How Much of Google Search Now Shows AI Overviews, and How Much Organic CTR Do They Take in 2026?

Google AI Overviews appear on 15.69% to 48% of tracked queries in 2025–2026 studies, depending on the keyword panel. Semrush’s December 2025 study of more than 10 million keywords settled at 15.69% in November 2025 after a July peak of 24.61%. BrightEdge’s February 2026 Generative Parser report measured 48% of its commercial tracker, up from about 31% in February 2025. Conductor’s 2026 AEO / GEO Benchmarks Report counted 5.5 million AI Overviews in 21.9 million U.S. searches between 15 September and 12 October 2025 — 25.11% of that mixed-intent panel.

Pew Research Center’s July 2025 analysis of 68,879 Google searches by 900 U.S. adults in March 2025 found an AI summary on 18% of those searches. Users clicked a traditional result on 8% of visits with a summary versus 15% without one, and they clicked a source inside the summary on 1% of those visits. Ahrefs’ February 2026 update of 300,000 keywords reported a 58% lower position-1 click-through rate in December 2025 versus the forecast without Overviews. That is the click cost on queries the Overview already occupies.

Want the full workflow on one page? Get the GEO Cold Start: The AI Citation Playbook — English + 中文, free download (create a free Sent2X account).

How often do Google AI Overviews appear, and why do the percentages disagree?

The coverage number moves with the sample. A long-tail mixed panel includes transactional and navigational queries that still skip the Overview. A commercial or question-heavy panel includes the queries Google summarizes first. Both figures can be correct for the set they measured.

Study	Window	Sample	Overview share
Semrush	January–November 2025	10 million+ keywords	6.49% (Jan) → 24.61% (Jul) → 15.69% (Nov)
Pew Research Center	March 2025 searches	68,879 Google searches, 900 U.S. adults	18% produced a summary
Conductor	15 Sep–12 Oct 2025	21.9 million U.S. searches	25.11% (Health Care 48.75%; Real Estate 4.48%)
BrightEdge	February 2025–February 2026	Commercial keyword tracker	~31% → 48%

Semrush also recorded a shift in intent. In January 2025, 91.3% of Overview-triggering queries were informational. By October that share was 57.1%, while commercial queries rose to 18.57%, transactional to 13.94%, and navigational to 10.33%. Coverage is no longer limited to definitional long-tail terms.

BrightEdge’s 12-month series crossed 40% in mid-2025 and reached 48% in February 2026, a 58% year-over-year rise on that tracker. The same report notes that about 52% of its tracked queries still show no Overview. Use your own query mix as the denominator. A recipe site, a B2B comparison library, and a local service page will not share one rate.

How much does an AI Overview reduce clicks for a page that already ranks number one?

Ahrefs compared December 2023 and December 2025 desktop Search Console CTR across 150,000 Overview keywords and 150,000 informational keywords without an Overview. Position-1 CTR on Overview keywords fell from 0.073 to 0.016. After applying the decline seen on non-Overview informational terms, the residual gap was 58%. Ahrefs framed that as 58 of every 100 historical position-1 clicks now staying on Google.

The same update published a position ladder. The Overview penalty shrinks down the page, but it does not disappear:

Organic position	CTR impact with an Overview (Ahrefs, Dec 2025)
1	−58.0%
2	−50.8%
3	−46.4%
5	−32.6%
10	−19.4%

Pew’s observed-behavior panel points the same direction with a different method. On visits with an AI summary, users clicked a traditional result 8% of the time and a summary citation 1% of the time. On visits without a summary, they clicked a result 15% of the time. They also ended the browsing session after 26% of summary pages versus 16% of traditional-result pages. Ranking first still puts the URL on the page. It does not put the URL above the answer block.

Why did Search Console impressions rise while clicks fell on the same ranking URLs?

Search Console does not split Overview impressions from blue-link impressions. When Google expands the set of queries that show a summary, a ranking page can collect more impressions while collecting fewer clicks. That is the “impressions up, clicks down” pattern operators report after an Overview lands on a cluster.

Seer Interactive’s April 2026 CTR study tracked 53 brands, 5.47 million queries, and 2.43 billion organic impressions from January 2025 through February 2026. Brand-cited Overview CTR fell 61% from Q3 to Q4 2025: 2.52% in September to 0.76% in November. Clicks on those pages did not collapse in lockstep. September recorded 398,798 clicks on 15.8 million impressions; October recorded 400,271 clicks on 33.1 million impressions. The denominator grew faster than the numerator.

Screen geometry adds a second cause. BrightEdge measured average Overview height at about 1,200 pixels in February 2026, up from about 1,050 pixels a year earlier, against a standard desktop viewport near 900 pixels. The first organic result sits below the fold when the block is present. Pew also found that Wikipedia, YouTube, and Reddit together accounted for 15% of sources listed in the summaries it examined — the same three domains that dominate standard results. A small site can lose the click even when its rank did not move.

Does getting cited inside an AI Overview restore the clicks that ranking used to produce?

Citation beats non-citation on the same Overview SERP. It does not restore the pre-Overview baseline. Seer’s 2026 update states that being cited delivered 120% more organic clicks per impression than not being cited on the same Overview-present SERP, and that cited pages still lagged no-Overview pages by 38%. In February 2026, organic CTR was 2.36% when an Overview was shown and 3.82% when it was not.

Ranking and citation are not the same contest. BrightEdge found that only about 17% of Overview-cited sources also ranked in the organic top 10, and that share stayed flat from February through July 2025. Healthcare overlap sat near 24%; finance sat near 11%. Five of six citations came from outside page one. That answers the rank-versus-citation question: position 1 is not a ticket into the block, and a page outside the top 10 is not automatically excluded.

Pew’s 1% click rate on summary links is the other bound. Most readers who see the block do not open a cited URL. Treat citation as a visibility and brand-search asset first, and as a residual click channel second. For ChatGPT, Perplexity, Gemini, and Claude referral share — a separate traffic pool from Overview CTR — use the existing AI referral traffic comparison.

Which query types should a small website treat as Overview-exposed first?

Question syntax and length are the cleanest free filters. Pew recorded an AI summary on 60% of searches that began with who, what, when, or why, on 53% of searches with 10 or more words, and on 8% of one- or two-word searches. Seer, across 30,842 informational queries, recorded an Overview on 95.4% of comparison queries (267 of 280) and 85.9% of question-format queries (1,214 of 1,413). Informational queries showed an Overview 36% of the time in that set, versus 8% for commercial and 5% for transactional.

Map those rates onto your library before rewriting every URL. Comparison and question pages are Overview-exposed by default. Transactional and local pages still sit closer to classic SERPs. Conductor’s industry split matches that pattern: Health Care 48.75%, Financials 25.79%, Utilities 25.4%, Consumer Staples 6.82%, Real Estate 4.48%. If a page answers a how/what/which question, write the first 80 words as a quotable answer with a named number and a source. If the page is a checkout or booking URL, keep ranking work on the blue link and track branded search as the Overview proxy.

If a page ranks and still never appears in the block, the failure is often extractability rather than rank. The diagnostic list in why content is ignored by ChatGPT, Perplexity, and AI Overviews covers robots.txt, passage structure, and source quality. For the weekly scorecard when Google will not split Overview clicks, use the GEO measurement workflow without click data.

Frequently Asked Questions

How often should you re-check which keywords trigger an AI Overview?

Re-run a fixed 20-query set every 14 days. Semrush’s 2025 series moved from 6.49% in January to 24.61% in July and 15.69% in November, so a quarterly snapshot misses the swing. Keep the same wording, country, and device. Log Overview present/absent, citation URL, and Search Console clicks for each query.

When is it worth rewriting a page for citation instead of chasing a higher rank?

Rewrite when the query already shows an Overview, your URL ranks in the top 10, and another domain occupies the citation. BrightEdge’s 17% top-10 overlap means rank work alone often fails. Add a 40–80 word answer, one sourced number, and a limitation line, then re-check the same query after 14 days.

Which free reports show Overview impact, and which paid tools add coverage data?

Google Search Console shows impressions, clicks, and CTR but does not isolate Overviews. A free check is a before/after CTR split on a saved query set. Ahrefs, Semrush, BrightEdge, and Seer-style panels add coverage and citation logs. Start with Search Console plus a 20-query manual SERP sheet before paying for prompt tracking.

How do you measure whether Overview citation work is paying back?

Track four numbers on a 14-day window: Overview presence rate, citation rate, Search Console clicks on the same URLs, and branded search or assisted conversions. Seer found cited pages earned 120% more clicks per impression than uncited peers on Overview SERPs, still 38% below no-Overview CTR. Keep the window fixed and disclose it.