The single most important trend: the share of AI Overview citations coming from Google's top-10 ranked pages fell from roughly 76% to 38% in about a year, per Ahrefs' own re-measurement. Ranking still helps, but it is no longer close to a prerequisite for citation.
Headline numbers
of AI Overview citations came from pages ranking in Google's top 10 — down from 76% roughly a year earlier.
the actual year-over-year decline, independently reported by the same source that produced the original figure.
What an AI Overview actually is, precisely
Worth defining precisely before the statistics, since imprecise usage is common. An AI Overview is a generated summary that appears above classic organic results for a qualifying query. It synthesizes an answer from multiple retrieved sources, and attaches citations to some or all of the claims within it.
It's a distinct product from Google's AI Mode, a separate, dedicated conversational search experience. See the dedicated AI Mode statistics page. It's also distinct from a classic featured snippet, which extracts one passage verbatim rather than generating new text. Treating these as interchangeable, as some coverage does, produces exactly the kind of conflated statistic this site's provenance audit warns against.
Citation-to-ranking overlap over time
This is the single most consequential trend statistic in AI-search SEO. A year ago, being cited in an AI Overview was closely tied to already ranking well on Google. That link has weakened substantially. It hasn't disappeared, but it's weakened enough that a page outside the top 10 now has a real, measurable path to citation it didn't have as clearly before.
AI Overview citations coming from Google's top-10 pages fell from ~76% to ~38% in about a year. Ranking well still helps — it's just no longer close to a prerequisite.
Share on XWhy the overlap likely fell, and why we say "likely"
Two plausible mechanisms have been proposed in industry discussion. Worth distinguishing them from confirmed causes, since neither has been directly tested. The first: Google may have deliberately broadened the source pool AI Overviews draw from, independent of the ranking algorithm behind the ten blue links, as the product matured past its early reliance on the same signals as classic search.
The second: AI Overviews may increasingly draw on a retrieval layer, potentially incorporating the same fan-out mechanism documented for AI Mode, that surfaces sources a purely ranking-based system wouldn't have selected. Both are Hypothesis on this site's evidence scale. No published study directly isolates the mechanism behind the observed decline.
What AI Overviews cite
AI Overviews draw on a broader source mix than classic organic results, with reported skew toward multimodal and video content. That pattern is consistent with Google's access to YouTube and its Knowledge Graph as retrieval inputs unavailable to other engines. The specific percentage breakdown by source type circulating in vendor coverage relies on a non-public corpus, so it's graded Partial here, rather than restated as a precise figure.
Click-through impact
Multiple reports describe reduced click-through to organic results when an AI Overview is present for a query, with figures in circulation ranging widely by study and definition — see the provenance audit's "58% problem" section for why we're not restating a single number here without the population it applies to attached. The dedicated, disambiguated treatment of click-through claims specifically is at AI search CTR statistics.
Trigger rate by query type
AI Overviews don't trigger uniformly across query types. Informational and "how-to" queries trigger them far more often than transactional or highly localized queries, per widely consistent, though largely vendor-sourced, Partial reporting. A rigorously sampled, disclosed-methodology breakdown by query category is one of the statistics pages planned once the Citation Index's query set is running.
Full statistics table
| Statistic | Value | Source | Grade |
|---|---|---|---|
| Citation-to-top-10 overlap, 2025 | ~76% | Ahrefs, Jul 2025 | Traceable |
| Citation-to-top-10 overlap, 2026 | ~38% | Ahrefs, Mar 2026 | Traceable |
| Multimodal/video source skew | Directionally reported, not precisely quantified here | Vendor-reported, non-public corpus | Partial |
| Query-type trigger-rate breakdown | Not independently verified at time of writing | Various, largely vendor-sourced | Partial |
How to monitor your own AI Overview exposure
Rather than relying solely on industry-wide figures, a site owner can build a rough, first-party picture. Use free tools already published on this site. The AI Overview Exposure Checker flags which of your own target keywords currently trigger an AI Overview. The AI Traffic-Loss Calculator translates that exposure into an estimated traffic-at-risk figure specific to your own query mix. That's a more actionable starting point than any aggregate percentage on this page.
A brief history of AI Overviews
AI Overviews emerged from Google's earlier "Search Generative Experience," or SGE, experiment. It was tested with a subset of users before rolling out broadly under the AI Overviews name. That lineage matters for reading older statistics.
Figures reported under the SGE label describe an earlier, more limited version of the product. They shouldn't be treated as directly comparable to current AI Overview measurements, without checking which version of the product was actually being tested at the time.
A worked example: reading one query correctly
Say you check a target keyword and find it triggers an AI Overview, but your page isn't cited. Under the old rule of thumb, ranking well was nearly a prerequisite, so a missing citation mostly meant "you're not ranking." Under the current 38% overlap, that's no longer the safe assumption.
Check your actual Google ranking for that query first. If you already rank in the top 10 and still aren't cited, ranking clearly isn't the blocker; something else about the page's structure or content likely is. If you don't rank in the top 10 either, both problems may be worth fixing, but they're not the same problem anymore.
Verification status
The overlap trend, 76% to 38%, is the strongest-sourced claim on this page: same organization, comparable method, disclosed sample sizes both times. Everything else here is presented with the grade it earned, not the grade a cleaner-looking bullet point would suggest.
Namdev, R. (2026). Google AI Overview statistics (v2). Retrieved from https://ritiknamdev.com/blog/google-ai-overview-statistics Published under CC BY 4.0 — reuse freely with attribution.
For the mechanism behind AI Mode specifically (a separate Google surface), see Google AI Mode statistics. For the provenance method behind these grades, see the statistics audit.