Your page ranks third for a query you spent two years earning. This morning someone searched it, read a paragraph Google wrote, glanced at three citation cards from sites you have never worried about — and left. Your ranking never moved. Your click never happened.
That paragraph was an AI Overview, and it now sits on top of a growing share of the queries you care about. You cannot opt out. What you control is whether you are one of the sources inside it.
This guide covers what the 2026 data says about how Google chooses those sources — and a six-step playbook for becoming one of them.
Why should I care?
Fair question — SEO has survived a dozen "this changes everything" moments. Two numbers make this one different.
The first is scale. At Google I/O in May 2026, Sundar Pichai announced that AI Overviews now has over 2.5 billion monthly active users. Fourteen months earlier, Google was quoting "more than a billion". That growth curve makes AI Overviews the highest-volume AI answer surface there is.
The second is clicks. Pew Research Center found that when an AI summary appeared on a results page, users clicked a traditional result on just 8% of visits, versus 15% when no summary appeared. Roughly half your expected clicks, gone. The full numbers get their own section later.
Put those together: the biggest AI surface in the market is eating your organic clicks, and it lives directly on top of your rankings. You cannot opt out. You can only show up in it. That is the discipline people now call AI Overviews SEO.
The core principle: extractable and trusted, wherever you rank
Google says there is no special optimization for AI Overviews — and the citation data says classic ranking is not the whole game either. The pages that win are extractable and trusted, wherever they rank.
The first half is Google's official position. Its AI features guide is blunt: "The best practices for SEO continue to be relevant because our generative AI features on Google Search are rooted in our core Search ranking and quality systems." The same guide states there are no additional requirements and no special optimizations for AI Overviews or AI Mode. Google's answer to "how do I rank in AI Overviews?" is, essentially, "do SEO."
The second half is what independent data adds: a majority of AI Overview citations now come from pages outside the top 10 — the numbers are two sections down. So being extractable (structured so a model can lift your answer cleanly) and being trusted (unambiguous about who you are and why you would know) decide who gets quoted from that wider pool. How to rank in AI Overviews is really two jobs: stay retrievable — the SEO you already do — and become quotable, which is what the rest of this guide is about.
What triggers an AI Overview?
Not every query gets one, and the share that do refuses to sit still.
Semrush, tracking more than 10 million keywords, watched the trigger rate go from 6.49% in January 2025 to a 24.61% peak in July 2025, then back down to 15.69% by November 2025 — "volatile growth" is their phrase for it. So treat any undated, single-number "X% of searches trigger an AI Overview" claim as a snapshot, not a fact.
The level matters less than the mix. In the same Semrush data, the intent profile of AIO-triggering queries flipped through 2025: informational queries fell from 91.3% of triggers in January to 57.1% by October, while commercial rose from 8.15% to 18.57% and transactional from 1.98% to 13.94%.
A follow-up Semrush study of 600,000+ keywords found the trend accelerating: between November 2025 and April 2026, AI Overview presence on commercial-intent queries grew another 71% (transactional presence fell 5%), with Finance queries growing 231.25%. In plain terms: AI Overviews started on homework questions and are moving onto the comparison queries where buying decisions get made — exactly where your brand needs to be the one quoted.
How AI Overviews choose citations (it's not classic ranking)
Google has explained the mechanics here — and independent studies have measured what those mechanics actually produce.
Start with Google's own words. Answers are grounded: Google's guide defines the "technique (also known as grounding) used to improve the quality, accuracy, and freshness of AI responses by relying on our core Search ranking systems to retrieve relevant, up-to-date web pages." (That is retrieval-augmented generation — the model writes from retrieved pages, not memory.)
Before it writes, the system also runs query fan-out: "a set of concurrent, related queries generated by the model to request more information and fetch additional relevant search results to address the user's query." (Your one search silently becomes many, each pulling its own results.)
Google states the consequence outright in its AI features documentation: "While responses are being generated, our advanced models identify more supporting web pages, allowing us to display a wider and more diverse set of helpful links associated with the response than with a classic web search."
How much wider? In March 2026, Ahrefs analyzed 863,000 keyword SERPs and four million AI Overview URLs and found that only 37.9% of pages cited in AI Overviews also rank in the top 10 for the same query — down from roughly 76% in their July 2025 analysis. One caveat Ahrefs raises itself: part of that drop reflects improved citation detection in their tooling, so read it as directional, not a clean like-for-like.
BrightEdge measured through a different lens in September 2025: overall overlap between AI Overview citations and organic rankings rose from 32.3% in May 2024 to 54.5% — yet only 16.7% of citations came from top-10 results, with Google preferentially sourcing pages ranked between positions 21 and 100.
Two studies, two methodologies, one synthesis: a top-10 ranking helps — the correlation is real, and by BrightEdge's measure strengthening — but it is neither necessary nor sufficient. Most citation slots go to pages the classic SERP would never show a user, and that gap is where the rest of this playbook lives. Nor is it unique to Google: the published research on which sources AI engines actually cite shows the pattern rhymes.
Step 1: Clear the eligibility bar
Everything starts with a binary check. Google's stated requirement, verbatim from its AI-features optimization guide: "To be eligible to be shown in generative AI features on Google Search, a page must be indexed and eligible to be shown in Google Search with a snippet, fulfilling the Search technical requirements."
Three things to verify this week:
- Indexed. Run your key pages through URL Inspection in Search Console. Not "discovered," not "crawled" — indexed.
- Snippet-eligible. The same preview controls that govern snippets govern AI features: nosnippet, data-nosnippet, max-snippet and noindex all apply. Audit for legacy tags — a forgotten max-snippet:0 quietly removes you from every AI surface Google runs.
- Actually retrievable. Content that only renders behind logins, heavy client-side scripts or interstitials cannot be fetched cleanly, so it cannot be grounded on.
Passing this bar buys you exactly one thing: candidacy. Candidacy is mandatory.
Step 2: Keep earning organic rankings
The "AI Overview vs SEO" debate frames them as rivals. The data reads more like a dependency. BrightEdge's overlap climbing from 32.3% to 54.5% says citations and rankings are converging, not diverging, and Google's line about generative features being "rooted in our core Search ranking and quality systems" explains why: same index, same quality signals, same retrieval.
Fan-out changes what "ranking" means, though: you now compete for the invisible cluster of subqueries the model generates. A page at position 40 for the head term but position 3 for a fanned-out subquestion is exactly the position-21-to-100 citation BrightEdge keeps finding. Practically:
- Cover the subquestions around your money topics as their own sections and pages, not throwaway paragraphs.
- Treat pages ranked 11–100 for target queries as citation candidates, not failures — improve them instead of writing them off.
- Keep doing the unglamorous work — internal linking, crawlability, earning links.
Step 3: Write answer-shaped content
A grounded model does not lift pages. It lifts passages. Answer-shaped content means the liftable passage already exists — the structure practitioners consistently report working:
- Direct answer first. Under every question-shaped heading, give the actual answer in the first two to three sentences, then add the depth. Not the reverse.
- Question-shaped H2s and H3s. Headings phrased the way people actually ask map neatly onto fanned-out queries.
- Lists and tables for anything enumerable. Steps, comparisons, prices, specs. Structure survives extraction; walls of prose get skipped.
- Self-contained sections. Each section should make sense alone — a retrieval system may only ever see that chunk.
None of this is a trick, which is why Google can endorse the spirit of it in its own guide: "Creating content that people find unique, compelling, and useful will likely influence your website's presence in generative AI search in the long run." It also pays twice: the same discipline is most of what wins in ranking in ChatGPT, which grounds its answers the same way through a different index.
See what AI says about your brand
Run the free audit and see what AI assistants tell buyers about businesses like yours.
Step 4: Add schema markup (in code, not screenshots)
Schema markup (structured data — machine-readable JSON-LD labels describing what your page is) carries no Google promise about AI Overviews specifically, so let's not pretend otherwise. What it verifiably does: qualifies pages for rich results and states without ambiguity who wrote what, when, for which organization. In a system deciding at retrieval time what to trust, ambiguity is a cost you do not need to pay.
A surprising number of the guides ranking for this exact topic present their schema examples as screenshots — images of code, published to teach machines to read text. Here is one you can select, copy and edit:
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Article",
"headline": "How to Appear in Google AI Overviews",
"author": {
"@type": "Person",
"name": "Jane Example",
"jobTitle": "Head of SEO",
"url": "https://www.example.com/authors/jane-example"
},
"publisher": {
"@type": "Organization",
"name": "Example Co",
"url": "https://www.example.com"
},
"datePublished": "2026-08-01",
"dateModified": "2026-08-01",
"mainEntityOfPage": {
"@type": "WebPage",
"@id": "https://www.example.com/blog/ai-overviews-guide"
}
}
</script>Swap the values, keep the types, validate with Google's Rich Results Test. Priorities: Article (or BlogPosting) on editorial pages, Organization sitewide, FAQPage anywhere you genuinely answer recurring questions.
Step 5: Lock down entity consistency and E-E-A-T
Grounding retrieves pages, but trust attaches to entities — the people and organizations behind them. The pattern practitioners report: consistently described entities get cited; vague ones get passed over for a source the model can pin down. The hygiene list:
- Real bylines with credentials. A named author with a role, and a reason they would know, on every substantive page.
- Author and About pages. State who you are and what you do — in the same words your schema and LinkedIn use.
- One description of your organization, everywhere. Same name, positioning and details across your site, directories and profiles. Contradictions read as noise.
- sameAs links in your Organization schema connecting the entity to its official profiles.
None of this is a confirmed ranking-factor list — anyone selling it as one is guessing. It is cheaper than that: it removes reasons to doubt you.
Step 6: Track your AI Overview presence
Rank trackers were not built for this: position 3 looks healthy while the AI Overview above it quotes three competitors — invisible exactly where the reader is looking. Google AI Overviews tracking means answering three questions per target query: does an AI Overview appear, are you cited, and who is cited instead?
Start manually: a spreadsheet, 20–50 target queries, checked weekly — our 10-minute AI visibility audit walks through the method. When the list outgrows the spreadsheet, that is what an AI visibility platform is for: Apex tracks prompts across AI Overviews, ChatGPT, Gemini, Perplexity and Copilot and turns the raw answers into share of voice and citation trends. Either way, run the free AI visibility check first — a baseline turns everything above from theory into a scoreboard.
What AI Overviews do to your click-through rate
Now the numbers that reframe your traffic reports.
Pew Research Center studied 900 U.S. adults across 68,879 searches in March 2025. On visits where an AI summary appeared, users clicked a traditional result 8% of the time, versus 15% without one. They clicked a link inside the AI summary on just 1% of those visits. And they ended their browsing session entirely on 26% of AI-summary visits, versus 16% otherwise. Google publicly disputed the study — worth knowing, though the independent data below points the same way. Summaries appeared on 18% of sampled searches at the time; coverage has been volatile since.
In February 2026, Ahrefs measured it longitudinally: across 300,000 keywords with Search Console data, an AI Overview's presence now correlates with a 58% lower click-through rate for the top-ranking page. Position-1 CTR fell from 0.073 in December 2023 to 0.016 in December 2025, against a forecast of 0.037 had AI Overviews not existed. Their April 2025 version of the same analysis had measured a 34.5% reduction — the correlation roughly doubled in ten months.
Read it straight: this is not "SEO is dead" — it is the impression migrating from the blue link into the answer box. The citation card is the new above-the-fold; the clicks that survive skew higher-intent, the pattern practitioners consistently report. Be present where the impression now happens.
Why it gets confusing: AI Overviews vs AI Mode vs Gemini
Google ships three AI surfaces with overlapping names, and plenty of teams are optimizing for the wrong one. One test separates them: did you have to go somewhere, or did it come to you? An AI Overview appears automatically at the top of a normal results page — you never asked for it. AI Mode is a tab inside Google Search that you deliberately switch to. Gemini is a separate standalone app that is not Search at all.
| Surface | How you reach it | Scale |
|---|---|---|
| AI Overviews | Appears automatically at the top of a normal results page for some queries | 2.5B+ monthly users (May 2026) |
| AI Mode | A tab inside Google Search you deliberately switch into | 1B+ monthly users in its first year |
| Gemini | A standalone app — not Search at all | 950M monthly users (Q2 2026) |
The scale figures are Google's own: AI Overviews at 2.5 billion monthly users, AI Mode having "already surpassed 1 billion monthly active users" in just a year, the Gemini app at 950 million as of Q2 2026. They are related by design: Google introduced AI Mode in March 2025 as expanding "what AI Overviews can do with more advanced reasoning, thinking and multimodal capabilities."
Do not carve the distinction into stone, though. On the Q2 2026 earnings call, Pichai spoke about bringing "AI Overviews and AI Mode into one seamless Search experience." Quietly good news: extractable, trusted content is positioned for whatever the merged surface becomes.
Where to focus first
Six moves, ranked by leverage per hour:
- Run URL Inspection on your ten most important pages — eligibility is binary; one stray noindex or max-snippet:0 zeroes everything else here.
- Rewrite your money pages answer-first — a liftable two-to-three-sentence answer under a question-shaped heading is the highest-yield content change.
- Ship Article, Organization and FAQPage schema as real code — it removes ambiguity about who is saying what, at near-zero cost.
- Publish author and About pages — trust attaches to entities the system can pin down.
- Baseline your AI Overview presence this week — the 10-minute audit needs only a browser; you cannot grow a share of voice you never measured.
- Keep funding classic SEO — every AI feature Google runs is still, in its own words, rooted in core Search ranking.
Frequently asked questions
How do I show up in Google AI Overviews?
Make sure your page is indexed and eligible to appear in Google Search with a snippet, keep earning organic rankings for the topic, and structure your content so the answer is easy to extract: a direct answer in the opening sentences, question-shaped headings, lists and tables, plus schema markup and consistent author and organization details. Then track which of your target queries trigger AI Overviews and whether you are cited, and iterate on the pages that miss.
Do AI Overviews use the same ranking as Google Search?
Partly. Google says its generative AI features are rooted in its core Search ranking and quality systems, but the overlap with classic rankings is loose: Ahrefs found only about 38% of pages cited in AI Overviews also rank in the top 10 for the same query, and BrightEdge found Google frequently sources citations from positions 21 to 100. Ranking helps, but AI Overviews regularly cite pages the classic top 10 would never show.
What is the difference between AI Overviews and AI Mode?
AI Overviews appear automatically at the top of a normal Google results page for some queries. AI Mode is a tab inside Google Search that you deliberately switch to for a fuller conversational experience. Gemini is a standalone app that is not Search at all. The first two are converging: at its Q2 2026 earnings call, Google spoke about bringing AI Overviews and AI Mode into one seamless Search experience.