Zero-click search is the fear that has shaped publisher strategy since AI answers moved to the top of results pages. New data published on November 2, 2026 adds an important twist. People now read the AI answer first, then take longer to decide where to go, and the clicks that remain look more deliberate. For publishers who depend on search traffic, that changes which pages deserve investment, how traffic should be measured and where trust is won.
This guide is based on "The quiet click," a study by Fractl published with Search Engine Land, plus Search Engine Roundtable's report on Google's brand results. Every number below comes from those sources, and we note where the study itself calls its figures directional.
Table of Contents
What Is Zero-Click Search?
A zero-click search is a search that ends without the user visiting any result. The answer appears on the results page itself, in a featured snippet, a knowledge panel or, increasingly, an AI Overview, and the searcher leaves satisfied or simply stops. For ad supported and affiliate sites, every zero-click search is a pageview that never happens.
The worry is easy to understand. If AI summaries answer more questions, fewer people need to visit the pages those summaries draw from. We covered the early traffic effects in our guide to how AI search is changing publisher traffic. The new study looks past raw click counts at what searchers actually do before they click.
What the Quiet Click Study Measured
According to the Fractl study published on Search Engine Land, the research had three parts. Fractl surveyed 1,007 U.S. adults about how much they trust AI generated search summaries. It analyzed a Datos clickstream panel of roughly 11,100 research and comparison keywords across 10 search sources from 2024 through 2026. It also ran a query audit of 28,327 queries covering about 862.6 million monthly searches in eight verticals.
One caveat matters for anyone quoting the results. The clickstream tables are equal weighted per keyword rather than by session volume, so the authors say the figures describe directional behavior across research queries, not population level click through rates. The study's own summary of the findings is direct: "the data suggests AI is becoming the first stop in search, but not necessarily the final authority."
Readers Still Verify When Stakes Are High
Trust in AI summaries has grown, but it has limits. In the survey, 64% of respondents said they trust AI generated search summaries at least somewhat, while 22% said they distrust them. Yet 77% said they had caught an AI summary being wrong at least once, and 45% said they trust AI results more than they did a year ago.
When an AI answer appears at the top of the results, the most common reaction was to read it and click a link to double check it. Only 16% said they read it and stop there. That is the core finding for publishers: the AI answer is often the first step, not the last one.
The stakes decide how far people go. For a quick, low stakes fact, 55% said an AI summary is enough. When comparing options, 48% said they need to click out, and 71% said they want a real source before a final decision or purchase. Health and wellness showed the strongest preference for a real source at 72%, followed by personal finance and banking at 53%, with education and shopping both at 27%.
Usage also shapes trust. Among people who recently used an AI assistant, 68% trust AI summaries at least somewhat, compared with 34% of non users. By generation, 81% of baby boomers expressed at least some trust in AI search summaries, compared with 58% of Gen Z.
Clicks Are Slower and More Deliberate
The behavioral data tells a more nuanced story than a simple rise in zero-click search. In the clickstream panel, sessions that ended without further action, which the study calls zero-click bounces, fell from 43% in 2024 to 34% in 2026. Continued searches climbed from 54% to 63%, while click through rates held steady at 3%.
The clicks that do happen take longer to arrive. Median time to click rose from 2.5 seconds to 6.0 seconds. Reads of 60 seconds or more increased from 9% to 10% of clicks, and visits under 10 seconds decreased from 71% to 67%. In other words, people are spending more time inside search, refining their queries and reading the AI answer, before choosing a page.
The study's interpretation is that the remaining clicks likely carry more intent than before. Its advice to marketers is blunt: "don't compete with AI on information it can summarize in two sentences." For a publisher, that means a basic definition page may lose its visit to the AI answer, while a detailed comparison or a page with original evidence can still earn a qualified reader.
Where the Escaping Clicks Go
Clicks that leave the starting search engine do not necessarily land on a publisher. In the panel, Google's other surfaces accounted for 35% of those destinations, followed by Amazon at 13%, Reddit at 7%, YouTube at 5% and Bing at 3%. About 40% of all exits were hops between search engines.
Publishers competed for a much smaller share. Destinations outside the starting search engine accounted for just 5% of the destinations users reached, down from 6% in 2024. NerdWallet, Forbes and PCMag each drew about 1% of exiting clicks, while Reddit drew more than any individual publisher.
When users did land on content, blog and resource pages accounted for 39% of clicks, followed by product pages at 31%, reviews at 20%, news at 4% and forums at 3%. The study also cites earlier Fractl research across 8,090 keywords and 25 verticals, which found just 7.2% of cited domains appeared in both Google AI Overviews and LLMs. Visibility in one AI system does not guarantee visibility in another.
The Formats That Rank and the Formats People Trust
For the research and comparison queries in the audit, one format dominated the organic slots directly beneath AI answers. Listicles claimed 81% of those winning slots and resource pages another 11%, together 92%. Forums and Reddit represented just 5%.
Trust told a different story. Forums and community threads topped the trust ranking at 21%, followed by news articles at 18%, and AI search summaries and official company websites at 17% each. Forums were also the most clicked format at 22%. Asked why they click after an AI answer, people pointed to a trusted expert or professional (24%), original data or research (23%), real customer experiences or reviews (17%) and a Reddit or forum thread with real people (15%).
The audit also modeled how likely queries were to trigger an AI Overview. Fintech carried the highest modeled likelihood at 46%, HealthTech and Insurance were near 40%, and Wellness was lowest at 34%. The authors stress these are modeled likelihoods, not observed live placements. For tactics on earning citations inside the answers themselves, see our guide on how to get cited by ChatGPT and Perplexity.
Google Is Also Rearranging Brand Results
The layout of results keeps shifting too. Search Engine Roundtable reported on November 2, 2026 that Google has moved AI Overviews for many brand related keywords to the right hand column of the results page, where previously they appeared at the top or middle of the main results. Barry Schwartz wrote that Google is essentially replacing brand knowledge panels with AI Overviews.
For publishers, the practical point is that the position of the AI answer varies by query type, so click patterns for branded and informational searches can move differently. Our earlier look at how AI Overviews affect SEO for blogging covers the basics of optimizing around these panels.
7 Smart Lessons for Publishers
1. Prioritize queries where verification matters
Comparisons, purchases, health and finance push readers past the AI answer. In the survey, 71% want a real source before a final decision or purchase.
2. Make the value beyond the snippet obvious
Original research, expert analysis, methodology, firsthand experience and specific recommendations give readers a reason to click that a summary cannot replace.
3. Show your trust signals
Name the expert, link to primary evidence and show when content was reviewed or updated, especially on high stakes topics.
4. Be selective about commodity content
A basic definition or low stakes fact may increasingly end inside the AI answer. Put depth where it can still earn a click.
5. Measure click quality, not just volume
If clicks are slower and short visits are falling, CTR alone tells less of the story. Track engagement, signups, return visits and assisted conversions by query intent.
6. Use the formats that hold the slot
Listicles and resource pages held 92% of the winning organic slots beneath AI answers in the audit. Well sourced roundups and resource hubs remain a strong format for consideration queries.
7. Earn the human signal
Expertise, original data and real experiences were the top reasons to click. Build those into your pages rather than chasing every informational keyword.
Zero-click search is real, but this study suggests it is not the whole picture. AI answers are absorbing the quick, low value visit, while readers who need proof still click and take more care choosing where. Publishers that build pages worth verifying against will keep earning those quieter, higher intent clicks.