Google AI Mode had more than 100 million monthly active users in the United States and India by July 2025, according to Alphabet’s Q2 2025 earnings transcript. For publishers, the commercial question is what that scale means for the readers who previously reached their sites through traditional search.
The studies cited below document substantial referral declines and lower click rates on results pages containing AI summaries. Each measures a different part of the shift: product usage, observed search behaviour or publisher traffic over time. Understanding those measures and their limits is essential before applying them to an individual blog.
AI Mode and AI Overviews are distinct experiences within Google’s search business. AI Overviews place summaries in search results; AI Mode provides a conversational search interface. So the AI Mode usage figure says nothing directly about how often people click through from AI Overviews.
For bloggers, the implications are practical: a business that depends heavily on search referrals needs to understand how those referrals are changing. The evidence gives real cause for concern about that model, though aggregate news-publisher data will not map neatly onto an independent site whose audience, content and sources of income may differ from those of the publishers measured.
What the traffic data actually shows
The decline has been building for more than a year, but the 2026 numbers have sharpened the picture considerably.
Chartbeat data published in the Reuters Institute’s 2026 trends report showed that Google search traffic to publishers fell by a third globally between November 2024 and November 2025, and by 38% in the United States. Google Discover referrals dropped 21% in the same period. Media executives surveyed by the Reuters Institute now expect search referrals to decline by a further 43% over the next three years. A fifth of respondents project losses above 75%.
A Pew Research Center analysis used browsing data from 900 U.S. adults and 68,879 unique Google searches. Users clicked traditional results on 8% of visits to pages with an AI summary, compared with 15% on pages without one. Links in the summaries received clicks on 1% of visits. Because Pew observed real browsing rather than running a controlled experiment, the gap is an association: it does not show that AI summaries alone caused the difference, or that any particular blog lost the same share of visits.
SE Ranking’s analysis of AI Mode citations found Google.com accounted for 17.42% of all citations in its sample. Of that Google.com subset, 59% pointed to organic Google search results. That works out to roughly 10.3% of all sampled citations. The study counted Google.com separately from other domains, including YouTube. The finding indicates that a share of AI Mode citations send users further into Google Search rather than out to external sites.
Which content gets hit hardest
The damage is not evenly distributed. Content categories that depend on informational queries — the bread and butter of most independent blogs — are absorbing the steepest losses.
Travel guides, recipe sites, product comparisons, how-to content, and educational resources face the greatest exposure because these are exactly the queries AI systems can resolve without requiring a click-through. When someone searches “best things to do in Lisbon” or “how to fix a leaking faucet,” AI Mode can synthesize an answer from multiple sources and present it as a complete, conversational response. The blog that originally researched and wrote that information may be cited in a footnote — or may not appear at all.
Chegg, the education platform, reported a 49% decline in non-subscriber traffic between January 2024 and January 2025, as described by Search Engine Journal. The company attributed pressure on its business to AI Overviews and filed an antitrust lawsuit against Google in February 2025. Those figures reflect one company’s experience, and its account of the cause forms part of its allegations; they are not a measure of the effect on educational sites generally.
Lifestyle and utility content — weather, TV listings, recipes, travel logistics — has been hit harder than hard news, partly because Google has so far largely exempted breaking news queries from AI Overviews due to the risk of hallucination. But that exemption is a policy decision, not a technical limitation, and there is no guarantee it will persist.
The counterargument — and what it misses
Google maintains that AI Overviews drive more clicks to supporting websites than traditional search when those sites appear within the overview. Some data supports a version of this claim. Research from Seer Interactive found that brands cited within AI Overviews earned 35% more organic clicks than brands that were not cited. AI Mode traffic that does reach external sites reportedly converts at 14.2% compared to 2.8% for traditional search traffic, with visitors spending 38% longer on site.
What that misses is that lower referral volume can still create a commercial problem even if the visitors who arrive are more engaged. The revenue effect depends on the site’s business model and on what those visitors do. A 50% drop in visits would imply 50% fewer affiliate clicks or signups only if the relevant rates per visit stayed constant; higher conversion could offset part of the decline. Display-ad income also depends on impressions per visit, ad rates and other factors. Publishers need to measure those outcomes rather than assume that traffic and every revenue stream move in identical proportions.
The Reuters Institute found that publisher confidence in journalism’s future has dropped to 38%, down from 60% four years ago. Most publishers surveyed plan to reduce investment in traditional Google SEO in 2026. The direction they are moving instead — original investigations, contextual analysis, community building, video, direct audience relationships — reflects a broad expectation among publishers that search traffic will keep shrinking.
What this means for bloggers right now
The strategic implications for independent publishers are straightforward, even if the execution is not.
The display-ad-plus-organic-search model is no longer a reliable foundation. Bloggers who depend on Google for the majority of their traffic and monetize primarily through ad networks face the most acute version of this problem, and the data suggests it will get worse before it stabilizes.
The publishers showing resilience in 2026 share common characteristics: they built email lists before the traffic decline began, they diversified into revenue streams that are audience-dependent rather than traffic-dependent — paid newsletters, memberships, digital products, direct sponsorships — and they invested in content formats that resist AI summarization: first-person expertise, original reporting, proprietary data, and subjective recommendations that require the reader’s trust in a specific author.
None of this is easy. And none of it guarantees survival. But the alternative — continuing to build primarily on a traffic source that is actively being redirected into a closed ecosystem — is a bet against the data.
The scale of AI search and the documented fall in referrals make this a serious distribution question for independent publishers. Industry-wide figures can only go so far. The useful question for each publisher is how its own traffic and revenue have changed, and which audience relationships can make the business less dependent on a single platform.
