AI Search Adoption Statistics (2026 Edition)

Zero-click behavior is not new — Google's Featured Snippets, Knowledge Panels, and direct answer boxes have been driving no-click queries for years. What has changed is the scale and the sophistication of the answer layer. AI Overviews can now resolve complex, multi-part questions that previously required a user to visit a page to get an answer. That expansion of zero-click surface area is what makes the 58.5% figure strategically significant in 2026 in a way it was not in 2019.

The geographic split is worth noting: EU searches show a slightly higher zero-click rate (59.7%) than US searches (58.5%), according to Semrush's data. The EU/US difference may reflect differences in query type distribution, device mix, or user behavior patterns rather than AI Overview prevalence specifically.

What the statistic means for brand strategy: the traditional model of "get clicks from search, convert clicks to customers" is operating in a shrinking slice of the search landscape. The other 58.5% of searchers are getting answers — some of which include brand names and recommendations — without ever visiting a website. The brands mentioned in those answers are getting something valuable even without a click: presence in the buyer's consideration set.

This is the most methodologically rigorous data point on the AI Overview impact question because it is based on behavioral tracking of real searches rather than panel surveys or modeled estimates. The Pew study followed actual participants and measured what they actually clicked — not what they said they would do. A sample of 68,879 searches is also large enough to be statistically meaningful.

Some important context around the 8% figure:

• This is an average across query types: Informational queries (how, what, why questions) likely show even lower organic CTRs when AI Overviews appear, while transactional queries (buy, price, near me) likely show less dramatic differences.
• 8% is still not zero: When AI Overviews appear, organic clicks do not disappear — they decrease significantly. Brands appearing in the first organic positions below an AI Overview still receive traffic.
• Being cited in the AI Overview itself changes the math: A brand whose content is cited as a source in the AI Overview benefits from visibility even for users who do not click through. This is why appearing inside the answer matters differently than appearing below it.
• The comparison baseline matters: 15% CTR on organic results when no AI Overview appears is already low by historical standards, reflecting years of Google features (Knowledge Panels, Local Packs, Shopping carousels) competing for click attention.

The Pew Research Center study is the primary academic-quality source for this figure. Attribution in any content citing this number should reference "Pew Research Center's July 2025 behavioral study" — not secondary summaries, which often strip the methodological context.

The 15%+ figure represents a significant expansion from the initial limited rollout of AI Overviews in mid-2024. Prevalence has trended upward as Google refines its triggering criteria and user experience. For brands whose target queries skew informational — research-phase queries where buyers are learning rather than buying — the AI Overview prevalence rate is likely meaningfully higher than the aggregate 15% figure.

The intent-based distribution is strategically important:

• Informational queries: "How does [X] work?" "What is [Y]?" "Why does [Z] happen?" — these trigger AI Overviews at high rates and show the largest organic CTR suppression when they appear.
• Commercial investigation queries: "Best [X] for [use case]" "Top [Y] platforms" — AI Overviews appear but less consistently; when they do appear, they may include explicit product or brand recommendations.
• Transactional queries: "Buy [X]" "[Y] pricing" "Order [Z] near me" — AI Overview prevalence is lowest, and Shopping carousels and local packs compete more directly for the click.

Semrush's data also notes that AI Overview prevalence varies by device, geography, and specific query phrasing — there is no single universal prevalence rate. Brands whose queries live primarily in informational intent territory should assume higher-than-average AI Overview exposure and plan accordingly.

This nuance is worth citing honestly because it changes the strategic framing. The headline "AI Overviews kill clicks" is a simplification. The more accurate framing from the Semrush/Datos keyword-level analysis is: AI Overviews shift the distribution of clicks — some users who would have clicked an organic result now read the AI Overview and leave; others who would have left with no click now use the sources cited in the AI Overview as a guide to what to click.

The practical implications of this nuance:

• Being cited as a source in AI Overviews may drive more clicks, not fewer: When your brand or page is cited with an explicit link in an AI Overview, that attribution can generate clicks from users who specifically want to go deeper on what the AI summarized from you.
• Query-level analysis matters more than averages: If your specific target queries show increasing zero-click rates, the aggregate nuance does not rescue you. Conversely, if your queries show AI Overviews driving more source-click behavior, the aggregate alarm may be overstated for your context.
• Informational vs. transactional intent diverges: Semrush's intent segmentation research separately notes that informational queries skew heavily zero-click while transactional queries much less so. The aggregate 58.5% zero-click figure is inflated by informational query behavior.

The honest statistical picture: zero-click behavior is real, significant, and growing — but the causal mechanism is more complex than "AI Overviews = fewer clicks." Brands that appear inside AI Overviews as cited sources experience a different outcome than brands that disappear entirely from the SERP's attention layer.

These figures should be cited with their limitations clearly stated. "Seer Interactive B2B client data" means: a specific set of B2B clients with specific offers, landing pages, and buying cycles. Conversion rates for AI-referred traffic in consumer e-commerce, SaaS self-serve, or media publishing will not necessarily match these figures. They are directional evidence of a real phenomenon, not universal benchmarks to copy into your board deck without context.

The structural reason the premium exists holds broadly even if the specific numbers vary:

• Users who ask ChatGPT or Perplexity for a vendor recommendation are further along the buying journey than users who type a keyword into Google. They are evaluating, not just exploring.
• An AI engine naming your brand in an answer constitutes an implicit recommendation, framing the subsequent visit as "confirming a trusted suggestion" rather than "cold evaluation."
• AI-referred visitors often arrive having already absorbed a synthesis of your competitive positioning from the AI answer — they know why the AI recommended you, which primes them to look for confirmation rather than explanation.

Additionally, Seer's broader AI Overview study from 2025 confirmed that different AI engines cite different mixes of domains. This means the brands winning ChatGPT mentions are not necessarily the same brands winning Perplexity citations — which is why per-engine measurement is more valuable than a single aggregate "AI visibility score."

This is the foundational academic study on the question of whether deliberate content changes can improve AI citation rates. It represents peer-reviewed academic research, not practitioner claims or agency case studies. The specific improvements the study measured:

• Adding statistics: Content that includes verifiable numerical data attributed to credible sources was cited more frequently in generated answers than equivalent content without statistics.
• Adding quotations: Quoting recognized external authorities within content improved citation probability — consistent with how academic and journalistic sourcing conventions signal trustworthiness.
• Improving fluency: Structurally clear, well-written prose extracted more cleanly than technically accurate but hard-to-parse content.
• Adding citations within content: Content that explicitly linked to primary sources and research was more likely to be cited in generated answers than unsourced assertions.

The "up to roughly 40%" figure represents the upper range of improvement across the techniques tested, not an average or guaranteed outcome. Individual results will depend on query type, competitor content quality, and how well the specific techniques are implemented. The GEO paper's value is methodological: it establishes that content decisions measurably affect AI citation rates, which validates the entire practice of deliberate AI search optimization.

For an explanation of how the GEO pipeline works from retrieval through citation, see our post: How Does Generative Engine Optimization Work?

What these statistics do not tell you: which specific prompts your buyers are asking AI engines, whether your brand appears in those answers, what your citation share is relative to competitors, or which of your pages are being cited. The statistics frame the urgency — you need to know the answers to those operational questions to act on that urgency.

A few strategic readings of the data together:

• The 58.5% zero-click figure + the 8% CTR with AI Overviews figure together mean: For informational queries (which make up a large share of total query activity), the traffic available to traditional organic results is declining structurally. Strategies that depend solely on organic click volume are operating in a shrinking market.
• The Seer conversion data means: Even as the volume of AI-referred visitors is currently small for most brands, the quality is high enough that building AI visibility is worth investment at much lower traffic volumes than traditional SEO would justify.
• The GEO paper means: The optimization is real — specific, executable content changes produce measurable improvements in AI citation rates. This is not wishful thinking or vendor marketing; it is peer-reviewed academic finding.

The strategic conclusion: measure your current AI visibility baseline, identify the highest-value prompts where you are currently absent, and build answer-first content that earns citations from ChatGPT, Perplexity, Gemini, Claude, and Google AI Overviews. That is exactly what our free AI visibility audit is designed to start.