, which makes manual sampling noisier. Start where you can measure, then add engines.

Sources

Updated October 1, 2026: added AI Mode and the Search Console AI control, corrected the Google-Extended and freshness claims, and linked every source.