How do you get cited in Google AI Overviews?
Rank well in Google organic first — it remains the strongest single predictor, though far weaker than it was — then structure content so individual passages answer questions on their own. AI Overviews are built on Google's own index, so unlike ChatGPT or Perplexity they still correlate meaningfully with ranking position. That correlation has been falling fast.
In mid-2025, roughly 76% of AI Overview citations came from pages ranking in Google's top 10. By early 2026 that figure had dropped to around 38% in Ahrefs' data, with some studies putting it lower still.
Why the correlation is weakening
The cause appears to be query fan-out. Rather than answering the question as typed, Google decomposes it into several related sub-questions, retrieves for each, and assembles an answer. A page that ranks nowhere for the original query can be the best available answer to one of the sub-questions — and get cited on that basis.
Two practical consequences:
Ranking is necessary less often than it used to be. There is a real path to citation for pages that don't rank on page one for the head term.
Breadth beats depth on a single term. Covering the adjacent questions around your topic gives you more chances to be retrieved for one of the sub-queries. This is a genuine change in how to plan content.
AI Overviews and AI Mode are not the same thing
This trips up nearly everyone, including people who track it professionally.
AI Overviews are the summary block on a standard results page. AI Mode is Google's separate conversational search surface. They frequently reach the same conclusion — semantic agreement has been measured around 86% — but cite the same URLs only about 13.7% of the time.
Google, in effect, disagrees with itself about which sources support the same answer. And AI Mode behaves much more like the standalone assistants: only a small fraction of its citations come from Google's organic top 10, versus a much larger share for AI Overviews.
If you're measuring, measure them separately. A tool reporting a single "Google AI" number is blending two surfaces that share almost no citations.
What AI Overviews draw on
Google's AI surfaces pull noticeably more from user-generated and video content than ChatGPT does. YouTube is structurally important. Reddit features heavily. Pew Research found the most-cited sources across AI summaries skew toward Wikipedia, YouTube and Reddit, with government sources appearing at roughly three times their rate in standard results.
AI Overviews also show the strongest preference for brand domains of the major engines — one analysis put brand domains at around 60% of citations, against much lower shares for ChatGPT and Perplexity — and provide the most clickable links. If your buyers are on Google, this is the AI surface where being cited is most likely to produce an actual visit.
The counterweight: AI Overviews don't appear on every query. Coverage estimates vary considerably by study and by vertical, so check your own terms rather than assuming.
What to do
1. Keep ranking. Weaker predictor than before, still the strongest one available for this surface specifically. Ordinary SEO is not optional here.
2. Answer directly, near the top. A self-contained forty-to-sixty-word answer under a descriptive heading. This is the same structure that wins featured snippets, and the signals overlap — AEO covers it in depth.
3. Cover the sub-questions. Take your main topic and write out the five to eight adjacent questions a searcher would ask around it. Each is a fan-out retrieval opportunity. Give each its own heading and direct answer.
4. Use real tables and lists. Tables are disproportionately extracted. Real <table> markup, not styled divs.
5. Get your entity signals consistent. Organization markup that agrees with your external profiles. Note that Google's guidance states no special markup is required for AI Overviews and that structured data should match visible text — schema helps entity resolution, it isn't an AI shortcut. The current schema picture.
6. Consider video. YouTube's presence in Google's AI surfaces is structural rather than incidental. If your category suits it, video is an underused route into this specific surface.
Measuring it
Search Console does not break out AI Overview appearances separately, which makes this harder than it should be.
Available proxies:
- Impressions rising while clicks stay flat on informational terms — the classic zero-click signature
- Manual checking of your priority terms, recorded on a schedule
- Third-party AI visibility tools, provided they separate AI Overviews from AI Mode
Set a baseline before you start work. Without one, a change in this surface is indistinguishable from ordinary ranking movement.
The zero-click question
Being cited in an AI Overview frequently means the user never clicks. That's real, and it's the honest cost of this surface.
Two reasons to pursue it anyway. AI Overviews link more readily than the other engines, so the click rate isn't zero. And absence has a cost of its own — if a competitor is cited in a block sitting above the organic results, your position 3 listing has been pushed below something that already answered the question.
Frequently asked questions
How do I rank in Google AI Overviews? Rank well organically first — it's still the strongest predictor for this surface — then structure content with direct, self-contained answers under descriptive headings, and cover the adjacent sub-questions Google's query fan-out retrieves for.
Do AI Overviews use my Google rankings? Partly, and less than before. Around 76% of AI Overview citations came from top-10 pages in mid-2025; by early 2026 that had fallen to roughly 38%. Query fan-out means pages that don't rank for the main term can be cited for a sub-question.
What's the difference between AI Overviews and AI Mode? AI Overviews is the summary block on a normal results page; AI Mode is Google's separate conversational search surface. They agree semantically around 86% of the time but cite the same URLs only about 13.7% of the time.
Does schema markup help with AI Overviews? Google states no special markup is required, and that structured data should match visible text. Clean markup helps entity resolution but is not an AI Overview shortcut.
Do AI Overviews reduce my traffic? On informational queries, often yes — the answer appears without a click. The signature is impressions holding steady or rising while clicks fall. AI Overviews do link more readily than other AI surfaces.
Can I opt out of AI Overviews? The nosnippet and max-snippet robots directives limit how your content can be used in snippets and AI Overviews, but applying them also restricts ordinary search snippets. It's a blunt instrument with a real cost.
Target the answer, not just the ranking
AI Overview Targeting restructures your priority pages for the surfaces that share signals — AI Overviews, featured snippets and People Also Ask — including the sub-question coverage that query fan-out rewards.
Entity & Schema Optimization aligns your Organization markup with your external entity footprint, which is what lets Google resolve who you are consistently across both AI surfaces.