Google AI Overviews have rewritten the rules of search, and the AI overviews statistics below show exactly how fast and how far. Pulled from Ahrefs, Semrush, Pew Research, NP Digital and other 2025-2026 datasets, this page rounds up the most important data on how often AI overviews appear, what they do to click-through rates, which sources Google cites, and how users actually behave. Every number is attributed to its source so you can act on it with confidence.
What Are Google AI Overviews?
A Google AI Overview (AIO) is an AI-generated summary that appears at the very top of the search results page, above the traditional organic listings. Powered by Google's Gemini models, it answers a query in a few sentences and links out to a handful of cited sources, replacing the old "ten blue links" experience for many informational searches.
First launched at Google I/O in May 2024 as the successor to the Search Generative Experience (SGE), the feature scaled rapidly. By May 2025 Google had expanded AI Overviews to more than 200 countries and territories in over 40 languages, and by 2026 the feature reaches roughly 2 billion users per month (Google; Digiday). Below, we break the data into the sections that matter most for SEO and marketing decisions.
Key AI Overviews Statistics for 2026 (Top Picks)
If you only remember a handful of numbers, make it these. They capture the scale, the traffic impact, and the citation behaviour of AI overviews heading into 2026:
- ~2 billion people use AI Overviews every month globally (Google / Digiday).
- AI Overviews now appear on roughly 20-25% of US searches and up to ~50-55% by some broad-tracker methodologies (Semrush; Ahrefs; Pew Research).
- AIOs grew 115% since March 2025, and AIO keyword share roughly doubled in the same window (Ahrefs).
- Organic clicks fall ~34.5% on a keyword when an AI Overview appears (Ahrefs, 300,000+ keywords).
- ~99% of AIO-triggering keywords are informational in intent (Ahrefs).
- Reddit is the single most-cited domain in AI Overviews (Ahrefs; Semrush).
- Only about 8% of visits with an AI summary led to a click on a traditional link, versus 15% without one (Pew Research Center).
Adoption & Rollout Statistics
AI Overviews went from experiment to default at remarkable speed. The rollout statistics show both explosive growth and the periodic dialling-back as Google tunes when summaries appear:
- Launched May 2024 at Google I/O; expanded to 200+ countries and 40+ languages by May 2025 (Google).
- AIO prevalence rose from 6.49% of queries in January 2025 to a 24.61% peak in July 2025, then stabilised near 15-16% by November 2025 (Semrush, 10M+ keywords).
- 115% growth in AI Overviews since March 2025, with AIO keyword share roughly doubling (Ahrefs).
- Mobile AIO presence grew 474.9% year over year (seoClarity).
- AI Mode reached 75 million daily users by December 2025 and 100M+ monthly active users across the US and India (Google / Search Engine Journal; TechCrunch).
The headline takeaway: AI overviews are no longer a fringe feature. They are a permanent, still-expanding layer of the search results page, and the variation between studies (20% to 55%) is mostly about whether the dataset is US-only, global, branded, or informational-heavy.
How Often Do AI Overviews Appear? (By Query Type & Industry)
AI overview frequency is wildly uneven. Informational, question-style, and longer queries trigger them constantly; transactional and ecommerce queries almost never do. Here is how the AI overview trigger rate breaks down by industry and query characteristics, compiled from NP Digital, Semrush, and Ahrefs datasets:
| Segment | AI Overview trigger rate | Source |
|---|---|---|
| Health | 60.7% | NP Digital (8M queries) |
| Home & Garden | 50.4% | NP Digital |
| Education | ~48-52% | SE Ranking |
| Transportation | 31.4% | NP Digital |
| Food & Beverage | 24.9% | NP Digital |
| Finance | 22.9% | NP Digital |
| Insurance | 19.0% | NP Digital |
| Technology | 17.4% | NP Digital |
| Travel / Hospitality | 11.3% | NP Digital |
| Automotive | 5.8% | NP Digital |
| Real Estate | 5.2% | NP Digital |
| Ecommerce / Retail | 2-4% | Search Engine Land; SE Ranking |
| Local services queries | ~2-7% | NP Digital; Ahrefs |
Query shape matters as much as topic. The longer and more conversational the query, the more likely an AI overview is to appear (Ahrefs):
- 1-word queries: ~9.5% trigger an AIO; 5-word: 27.6%; 7+ words: 46.4% (Ahrefs).
- 8+ word queries are 7x more likely to surface an AI Overview (Search Engine Land).
- Technical / jargon-heavy queries are 48% more likely to get an AIO (Search Engine Land).
- About 99% of AIO-triggering keywords carry informational intent; commercial (~18.6%), transactional (~13.9%) and navigational (~10.3%) queries trigger them far less (Ahrefs; Semrush).
- Roughly 81% of AIO result pages also include a "People Also Ask" box (Ahrefs).
AI Overviews CTR & Traffic Impact Statistics
This is the section that keeps marketers up at night. When an AI overview appears, fewer people click through to the websites below it. The click-through rate (CTR) impact data is consistent across independent studies, even if the exact percentage varies by methodology:
| Metric | Impact | Source |
|---|---|---|
| Click drop when AIO appears (per keyword) | -34.5% | Ahrefs (300K+ keywords) |
| Industry-wide average CTR decline | -15.49% | Amsive (700K keywords) |
| AIO + featured snippet combined | -37.04% | Amsive |
| Non-branded keyword CTR drop | -19.98% | Amsive |
| Position #1 informational CTR (pre- vs with-AIO) | 5.6% → 3.1% (-44.6%) | Ahrefs |
| Clicks on a result when AIO present | 8% of visits | Pew Research Center |
| Clicks on a result with no AIO | 15% of visits | Pew Research Center |
| Branded query CTR (the exception) | +18.68% | Amsive |
A few nuances are worth flagging. Pew Research found that only about 1% of users click a link inside the AI Overview itself, and roughly 26% of sessions are abandoned after an AIO appears. Around 58% of US Google searches now end without any click at all (SparkToro / Superprompt). The pain is heaviest on desktop (~50% CTR reduction) and lighter on mobile (~30%) according to Harton Works and Conductor. The lone bright spot: branded and cited results can actually gain CTR, which is why earning citations is now a core SEO objective.
Which Sources Get Cited in AI Overviews?
If clicks are scarcer, being cited inside the AI Overview is the new prize. Citation data reveals a strong preference for community platforms, encyclopaedic references, and high-authority domains. Here are the most-cited domains in Google AI Overviews:
| Domain | Citation share | Source |
|---|---|---|
| reddit.com | ~9.7% (most cited) | Semrush |
| linkedin.com | ~8.9% | Semrush |
| wikipedia.org | ~7.6% | Semrush |
| medium.com | ~4.8% | Semrush |
| youtube.com | ~4.2% | Semrush |
| nih.gov | ~4.1% | Semrush |
| forbes.com | ~3.4% | Semrush |
| Quora | ~4% | Ahrefs |
Other citation patterns to plan around:
- The top 50 domains capture about 30% of all AIO mentions (Ahrefs) — citation is concentrated among authoritative sites.
- About 40% of AIO sources rank positions 11-20 (page 2), not the top 10 — you do not need to rank #1 to be cited (Ahrefs).
- Short AIOs (under 600 characters) cite around 5 sources; long ones (6,600+ characters) cite up to ~28 sources (Search Engine Journal). The median AIO cites 13.34 links, with one outlier citing 95 (SE Ranking).
- 88% of AI summaries cite three or more sources; only ~1% rely on a single source (Pew Research Center).
- Brands cited in an AI Overview earn roughly 120% more organic clicks per impression than uncited brands on the same query (Surfer SEO analysis).
User Trust & Behaviour Statistics
How people read, trust, and act on AI overviews shapes everything downstream. The behavioural data shows users are reading less, trusting cautiously, and clicking rarely:
- 7 in 10 users read only the first lines of an AI Overview before moving on (Growth Memo).
- 88% of users click "show more" to expand a truncated AIO (Search Engine Land).
- 70% of consumers say they at least somewhat trust generative AI results, while 75% are concerned about AI-driven misinformation (Search Engine Land; Forbes).
- 79% of consumers expect to use AI-enhanced search within a year (Search Engine Land).
- AI Overviews occupy about 42% of the desktop screen and 48% on mobile, with ~81% of AIO queries done on mobile (Botify; Ahrefs).
- Accuracy remains a concern: 75% of users reported spotting at least one major error in AIO answers, with 51% citing inaccurate answers (NP Digital, 1,000 respondents).
AI Overviews vs Traditional SERP Features
AI Overviews do not just sit above the SERP — they reshape the rest of it. When an AIO appears, other rich results get squeezed out, and zero-click behaviour rises. The contrast between an AIO SERP and a traditional one is stark:
| SERP element / behaviour | With AI Overview | Source |
|---|---|---|
| Sitelinks | 96% fewer | Ahrefs |
| Video results | 84% fewer | Ahrefs |
| Knowledge Panels | 75% fewer | Ahrefs |
| People Also Ask co-occurrence | ~81-98% of AIO pages | Ahrefs; SE Ranking |
| Zero-click searches | ~58% of US searches | SparkToro |
| Median AI summary length | 67 words | Pew Research Center |
For comparison, ChatGPT and Perplexity behave differently from Google's AIO. They respond to nearly 100% of prompts, while Google shows an AIO on roughly 58% of the same set (SE Ranking). ChatGPT leans heavily on Wikipedia (~48% top citation), Perplexity leans on Reddit (~47%), while Google AI Overviews use the widest range of sources of any major AI search engine (Search Engine Land; SE Ranking).
What These AI Overviews Statistics Mean for SEO
The data points to a clear strategic shift: visibility is no longer just about ranking #1 — it is about being the source the AI summarises and cites. With clicks compressing, the goal becomes earning the citation, capturing branded demand, and defending the query types that still send traffic. The implications for any SEO strategy are significant:
- Informational content is most exposed. With ~99% of AIO triggers being informational, top-of-funnel guides will lose direct clicks — but they remain the best route to being cited.
- Transactional and local queries are safer. Ecommerce (2-4%) and local services (~2-7%) rarely trigger AIOs, so commercial pages retain more of their traditional CTR.
- Citation beats ranking. Because ~40% of cited sources sit on page two, strong on-page SEO and clear, quotable answers can win citations even without a #1 ranking.
- Brand demand becomes a moat. Branded queries are the one segment where CTR actually rises with AIOs, making brand-building a measurable SEO defence.
How to Optimise for AI Overviews (Actionable Steps)
You cannot opt out of AI Overviews, but you can position your content to be the source Google cites. Based on what the citation data rewards, here is a practical, ordered playbook:
- Answer the question in the first 1-2 sentences. Lead each section with a concise, 40-55 word direct answer that an AIO can lift verbatim.
- Structure for extraction. Use clear H2/H3 headings, bulleted lists, and comparison tables — the formats AIOs pull from most. Strong technical SEO ensures Google can crawl and parse them.
- Target question and long-tail queries. Since 7+ word and question queries trigger AIOs most, build content around them. Solid keyword research surfaces the conversational phrases worth owning.
- Demonstrate E-E-A-T. Cite primary data, name authors, and show expertise — Google favours trustworthy sources for AIO citations. Earn mentions on Reddit, LinkedIn, and other high-citation platforms.
- Win featured snippets first. Snippet-optimised pages are prime AIO citation candidates; our guide to mastering featured snippets covers the formatting that works.
- Track AIO visibility, not just rank. Monitor citation share and zero-click impressions, and use analytics to separate AIO-exposed queries from protected ones.
- Double down on commercial and branded pages. Reinforce the transactional and branded queries that AIOs rarely touch with strong content marketing to capture the clicks that still convert.
For deeper context on how AI is reshaping organic search, see our breakdown of AI and SEO.
Frequently Asked Questions
How often do Google AI Overviews appear?
Estimates range from roughly 15-25% of US searches (Semrush, Ahrefs) up to ~50-55% by broader trackers, depending on methodology. Frequency is highest for health (60.7%) and informational, question-style, and long-tail queries, and lowest for ecommerce (2-4%) and local services (NP Digital; Ahrefs).
Do AI Overviews reduce website traffic?
Yes. Ahrefs found organic clicks drop ~34.5% on a keyword when an AI Overview appears, and Pew Research recorded clicks on traditional links falling from 15% to 8% of visits when a summary is present. Impact is heaviest on desktop and informational queries.
Which sources do AI Overviews cite most?
Reddit is the single most-cited domain (~9.7%), followed by LinkedIn (~8.9%), Wikipedia (~7.6%), Medium, YouTube, and NIH.gov (Semrush). About 40% of cited sources rank on page two, so a #1 ranking is not required to be cited.
How many users actually trust AI Overviews?
About 70% of consumers say they at least somewhat trust generative AI search results, but 75% are concerned about misinformation, and 75% reported spotting at least one major error in AIO answers (Search Engine Land; Forbes; NP Digital).
How do I get my content cited in AI Overviews?
Lead with a concise direct answer, structure content with headings, lists, and tables, target question and long-tail queries, demonstrate E-E-A-T with original data, and win featured snippets — the formats and signals AIOs pull from most.
Turn AI Overview Data Into a Search Strategy
AI Overviews are not the end of SEO — they are a reshaping of it, and the brands that adapt to being cited will win the visibility that clicks alone used to provide. At D'Marketing Agency, we help businesses audit their AIO exposure, protect their highest-converting queries, and build the citation-worthy content that earns a place inside Google's answers. Use the quote form on this page to request a free consultation and a tailored AI-search readiness plan.
