Pew Study: Google AI Overviews Choke Web Traffic, Endangering Open Web
Pew data confirms AI Overviews are devastating website traffic, imperiling the ad-supported open web's future.
July 23, 2025

A landmark study from the Pew Research Center has provided stark, quantitative evidence for a shift in online behavior that publishers and content creators have long feared, revealing that Google's AI-generated summaries are drastically reducing the flow of traffic to external websites. The research indicates that when Google presents an AI Overview at the top of its search results, users are significantly less likely to click through to traditional web links, with a vanishingly small number engaging with the source links embedded within the AI summaries themselves. This fundamental change in user interaction with the world's dominant search engine signals a potential existential threat to the ad-supported open web, forcing a reexamination of the symbiotic relationship between Google and the creators who supply its informational lifeblood.
The data, gathered by tracking the browsing habits of 900 U.S. adults during March 2025, paints a clear picture of user disengagement from the traditional "blue links." According to the Pew study, when a search page included an AI Overview, users clicked on a standard website link only 8% of the time.[1][2] In stark contrast, on pages that returned only traditional search results, that figure nearly doubled to 15%.[1][2] The most alarming statistic for publishers, however, is the click-through rate for the source links provided within the AI-generated summaries. The research found that a mere 1% of users who encountered an AI Overview clicked on one of the cited source links.[1][3][4] This suggests that users are increasingly satisfied with the synthesized answer provided by Google, feeling little need to explore the original source material for deeper context, verification, or further information.
The Pew analysis, which covered nearly 69,000 unique Google searches, found that AI Overviews appeared in roughly one out of every five searches, or 18% of the time.[5][3] This figure rises dramatically with the complexity of the query. While only 8% of searches with one or two words triggered an AI summary, that number jumped to 53% for queries containing ten or more words.[6][2][7] Question-based searches, such as those beginning with "what" or "how," were also far more likely to generate an AI response.[6][7] The study also observed that the presence of an AI summary correlated with a higher rate of search session abandonment. On pages featuring an AI Overview, 26% of user sessions ended without any clicks, compared to 16% on pages with only traditional results.[5][1][6] This suggests that for a significant portion of users, the AI-generated answer is the final destination, not a gateway to further online exploration.
The implications of this traffic shift are profound and far-reaching, particularly for the digital publishing and media industries, which have long relied on Google search as a primary driver of audience and revenue. For years, the implicit contract of the web was that creators produce high-quality content, and search engines direct users to that content, allowing publishers to monetize through advertising or subscriptions. AI Overviews disrupt this model by effectively scraping and summarizing publisher content to provide direct answers on the search results page, keeping users within Google's ecosystem. This validation of long-held publisher fears comes at a time when many news outlets are already in a precarious financial state. The phenomenon of "zero-click searches," where a user's query is answered on the results page itself, has been a growing concern, and AI Overviews appear to be accelerating this trend dramatically.[8]
In response to the study, Google has disputed the methodology, claiming it is flawed and not representative of overall search traffic.[9] The company maintains that it continues to send billions of clicks to websites daily and has argued in the past that AI features can create new opportunities for users to connect with a greater diversity of websites, especially for more complex questions.[9][10] However, Google has not released public data to counter the specific findings of the Pew study or to demonstrate the volume of traffic originating directly from AI Overviews.[5][11] The study also highlighted which platforms are most frequently cited by the AI, with Wikipedia, YouTube (a Google property), and Reddit collectively accounting for 15% of the sources listed in the summaries.[5][7][12] This concentration of sourcing could further consolidate traffic around a few dominant platforms, to the detriment of smaller, independent publishers.
In conclusion, the Pew Research Center's findings provide the most comprehensive data yet on the tangible impact of generative AI on search engine behavior and the broader web ecosystem. The dramatic reduction in clicks to both traditional search links and the sources cited within AI Overviews confirms the publishing industry's deepest anxieties about its future in an AI-dominated landscape. As users increasingly find their answers without leaving Google, the economic model that has sustained a diverse and open web for decades faces an unprecedented challenge. The debate is no longer about if AI will change the internet, but by how much, and what will be left of the open web once the transformation is complete.