Skip to main content
Answer Engineered

Verified: 2026-07-18By Answer Engineered Research

The Position-1 CTR Collapse: From 7.3% to 1.6% in Two Years

Ahrefs 300K-keyword data: position-1 CTR on AI Overview keywords fell from 7.3% to 1.6% between Dec 2023 and Dec 2025. The zero-click economics behind the drop.

Published July 18, 2026

What exactly did Ahrefs measure?

The Ahrefs study tracked 300,000 keywords — 150,000 that trigger an AI Overview and a 150,000-keyword control set that does not — and compared position-1 organic CTR across two years. On the AI Overview set, position-1 CTR fell from 7.3% in December 2023 to 1.6% in December 2025. To isolate how much of that is attributable to AI Overviews rather than other SERP changes, Ahrefs modeled a counterfactual: without an AI Overview present, position-1 CTR in December 2025 would have been approximately 3.7%. The actual 1.6% against that 3.7% counterfactual is the 58% reduction Ahrefs attributes to the AI Overview itself. Ahrefs notes the figure will likely grow.

This is the second wave of the same measurement. An earlier Ahrefs study using a March 2024 versus March 2025 window found a 34.5% drop; the 58% figure is the newer, larger, and longer-window result, and supersedes it. The methodology is US-centric and applies to informational, AIO-triggering queries — the number should not be quoted as a universal CTR change across all search.

Was search traffic healthy before AI Overviews?

No — and the baseline matters for reading the drop correctly. SparkToro’s 2024 clickstream analysis found 68.01% of US Google searches already ended without any click to the open web. And Ahrefs’ own 14-billion-page study found 96.55% of pages receive zero organic traffic from Google at all — December 2023 data, sampled toward the quality side of the web, and never updated, but a sobering as-of-2023 baseline. The same vendor’s ranking research shows only 1.74% of new pages reach the top 10 within a year across all URL types (6.11% for the non-empty English-content subset).

The AI Overview era did not break a healthy system. It compressed the payout of the narrow slice that was still paying.

What does this look like in publisher traffic?

Two independent data releases sketch the downstream effect. Google search referrals to publishers fell 33% globally between November 2024 and November 2025 across more than 2,500 sites, per Reuters Institute analysis of Chartbeat data. A separate Chartbeat pull reported in March 2026 showed the pain distributed unevenly: small publishers down 60%, large publishers down 22%. These are two distinct releases with different windows — the small-publisher timeframe is disputed even in secondary coverage — and should never be quoted as one study. But both point the same way, and both are consistent with what a 7.3-to-1.6 CTR change at the top of the SERP would predict.

Scale explains the speed. Google AI Mode reached 1 billion monthly active users within 12 months of its May 2025 US launch, and ChatGPT stood at 900 million weekly active users as of May 2026. The surfaces eating the clicks are already mass-market.

Is there any upside in the same data?

Yes, and it is the most underreported number in this story. Seer Interactive’s study of 53 brands and 5.47 million queries found that pages actually cited inside an AI Overview saw a 91% CTR uplift versus the same organic position without one. The AI Overview does not delete the click market; it re-routes it through the citation slot. The losers are ranked-but-uncited pages — and that group is growing, since top-10 organic pages’ share of AIO citations fell from 76% to 38% between July 2025 and March 2026 in Ahrefs’ 863,000-keyword data (with Ahrefs’ own caveat that improved parsing accounts for part of the measured drop).

There is also a quality effect on the clicks that survive. Orbit Media’s 2026 analysis measured ChatGPT referral traffic converting at 15.9% versus 1.76% for organic search. Fewer visitors arrive, but the ones who do have already had their comparison shopping done by the answer engine.

What are the zero-click economics that follow?

Three consequences fall out of the arithmetic.

Traffic-based revenue models need lower baselines. Any plan built on pre-2024 CTR curves overstates expected visits on informational queries by a factor of roughly four at position one, per the Ahrefs data above. Display-advertising economics, which pay per pageview, absorb that reduction directly.

The competitive unit shifts from the ranking to the citation. If being cited carries a 91% CTR premium and being merely ranked carries a 58% AI-attributable penalty, the rational optimization target on AIO queries is the citation, not the position. The two are increasingly independent: 61.7% of AI citations point to URLs outside the organic top 100 in Profound’s 340-million-prompt dataset.

Citation positions are rented, not owned. Unlike a defensible ranking, a citation slot churns constantly: SISTRIX measured a 56% weekly domain replacement rate in Google AI Mode, and after Gemini 3 took over AI Overviews on 2026-01-27, 42.4% of previously cited domains dropped out in SE Ranking’s 100,000-keyword study. The operational consequence is that citation visibility is a monitoring discipline, not a one-time achievement — the cost structure of search visibility shifts from campaigns toward continuous measurement.

Impressions decouple from value. A page can be read by an answer engine, summarized, and credited without a visit ever registering. The measurable click becomes a conservative floor on actual reach — an argument for tracking citation presence directly rather than inferring visibility from analytics alone, especially since most AI-referred visits arrive with no referrer header and land in analytics as “Direct.” First-party tooling is starting to close the gap: Google Search Console added a generative-AI performance report in June 2026, though it reports AI Overviews and AI Mode impressions combined rather than separately, and the rollout remains limited as of July 2026.

The bottom line

As of July 2026, position one on an AI Overview keyword pays roughly one click where it paid four and a half two years earlier. The strongest response the data supports is not abandoning search — it is repricing it: build for the citation slot that carries the premium, assume the click volume of 2023 is not coming back, and measure AI surfaces directly instead of reading their effects off a shrinking analytics shadow.

Sources

Monthly AI Visibility Index

Get the monthly index

One email a month: who AI engines cite, who they dropped, and the verified numbers behind each shift. No spam, unsubscribe any time.