AI Overview Click-Through Rate Statistics: What the Independent Data Actually Says

Most stats roundups on AI Overviews quietly mix two very different things: independent research from organizations with no product to sell, and vendor studies from SEO tools that benefit from the panic. They cite both with the same weight. You shouldn't.

So here is the short answer to the ai overview click-through rate statistics question, graded by who measured it. Pew Research Center found that when an AI Overview appears, users click a traditional result in 8% of visits versus 15% without one, and click a cited source just 1% of the time (Pew, July 2025). Ahrefs, working from real Google Search Console data, measured a 34.5% lower CTR for the top-ranking page when an AI Overview was present, then updated that to a 58% decline at position one in December 2025 (Ahrefs original study, Ahrefs December 2025 update). Those four numbers are the load-bearing ones. Everything else in this piece is context for how to read them.

TL;DR

• Pew (independent, July 2025): clicks fall from 15% to 8% when an AI Overview appears; 1% click a cited source.

• Ahrefs (GSC data, original study): top-ranking page sees a 34.5% lower CTR with an AI Overview present.

• Ahrefs (GSC data, December 2025 update): position one −58%, position two −50.8%, position three −46.4% CTR.

• Vendor data (Semrush, flagged): AI-search visitors valued at 4.4x an organic visitor; AI search projected to overtake traditional search by early 2028.

• Operator read: impressions rising, clicks falling, intent quality rising on the clicks that survive.

1. The headline numbers, graded by who measured them

Before we go through the studies, a short note on grading. Independence here means three things: the researcher has no commercial product riding on the answer, the methodology is published, and the sample is either nationally representative or drawn from production search data the researcher doesn't control.

Notice what's missing from most roundups you'll read this quarter: that fourth column. The grade. Treat it as the first filter when you decide what to plan against.

Independent (Pew) and GSC-based (Ahrefs) CTR declines, with source grade. Higher bar = larger CTR loss.

2. Pew Research: the 15% → 8% drop, and the 1% citation click

The Pew Research Center finding is the most-cited zero-click search statistic of 2026 because it's the only one drawn from a nationally representative US sample.

Pew tracked 900 US adults across 68,879 Google visits in March 2025 (full methodology here). When an AI Overview appeared, users clicked a traditional organic or sponsored result in 8% of visits. When no AI Overview appeared, they clicked one in 15%. That's not a relative drop measured against rankings — it's a behavioral finding measured against real human visits.

The quieter number is the one operators should fixate on: of the sources cited inside the AI Overview itself, users clicked through to them 1% of the time. Getting cited isn't getting clicked. We've written about how to actually get cited in Google's AI Overviews, but be honest with yourself about what citation buys you — brand impressions and reference-grade authority, not a traffic line on your dashboard.

3. Ahrefs: the GSC-based CTR decline, with methodology you can audit

Ahrefs is a vendor. They also run the largest publicly documented study of AI Overview impact using actual Google Search Console data, which is why their numbers belong in the high-independence tier despite the vendor label.

Their original study analyzed 300,000 keywords and found the top-ranking page lost 34.5% of its CTR when an AI Overview appeared above it (original methodology). Their December 2025 update, drawing on more GSC properties and a maturing AI Overview rollout, sharpened the picture by position:

• Position one: −58% CTR

• Position two: −50.8% CTR

• Position three: −46.4% CTR

Read the December 2025 update for their cohort logic. The methodological caveat worth carrying: this is relative CTR loss on queries where AI Overviews appear, not your overall organic CTR across the site. If AI Overviews trigger on 30% of your impressions, your blended decline is smaller than the headline number suggests — though still substantial.

4. The Great Decoupling: impressions up, clicks down

If you've watched your Search Console graph diverge over the past six months — the blue line climbing while the green line falls — that's the pattern these studies describe at scale.

Google is showing your content to more people because AI Overviews pull from a wider pool of sources than the traditional ten blue links ever did. The Pew and Ahrefs numbers tell you what happens after the impression: fewer of those people need to click to get their answer. Impressions are no longer a leading indicator of sessions. They're a leading indicator of brand surface area — which matters, but doesn't pay the hosting bill.

In our SEO work, we now report impressions and clicks on separate axes, with a third line tracking qualified sessions — visitors who scroll past the fold, read more than 30 seconds, or hit a conversion event. The headline CTR number is becoming less informative every quarter. The qualified-session line is becoming more so.

5. The value side: what an AI-search visitor is worth (vendor data, flagged)

Here's where the source grading matters most. The most-cited statistic for AI-search visitor value comes from Semrush, which sells AI-search SEO tooling. Their study values an AI-search visitor at 4.4x an organic visitor and projects AI search will overtake traditional search for studied topics by early 2028 (Semrush study).

The 4.4x number is directionally consistent with what we see in client analytics — AI-search visitors arrive with more context, ask sharper questions, and convert at higher rates than cold organic — but Semrush hasn't published its full sample composition, and a vendor with a tool to sell has every incentive to make the multiple look generous. We treat it as a useful hypothesis, not a planning input. If you're modeling pipeline against AI-search traffic, hedge that multiple down before you commit headcount.

The early-2028 forecast is a forecast. Semrush forecasts it. Nobody has lived through it yet.

6. What these numbers mean for a small site's KPI targets

If you're running a site under 100,000 monthly sessions, the right response to these statistics is recalibration, not retreat. Three concrete moves we're making with clients this quarter:

First, rebaseline. Pull your last 12 months of GSC data and tag queries where AI Overviews now appear. Your CTR target for those queries should be roughly half what it was 18 months ago. That's not failure; that's the new floor. Plan against it.

Second, shift the editing budget from new posts to depth on the posts that already rank. The Ahrefs position-one drop hurts most if you're at position one on thin content, because thin content is exactly what AI Overviews summarize away. We've written the playbook for updating old blog posts — the short version is that depth, original data, and named expertise are what survive the summary.

Third, track citation rate as a separate KPI from click rate. A 1% click-through on cited sources sounds bleak until you remember that being cited in an AI Overview puts your brand name in front of a user who is, by definition, mid-decision. That impression has value even when the click doesn't fire.

FAQ

How much do AI Overviews reduce click-through rates in 2026?

The most-cited independent numbers: Pew Research found organic clicks fall from 15% to 8% when an AI Overview appears (July 2025). Ahrefs measured a 58% CTR drop at position one in its December 2025 GSC-based update, down from 34.5% in its original 300,000-keyword study.

What are the most reliable zero-click search statistics for 2026?

Pew Research's July 2025 study is the highest-independence source: 900 US adults, 68,879 tracked Google visits, nationally representative. Ahrefs' December 2025 GSC update is the best vendor-run dataset because the methodology is public and the sample is real Search Console data, not modeled estimates.

Are AI Overviews killing SEO traffic for small sites?

Not uniformly. Sites publishing thin, summary-style content are losing the most because AI Overviews replace exactly that content type. Sites with original data, named expertise, and depth see impressions rise while CTR falls — a net pattern that rewards qualified sessions over raw clicks. Plan against qualified sessions, not blended CTR.

What's the click-through rate on sources cited inside an AI Overview?

Pew Research found users click through to a cited source roughly 1% of the time when an AI Overview is shown (July 2025). Getting cited generates brand impressions and reference authority, but it doesn't drive meaningful direct traffic. Treat citation as a brand metric, not a traffic channel.

Is AI search going to overtake traditional search?

Semrush forecasts AI search will overtake traditional Google search for studied topics by early 2028, based on their 2025 vendor study. That's a forecast from a company selling AI-search SEO tools, so treat it as directional. Independent research hasn't published a comparable projection yet.

What to do this week

Pull your top 50 ranking URLs from Search Console. Tag the queries where AI Overviews now appear. Cut your CTR target on those queries in half. Then pick the five posts with the most impressions and the steepest click decline, and rewrite them for depth — original data, named author, specific numbers — not breadth.

If you want a second pair of eyes on which posts to prioritize, send us the list. We'll read it the same way we read our own.