---
title: "AI Overview citations: how often is a Google top-10 site also cited? 30.7% in the US"
description: "AI Overview citations vs Google rankings: in the US, 30.7% of top-10 domains are also cited (69 searches). By position, query type and market, with data."
url: "https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/"
updated: "2026-10-02"
---

# AI Overview citations: how often is a Google top-10 site also cited? 30.7% in the US

> TIN read the Google results and AI Overview for 438 buyer searches in seven markets. In the US, 30.7% of top-10 organic domains are also cited in the AI Overview (152 of 495, across 69 searches), positions 1 and 2 most often, and 63.0% of cited domains are not in the top 10. The six European markets are shown one by one with their n.

- Study type: Recon measurement by The Internet Ninja. TIN is the instrument; this is a measurement, not an adjudication of any claim.
- Edition: 2026-10-02
- n = 1760
- Population: Every organic row and every cited domain on the 151 usable Google results (of 438 frozen buyer phrases in 7 markets) read through Ahrefs SERP overview on 2026-10-01 and 2026-10-02; results more than 30 days old and phrases with no stored result are counted and excluded. (from 438 serp-overview calls (recon/google-top10-vs-ai-overviews-2026/serp/), 151 usable, 113 with an AI Overview; rows built by analysis/build_data.py with the rules of analysis/analyze.py.)
- Pre-registered: method file `recon/google-top10-vs-ai-overviews-2026/method.md` committed at [`ac309177d4c1`](https://github.com/ManeeshBehera/plan-2k27-1M/commit/ac309177d4c1f44b621a651d92f12afe83371b37) before collection
- Dataset: [data.csv](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv), 1760 rows, one per entity (CC BY 4.0)
- Canonical URL: https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/

## Key findings

| # | Finding | Value | n |
| --- | --- | --- | --- |
| 1 | In the US, 30.7% of top-10 organic domains are also cited in the AI Overview (152 of 495, 69 searches) | 30.7% | 69 |
| 2 | US organic positions 1 and 2 are also cited in 44.9% and 46.4% of searches; position 7 in 21.1% | 44.9% and 46.4% | 69 |
| 3 | 63.0% of the domains US AI Overviews cite are not among that search's top-10 organic domains (259 of 411) | 63.0% | 69 |
| 4 | Matched by exact URL, at least 25.7% of US top-10 pages are the page the AI Overview cites (118 of 460) | at least 25.7% | 69 |
| 5 | The US overlap is between 28.8% and 31.5% for each of the four query types | 28.8% to 31.5% | 69 |
| 6 | Six European markets, each shown alone with its n: 4 to 10 searches with an AI Overview per market | 4 to 10 | 44 |

When a page ranks in Google's top 10, how often is the same site also cited in that search's AI Overview? We wrote 240 buyer intents across 12 industries and four kinds of query (choosing, comparing, pricing, doing), found the phrase people actually search for each one in seven markets, and read the Google results and AI Overview for every phrase on 2026-10-01 and 2026-10-02 [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). The method was committed before any result was read.

The US is the headline: 69 searches carried an AI Overview, with 495 top-10 organic domains between them [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). The six European markets (France, Spain, Italy, Germany, Portugal, the Netherlands) had 4 to 10 such searches each, too few to support a headline, so each one is shown on its own with its n, and none is pooled with another or compared [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv).

## 1. In the US, 30.7% of top-10 organic domains are also cited in the AI Overview

Across the 69 US searches with an AI Overview, 152 of the 495 top-10 organic domains (30.7%) are also cited in that search's AI Overview [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). It varies a lot from search to search: the median search shares 29%, the range runs from 0 to 86%, and 12 searches share no domain at all [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv).

| Measure | US, 69 searches with an AI Overview |
|---|---|
| Top-10 domains also cited | 30.7% |
| Median search | 29% |
| Searches sharing no domain | 12 |

## 2. The first two organic results are the ones most often cited (US)

Of the 69 US searches, the domain in position 1 is also cited in the AI Overview in 44.9% and position 2 in 46.4% [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). From position 3 the share falls, to 21.1% at position 7 (57 searches reach a 7th result) [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv).

The chart stops at position 7, the last with at least 30 searches behind it; position 8 (26 searches, 19.2%) is left out of the chart [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). No US search returned a 9th or 10th organic result: SERP features take those slots, so "top 10" here means the organic results the page actually showed, up to 8 [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv).

## 3. Most AI Overview citations come from outside the top 10 (US)

Turned the other way: of the 411 distinct domains the US AI Overviews cite, 63.0% are not among that search's top-10 organic domains [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). That includes 35 opaque Google redirect links, which cannot be matched to any page and so count as outside. Ranking in the top 10 and being cited overlap, but most of what an AI Overview cites sits outside the first page of results.

## 4. Matching exact pages, not domains: at least 25.7% (US)

The headline matches by domain. Matching by exact URL instead, at least 25.7% of US top-10 organic pages (118 of 460) are the very page the AI Overview cites [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). It is a lower bound: by a fixed rule, 35 of the 495 organic URLs are flagged as cut short to a section or a home page and left out of this share, and the rule cannot catch every cut URL [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv).

## 5. The overlap is about the same for every query type (US)

The overlap barely moves with the kind of query, from 28.8% for doing queries (9 searches) to 31.5% for pricing (13) [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). Business queries share 30.2% (62 searches); consumer queries 35.4%, on only 7 [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). Each industry has 4 to 8 searches, too few for a finding, so industries are listed for reference only.

| Query type | Searches | Also cited |
|---|---|---|
| Choosing | 33 | 31.0% |
| Comparing | 14 | 30.6% |
| Pricing | 13 | 31.5% |
| Doing | 9 | 28.8% |
| Business (B2B) | 62 | 30.2% |
| Consumer | 7 | 35.4% |

| Industry | Searches | Also cited |
|---|---|---|
| Banking | 6 | 37.8% |
| Education | 8 | 29.3% |
| Energy utilities | 4 | 26.7% |
| Healthcare | 7 | 30.8% |
| Hospitality travel | 4 | 50.0% |
| Insurance | 6 | 25.6% |
| Logistics | 6 | 30.8% |
| Manufacturing | 7 | 33.3% |
| Professional services | 6 | 26.2% |
| Real estate | 4 | 38.7% |
| Retail ecommerce | 6 | 17.9% |
| SaaS | 5 | 25.7% |

## 6. The six European markets, each on its own

Each market below stands alone, with its own n. The markets are not pooled into an "EU" figure, and none is compared with the US or with another market: with 4 to 10 searches each, a gap between two markets would say more about the few searches drawn than about the market [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). Every per-market measure, including position, is in the downloadable data.

| Market | Searches with an AI Overview | Top-10 domains also cited |
|---|---|---|
| France | 10 | 37.2% |
| Spain | 9 | 44.3% |
| Italy | 8 | 33.3% |
| Germany | 9 | 43.5% |
| Portugal | 4 | 10.0% |
| Netherlands | 4 | 40.6% |

## 7. How often the crawler recorded an AI Overview

The share of each market's usable searches where the crawler recorded an AI Overview [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). It is not a ranking of markets. Some European rollouts first showed AI Overviews only to signed-in adults, and the crawler sees a logged-out page, so a lower figure means "recorded by the crawler", not "Google does not show AI Overviews here".

| Market | Usable searches | AI Overview recorded |
|---|---|---|
| United States | 91 | 75.8% |
| France | 13 | 76.9% |
| Spain | 11 | 81.8% |
| Italy | 8 | 100.0% |
| Germany | 13 | 69.2% |
| Portugal | 4 | 100.0% |
| Netherlands | 11 | 36.4% |

## What this suggests if you want to be cited in AI Overviews

These are associations in one snapshot, not causes.

- A first-page ranking and an AI Overview citation overlap, but only partly: 30.7% of US top-10 domains are also cited (152 of 495), and 63.0% of cited domains are not on the first page (259 of 411) [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). Track both, because one is not a stand-in for the other.
- The top two organic positions are the ones most often cited in the US: 44.9% and 46.4% of searches (31 and 32 of 69) [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv).
- The pattern holds across choosing, comparing, pricing and doing queries, so it is not specific to one stage of a buyer's search.

Related on TIN: [which sites AI engines cite, by engine and country](https://theinternetninja.com/recon/ai-citation-sources-2026/) and [what the retrieval record shows about AI search optimization](https://theinternetninja.com/blog/ai-search-optimization-what-the-retrieval-record-shows-2026/).

## How this was measured

1. **Intents.** 240 buyer intents: 12 industries, 4 query types (choosing, comparing, pricing, doing), 5 each. Frozen before any search phrase was chosen.
2. **Phrases.** For each intent and market, the highest-volume real search query that names the intent's topic and carries its query-type word, from Ahrefs Keywords Explorer, picked by two blind passes that never saw a search result. Kept: US 174 of 240, France 57, Spain 33, Italy 47, Germany 68, Portugal 18, the Netherlands 41 (438 in all). The first rule, one written phrase per intent kept only if Ahrefs held its volume, had kept US 61, France 10, Spain 7, Italy 3, Germany 11, Portugal 2 and the Netherlands 6.
3. **Freeze.** The phrases, rules and data source were hashed in FREEZE.md and committed before any search result was read. The European markets fell under the pre-set floor of 120 phrases, so FREEZE.md recorded before collection that none would be pooled.
4. **Collection.** One Ahrefs SERP overview call per phrase (438), in one window on 2026-10-01 and 2026-10-02: the top organic results, the AI Overview, and the sites it cites. 227 phrases returned no stored search result and 60 results were more than 30 days old; both are counted and left out, never replaced [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv). 151 usable searches remain, 113 of them with an AI Overview [source](https://theinternetninja.com/recon/google-top10-vs-ai-overviews-2026/data.csv).
5. **Matching.** A top-10 result counts as cited when its registrable domain matches a cited domain on the same search. On YouTube, Reddit, Wikipedia and other platforms where one domain hosts unrelated authors, only the exact URL counts. Position is the organic order on the page. Duplicate organic rows (18) are removed.
6. **Data notes.** Two US searches (S4-169, S4-233) disagree between Ahrefs' AI Overview marker and its citation rows; the citation rows decide, as the pilot set. AI Overview presence uses the citation rows. The full SERP-by-row data is in the download.

## Method record

Amendments, each landed before any outcome data:

- `recon/google-top10-vs-ai-overviews-2026/amendment-1.md` at [`d23d033683e1`](https://github.com/ManeeshBehera/plan-2k27-1M/commit/d23d033683e10fdf74a3ca92cb85dcf231f097df)
- `recon/google-top10-vs-ai-overviews-2026/amendment-2.md` at [`d23d033683e1`](https://github.com/ManeeshBehera/plan-2k27-1M/commit/d23d033683e10fdf74a3ca92cb85dcf231f097df)
- `recon/google-top10-vs-ai-overviews-2026/amendment-3.md` at [`1409e30935f1`](https://github.com/ManeeshBehera/plan-2k27-1M/commit/1409e30935f1c476b0ae5d8137fdc475ee889b42)
- `recon/google-top10-vs-ai-overviews-2026/FREEZE.md` at [`3c6108e7b7f4`](https://github.com/ManeeshBehera/plan-2k27-1M/commit/3c6108e7b7f42269229d1345ecf70eb9e60e1667)
- `recon/google-top10-vs-ai-overviews-2026/amendment-4.md` at [`f181f214dd9e`](https://github.com/ManeeshBehera/plan-2k27-1M/commit/f181f214dd9e7846cf4a7477ce510fa86e1c94ea)
- `recon/google-top10-vs-ai-overviews-2026/amendment-5.md` at [`789661960cec`](https://github.com/ManeeshBehera/plan-2k27-1M/commit/789661960ceccac4caa8c80af56247e96f345919)

| Metric | Unit | Tool | Computation |
| --- | --- | --- | --- |
| Top-10 domains also cited | % | recon/google-top10-vs-ai-overviews-2026/analysis/analyze.py | Among searches with an AI Overview, organic rows whose registrable domain (Public Suffix List; exact URL on 12 platforms) matches a domain the AI Overview cites, divided by the organic rows returned (amendment-2 B3, B5). |
| Share cited at an organic position | % | analysis/analyze.py | The same match, per organic ordinal (the first organic row is position 1). Positions with fewer than 30 searches are printed, not charted. |
| Cited domains outside the top 10 | % | analysis/analyze.py | Distinct cited domains not among the search's top-10 organic domains, divided by all distinct cited domains; opaque Google redirects count as outside (B5). |
| Exact-URL match (lower bound) | % | analysis/analyze.py | Organic URLs equal to a cited URL after normalisation, over unflagged rows only; URLs cut to a section or home page are flagged and left out (B4). |
| AI Overview recorded | % | analysis/analyze.py | Usable searches with at least one AI Overview citation row, per market; never ranked across markets (amendment-1 A4). |

Design: correlational. The findings report what the measured entities have in common, never that a trait produced an outcome.

## What this study cannot say

- One crawl window (2026-10-01 and 2026-10-02). AI Overview answers vary between runs; this is a snapshot.
- The intents are TIN's buyer queries in its four channels. They are not a random sample of all searches.
- The source is a crawler's view of the results page, not a real user's personalised result.
- Ahrefs does not publicly document what its AI Overview fields mean, nor the device (desktop or mobile) its crawler uses. The field meanings here come from TIN's own pilot on off-study keywords.
- Picking the highest-volume matching query tilts the sample toward the most-searched phrasing of each intent. The study describes what is searched most for each intent, not the long tail. Phrases were re-sourced after the first rule left the European markets too small; the first rule's counts are printed in the method.
- The US headline rests on 69 searches with an AI Overview. The European markets have 4 to 10 each, so their figures are shown alone, with n, and are not compared.
- In the US, no search returned more than 8 organic results; "top 10" means the organic results the page showed.
- The 101 searches collected first (S4-001 to S4-062) were read with every field Ahrefs offers and the rest with the four fields the analysis uses (amendment 4). No measure is computed differently.
- A ZAK study published on 14 July 2026 looked at a related question. It is mentioned for context only; none of its figures are compared with these.

## Questions this study answers

**How often are Google top-10 sites in AI Overview citations?**

In TIN's study, 30.7% of the time in the US: 152 of 495 top-10 organic domains, across 69 searches with an AI Overview, were also cited in that AI Overview. The share per search ranged from 0 to 86%, with a median of 29%.

**Does ranking first on Google mean the AI Overview cites you?**

Not usually. In the US, the domain at organic position 1 was also cited in 44.9% of searches with an AI Overview (31 of 69), and position 2 in 46.4% (32 of 69). This is an association in one snapshot, not a cause.

**Where do the other AI Overview citations come from?**

Mostly from outside the top 10. Of 411 distinct domains cited in US AI Overviews, 259 (63.0%) were not among that search's top-10 organic domains.

**Does the type of query matter?**

Very little in this study. The US overlap was 31.0% for choosing queries, 30.6% for comparing, 31.5% for pricing and 28.8% for doing, on 33, 14, 13 and 9 searches.

**What about Europe?**

France, Spain, Italy, Germany, Portugal and the Netherlands each had 4 to 10 searches with an AI Overview, too few for a headline. Each is shown on its own with its n, and none is pooled or compared.

**How was this measured?**

TIN froze 240 buyer intents, picked the most-searched real phrase for each in seven markets with two blind passes, and read each phrase's Google result and AI Overview through Ahrefs on 2026-10-01 and 2026-10-02. The method was committed before any result was read.

## Related

- [ChatGPT ranks passages, not pages: what its retrieval stream shows, and the numbers nobody can source](https://theinternetninja.com/blog/ai-search-optimization-what-the-retrieval-record-shows-2026/)

## Editions and errata

This edition: 2026-10-02. A refresh republishes on the same URL, and every prior edition's dataset stays downloadable.

No erratum.