Google AI summaries cut CTR to 8%

Google search traffic used to work like a rough bargain. Publishers made pages, Google ranked them, and some share of users clicked through. AI Overviews change that bargain because the answer now often sits on Google’s page, not yours. The big new thing here is that there’s finally a widely cited public number for the damage: Pew Research found users clicked a standard search result 8% of the time when an AI summary appeared, versus 15% when it didn’t. (pewresearch.org) ### What exactly happened? Pew looked at the browsing behavior of 900 U.S. adults who agreed to share tracked web activity, then matched that against Google searches from March 2025. The dataset covered 68,879 unique Google searches, and 12,593 of them produced an AI summary. That gave the study something publishers usually don’t have — a way to compare search behavior with and without AI Overviews at scale. (pewresearch.org) ### Why is 8% such a big deal? Because the comparison number is 15%. On search pages without an AI summary, users clicked a result link nearly twice as often. On pages with an AI summary, they clicked links inside the summary itself only 1% of the time. Users were also more li(pewresearch.org) trip. (searchengineland.com) ### Is this just a layout tweak? Not really. The screen design matters, but the deeper change is that Google is answering the query inside its own interface. Pew’s data suggests the user often gets enough from that synthesized response not to leave Google at all. An April 2026 European Parliament briefing goes further and frames this (searchengineland.com)original publisher to the platform that summarizes it. (pewresearch.org) ### What kind of searches trigger this? Not every query gets an AI Overview. Pew found about 18% of all Google searches in the sample triggered one, but the odds jumped on longer and more natural-language queries. Only 8% of one- or two-word searches got a summary, while 53% o(pewresearch.org)ory queries publishers have historically targeted. (searchengineland.com) ### What does Google say back? Google disputes the study. Its public line is that the methodology is flawed and the query set isn’t representative of overall Search traffic. Google also says it still sends billions of clicks to websites every day and argues that AI features create new chances for people to connect with sites, especial(searchengineland.com) relevant links and can send users to a greater diversity of websites. (searchengineland.com) ### Why are publishers so frustrated? Because even if Google is right in aggregate, individual publishers still need to know what changed for them. And Google’s own documentation says AI feature traffic is bundled into normal Search reporting. There’s no clean built-in breakout showing what came from classic blue links versus AI-drive(searchengineland.com)e from AI summaries themselves. (developers.google.com) ### Why is regulation entering the picture? Because this stops looking like a normal product tweak once it affects revenue, visibility, and which sources get surfaced. The European Parliament briefing argues the issue isn’t only copyright or ad money — it’s also media pluralism and democratic resilience, especially for smaller and regional publishers that depend on search discovery. That’s a much bigger fight than SEO. (europarl.europa.eu) ### Bottom line? The 8% number matters because it turns a vague publisher complaint into a measurable behavior shift. Users are clicking less when Google answers more. And until Google breaks out AI-driven search traffic clearly, the people losing distribution are stuck arguing about a black box. (pewresearch.org)ars-in-the-results/))