What is a zero-click search, and which SERP features cause it?
A zero-click search is a search that ends without a click from the search engine results page (SERP) to a website. Instead of routing users to a source page, the search engine shows the answer itself, so the user finds what they need in the search results, refines the query, or leaves. Several SERP features can answer a query directly:
- AI Overviews and AI Mode in Google, and AI-generated answers in other search engines and assistants, such as Bing and Copilot.
- Featured snippets, which Google describes as boxes that show the descriptive snippet before the title and link.
- Knowledge panels, which are information boxes about entities in Google's Knowledge Graph, such as a company, a person, or a place.
- Direct answers, such as definitions, conversions, and quick facts.
The clearest examples are practical questions. For local businesses, opening hours and directions appear on the results page itself. Quick questions, such as a currency conversion or the definition of a word, are answered directly, so there is nothing left to click. Voice search and voice assistants that read out one answer are the extreme case.
Featured snippets are the oldest of these SERP features. Google says its systems choose a featured snippet automatically, that you can't mark up a page to earn one, and that the nosnippet setting keeps a page out of snippets. A featured snippet can satisfy the query without a visit, because the answer comes before the title and link.
Zero-click searches aren't new. Traditional search engines have answered quick questions with snippets and knowledge panels for years. What changed is that AI-generated answers now cover research questions too, so a zero-click result can be the first time a buyer hears your name. Zero-click behavior is now part of normal search behavior, so plan for it instead of treating it as an anomaly. Definitions also differ between studies. SparkToro, for example, counts a click on an ad or on a Google service such as Maps or YouTube as a click, and a follow-up search as no click, so compare numbers within one study and not across them.
How many zero-click searches are there, and what do AI summaries change?
About two in three US Google searches ended without a click in early 2026, according to SparkToro's clickstream study, and Pew found that an AI summary lowers clicks further. Google says organic click volume to websites has been relatively stable.
| Source | Data and period | Finding | Limit to keep in mind |
|---|---|---|---|
| SparkToro, June 2026 | Similarweb browser clickstream, US, January to April 2026 | 68.01% of Google searches were zero-click searches, against 60.45% in 2024 | The 2024 and 2026 figures come from different panels, and Google's mobile app is not covered |
| Pew Research Center, July 2025 | Browsing of 900 US adults in March 2025, covering 68,879 Google searches | An AI summary appeared on 18% of searches. People clicked a traditional result in 8% of visits with a summary and 15% without, clicked a link inside the summary in 1%, and ended the session in 26% of visits with a summary against 16% without | One month, one search engine, and adults who agreed to be tracked |
| Google, August 2025 | Google's own data | Total organic click volume to websites has been relatively stable year over year, and average click quality has increased | Google published no figures or method |
| Gartner, March and May 2026 | Survey of about 645 B2B buyers, August to September 2025 | 45% used AI during a recent purchase, and 69% prefer to validate AI-generated insights with sales reps | Self-reported, and the releases give no industry or region split |
The sources measure different things. SparkToro counts all searches in a browser panel. Pew compares visits to search results pages with and without an AI summary. Google describes click volume across all sites and publishes no figures. Pew's numbers show the mechanism: when a summary appears, search engine users click a traditional result less often, almost no one clicks a link inside the summary, and more people end the session. Google's statement and SparkToro's estimate can both hold if total search volume grows while the share of searches that send a click falls. Both studies look at Google, so they say nothing about how people behave on other search engines, such as Bing.
Use of AI-generated summaries is growing. At Google I/O on May 19, 2026, Sundar Pichai said Google AI Overviews had passed 2.5 billion monthly active users and AI Mode had passed 1 billion. SparkToro's panel saw only 0.34% of US searches move into AI Mode between January and April 2026. The two figures count different things: monthly active users in one case, and a share of US searches in a browser panel that leaves out Google's mobile app in the other.
None of this describes your site. Whether your own clicks are falling depends on which search queries you rank for and which of them now show an answer. The next section shows which queries are most exposed, and the sections after it cover how to read Search Console.
Which search queries are most likely to show an AI summary?
Longer, question-style search queries. In Pew's March 2025 data, 53% of searches with ten or more words produced an AI summary, against 8% of searches with one or two words.
| Type of search query | Share that produced an AI summary |
|---|---|
| One or two words | 8% |
| Ten or more words | 53% |
| Contains both a noun and a verb | 36% |
| Starts with who, what, when, or why | 60% |
Pew's data covers all topics, not B2B software, so the B2B reading is ours. Long research questions, such as how to compare two platforms or what a tool in your category should include, are the most exposed. Short brand-name searches are the least exposed. Treat that as a hypothesis to test on your own search queries, because Pew measured AI summaries and not zero-click searches directly.
Search Console can run the test. A practical first pass is to sort queries by impressions and look for those whose click-through rate fell while impressions held or rose, because those are your likeliest zero-click queries. Then filter the query dimension with a regular expression that matches question words, and compare two date ranges to see whether clicks fell more on those search queries than on the rest. Google's documentation describes both the regular expression filter and the date comparison.
Why do zero-click searches matter differently for B2B SaaS?
Because the buyer who doesn't click today may still buy. B2B purchases involve several people and several months, and much of the research happens before anyone contacts sales. Gartner's survey of about 645 B2B buyers, fielded in August and September 2025, found that 45% had used AI during a recent purchase, primarily to gather information on vendors and products, and that buyers used an average of seven information sources. Gartner also reported that 69% of buyers prefer to validate AI-generated insights with sales reps.
That changes what a lost click means. Digital marketing has long treated a click as a stand-in for interest, and zero-click searches weaken that stand-in. If your brand appears in an AI answer while a buyer is learning about the category, the visit may come later as a branded search, a direct visit, or a demo request in which the buyer says an assistant recommended you. If the answer names a competitor instead, you lose more than a visit. Because buyers consult multiple sources, a zero-click answer is one touchpoint in long customer journeys, not the whole story.
Clicks should still matter most on late-stage queries, such as pricing, comparisons, integrations, and security, where buyers need detail that a summary rarely gives. Google makes a related claim about quality. It says the clicks that do arrive from results with AI Overviews tend to be higher quality, with people more likely to spend time on the site, but it publishes no figures you could check against your own data. Our posts on what B2B buyers need from your website and on SaaS pricing pages cover the late-stage pages in more detail.
What should you measure instead of clicks alone?
Measure exposure, demand, and outcomes, and read them together. Website traffic stays on the list, and organic traffic is still the starting point, but neither can be the only signal. A digital marketing report built only on sessions and clicks will show a drop that pipeline may not confirm.
| Metric | Where to find it | What it tells you | Caveat |
|---|---|---|---|
| Impressions and click-through rate by page | Search Console, Performance report, Web | How often you appear in search results, including inside AI features | An AI Overview counts as one position shared by every link in it |
| Generative AI impressions | Search Console, Generative AI performance report | How often links to your site appear in AI Overviews and AI Mode | Impressions only, no clicks, and already part of the Web totals |
| Branded search | Search Console, branded queries filter | Whether demand for your name is growing | Unavailable to sites with few impressions, has data only from March 11, 2025, and some queries are misclassified |
| AI citations in Bing and Copilot | Bing Webmaster Tools, AI Performance report | Total citations, cited pages, and a sample of grounding queries | Announced in public preview, and shows no placement or ranking |
| AI assistant referrals | GA4, AI Assistant channel | Visits from ChatGPT, Gemini, Copilot, and similar AI tools | Excludes AI Overviews and AI Mode, which count as Organic Search |
| Self-reported source | A "How did you hear about us?" field on demo forms | Influence that tracking can't see, such as an assistant recommending you | Answers can favor the latest or most memorable touch |
| AI mention rate | A fixed set of buyer prompts, checked monthly | Your query coverage: the share of prompts where assistants name your brand | Answers vary between runs |
| Demo requests and pipeline | CRM, by first source and self-reported source | Whether visibility turns into revenue | Long sales cycles delay the signal |
Google suggests tracking conversions and time spent on your site in other tools, such as Google Analytics, alongside Search Console. Direct conversions from organic search, such as a demo request on the landing page, will undercount influence when buyers read the answer first and return later by another route. Our posts on AI traffic in GA4, attribution without an attribution tool, and a 30-prompt method for tracking brand mentions cover three of these rows in detail.
How do you read Search Console when clicks fall but impressions rise?
First check what changed in how results are counted, and which SERP features now appear on your key queries, because average position and click-through rate can move without any change in your rankings. Google's documentation says an AI Overview occupies a single position, and every link in it shares that position. A click on a link to an external website in an AI Overview counts as a click. AI Mode follows the same position methodology as a regular results page, and a follow-up question in AI Mode counts as a new query.
| Pattern | Possible reading | What to check next |
|---|---|---|
| Impressions up, clicks flat, click-through rate down | You appear in more answers that don't need a click | The Generative AI report for the same pages |
| Branded impressions up, non-branded clicks down | Answers send buyers to search for you by name | The branded filter, and demo requests from branded search |
| Non-branded clicks down on guides only | Early research questions are answered on the page | Conversions from those pages, not just traffic |
| Clicks down on one query while impressions hold | A new featured snippet or AI Overview now answers it | The live results page for that query |
| Clicks down on pricing or comparison pages | A real loss of late-stage traffic | Rankings, title links, and competitors' pages |
These readings are starting points, not diagnoses. Each one needs a second data source before you act on it.
Google also added a Generative AI performance report, which it rolled out to all websites worldwide on August 31, 2026. It shows impressions from AI Overviews and AI Mode by page, country, date, and device, with a filter for text and multimodal searches. It shows no clicks, it may not appear for sites with few AI impressions, and Google says it includes data from the Web search type, so don't add its numbers to your Web totals.
A rough check is to divide a page's generative AI impressions by its Web impressions for the same dates. A rising ratio on pages whose clicks are flat suggests more of your search visibility is happening inside answers. Treat it as a direction and not an exact share, because Google says the newest data can be preliminary.
How should your content strategy respond to zero-click searches and AI search?
Write pages that search engines and AI answers can use, and keep pages that buyers still need to visit. A good content strategy does both. Google's guide to generative AI search treats optimizing for it as SEO, so this work belongs in your SEO strategy, not beside it. Marketers increasingly call it generative engine optimization (GEO), but the underlying search engine optimization practice is the same.
Answer clearly and early. Google says there are no special optimizations for AI features: a page has to be indexed and eligible to appear with a snippet. Microsoft's guidance adds that AI assistants split a page into smaller pieces, and that clear headings mark where one idea ends and the next begins. A question-based heading with a direct answer in the first lines gives each piece a label, and it is also the kind of page that can be selected for a featured snippet.
Choose content formats that answers can reuse. Microsoft's guidance lists Q&A blocks among the formats that assistants can reuse easily, so structure content in short sections that each answer one question. Match each page to user intent: an early research question deserves a direct answer, while a pricing or comparison page deserves detail.
Make facts attributable. A dated figure with a named source is easier for an answer to cite and easier for a reader to check.
Invest in late-stage pages. Pricing, comparison, integration, and security pages, and customer stories, hold the detail a summary can't replace.
Accept that some answers won't earn a click. If an AI summary names your company correctly and the next step is easy to find, a zero-click answer still does part of your marketing work. Zero-click opportunities are worth taking when the answer is accurate, and not worth forcing when the only way to earn the click is to hide the answer.
How do you protect brand visibility in AI responses and knowledge panels?
Make your brand easy to identify, then check how it appears.
Start with structured data. Schema.org is the shared vocabulary for embedding structured data in a site, and Organization is one of its types. Organization markup helps search engines understand which entity your site represents: Google says some properties help it tell your company apart from others, and a logo property can change the logo shown in your knowledge panel. Microsoft says schema helps search engines and AI systems understand your content, while Google says no special schema markup is needed for its AI features. Google also says structured data must match what readers can see on the page, and FAQ rich results stopped appearing in Google Search in May 2026, so don't add markup for its own sake.
Check your knowledge panel appearances. Google describes knowledge panels as information boxes about entities such as organizations, generated automatically when enough information exists on the open web. An entity has to be in Google's Knowledge Graph to trigger one, and the graph compiles facts from public sources, licensed data, and information that content owners send. An official representative can claim a panel and suggest edits, and Google says it may remove information from a knowledge panel that is demonstrably false or outdated when reliable evidence supports that. Search your company name on a schedule, and note whether a panel appears and whether it describes your brand accurately.
Keep entity data consistent. Accurate company profiles and content that answers questions directly are the basics of a zero-click strategy. Use the same company name, logo, and description on your site, in your Organization markup, and on profiles such as LinkedIn, G2, and Crunchbase, so that search engines and AI systems reading them find the same description of your brand.
Build brand demand. Branded search is one of the few signals that can show whether zero-click visibility turns into interest in your company, though it shows correlation and not proof. Being named in AI-generated responses may build awareness even without a click, but that is hard to prove, so test it against branded search and self-reported source. Whenever your brand appears in an answer, a clear next step on your own site, such as a pricing page, gives the interest somewhere to go.
How do you report the impact of zero-click searches to leadership?
In three layers, so that a digital marketing report shows a drop in organic traffic in context and not as a failure.
- Exposure: search impressions, generative AI impressions, and AI citations, for the pages that matter commercially.
- Demand: branded search impressions and clicks, direct traffic to key pages, and AI assistant referrals.
- Outcomes: demo requests and pipeline by first source and by self-reported source.
Add one line on what changed in the search results themselves, such as a new AI Overview or another SERP feature on a key query. In a zero-click world, the useful question over a quarter isn't whether clicks fell. It's whether exposure and demand held while pipeline grew.
We report on organic search and AI visibility for B2B SaaS companies, so we have an interest in this topic. For how we measure mentions and citations, see our AEO agency page.
Questions about zero-click searches
What is a zero-click search?
A search that ends without a click from the results page to any website. The searcher gets the answer from a featured snippet, knowledge panel, direct answer, or AI summary, refines the query, or leaves.
How many Google searches are zero-click?
SparkToro's clickstream study estimated that 68% of US Google searches ended without a click from January to April 2026, against 60% in 2024. The two years use different data panels, so treat the comparison as rough.
Which search queries are most likely to get an AI summary?
Long, question-style queries. Pew found AI summaries on 53% of searches with ten or more words, 60% of searches that began with who, what, when, or why, and 8% of one- or two-word searches.
Do clicks from AI Overviews show up in Search Console?
Yes. Google counts a click on a link to an external page in an AI Overview or in AI Mode as a click in the Performance report. An AI Overview occupies a single position, and every link in it shares that position.
Can you see AI Overview impressions in Search Console?
Yes, in the Generative AI performance report, which Google rolled out to all websites worldwide on August 31, 2026. It shows impressions by page, country, date, and device, but not clicks, and sites with few AI impressions may not see it.
Are zero-click searches bad for B2B SaaS?
Not automatically. Pew found fewer clicks when an AI summary appears, while Google says overall organic clicks have been stable. The effect depends on which queries lose clicks: early research questions or late-stage pages such as pricing and comparisons. Track impressions, branded search, and demo requests, not clicks alone.
Sources
- Pew Research Center: Google users are less likely to click on links when an AI summary appears in the results Accessed October 19, 2026
- SparkToro: In 2026, less than one-third of Google searches still send a click Accessed October 19, 2026
- Google, The Keyword: AI in Search is driving more queries and higher quality clicks Accessed October 19, 2026
- Google, The Keyword: Google I/O 2026, Sundar Pichai's opening keynote Accessed October 19, 2026
- Gartner: Sales survey finds 67% of B2B buyers prefer a rep-free experience (March 9, 2026) Accessed October 19, 2026
- Gartner: Survey finds 69% of B2B buyers turn to sales reps to validate AI-generated insights (May 20, 2026) Accessed October 19, 2026
- Search Console Help: What are impressions, position, and clicks? Accessed October 19, 2026
- Search Console Help: Generative AI performance report (Search) Accessed October 19, 2026
- Search Console Help: Performance report (Search results): Dimensions and data groupings Accessed October 19, 2026
- Search Console Help: Performance report (Search results): Common tasks and use cases Accessed October 19, 2026
- Google Search Central: AI features and your website Accessed October 19, 2026
- Google Search Central: Optimizing for generative AI search Accessed October 19, 2026
- Google Search Central: Featured snippets and your website Accessed October 19, 2026
- Google Search Central: Organization structured data Accessed October 19, 2026
- Google Search Central: Documentation updates Accessed October 19, 2026
- Knowledge Panel Help: About knowledge panels Accessed October 19, 2026
- Knowledge Panel Help: How Google's Knowledge Graph works Accessed October 19, 2026
- Microsoft Advertising: Optimizing your content for inclusion in AI search answers (October 8, 2025) Accessed October 19, 2026
- Bing Webmaster Blog: Introducing AI Performance in Bing Webmaster Tools Public Preview (February 10, 2026) Accessed October 19, 2026



