SEO SEO-06

Zero-Click Search: How to Build Visibility When the Click Is No Longer the Only Goal

More searches are resolved without leaving Google, but the studies use different methods and can't be compared with each other. How to sort your demand by the value of the click, and decide where to be the source, the brand or the destination.

Published
Read 14 min
Table of contents

When a marketing director sees organic traffic fall, the usual reflex is to look for a culprit: the algorithm, AI Overviews, the content. That diagnostic exercise is necessary, and we cover it separately in Why Your Organic Traffic Is Dropping Even Though Your SEO Is Working. This article starts from a different, more strategic question: assuming more and more searches get resolved without leaving the search engine, where does value remain, and how do you capture it?

Our thesis has three parts. First, the available evidence points in one direction (fewer clicks per search when direct answers appear), but each study uses different data and definitions, so no single figure works as a forecast for your business. Second, not all zero-click searches cost the same; demand has to be sorted by the value of the click. Third, depending on the type of demand, the goal shifts among being the source, being the brand or being the destination. In all three cases, you have to measure in units other than the click.


What the Data Says (and Why the Studies Can’t Be Compared)

The most-cited zero-click studies answer different questions with different methods. This table puts them side by side, with date, data source and stated limits:

Zero-click and AI Overviews studies: what each measured and what limit it states
Study What it measured Main finding Stated or relevant limit
SparkToro and Datos (2024) Clickstream panel (Datos), US and EU, September 2022 to May 2024 58.5% of US searches and 59.7% of EU searches ended without a click to the open web; about 360 clicks to the open web per 1,000 US searches Minimal iPhone coverage; mobile data only from January 2024; ad blockers may undercount paid clicks
SparkToro and Similarweb (2026) Similarweb desktop and mobile web panel, US, January to April 2026 68.01% of searches without a click, versus 60.45% cited for 2024 Excludes the Google mobile app (likely higher there); authors admit that comparing different panels is "apples and oranges"
Pew Research Center (July 2025) Real browsing of 900 US adults in March 2025 and 68,879 Google searches Click on a traditional result in 8% of visits with an AI summary versus 15% without; click on a link in the summary in 1% US only and Google only; summaries were collected afterward with a scraping service (April 2025) and may vary
Ahrefs (February 2026) 300,000 keywords, average desktop CTR from aggregated Search Console data, December 2023 versus December 2025 An AI Overview is associated with a 58% lower position 1 CTR Correlation plus a counterfactual estimate; desktop only; study by an SEO tool vendor
Seer Interactive (April 2026) 53 brands, 5.47 million queries, 2.43 billion organic impressions, January 2025 to February 2026 CTR with an AI Overview hit a low of 1.31% in December 2025 and 2.36% in February 2026; on informational queries with an AI Overview, brands cited averaged 2.07% organic CTR in 2025 versus 0.94% for those not cited Authors state they cannot claim causation; one account weighs heavily on the 0.94% (about 1.61% without it); study by an SEO consultancy

Sources: SparkToro (2024, 2026); Pew Research Center (2025); Ahrefs (2026); Seer Interactive (2026). Consulted October 9, 2026. Each study refers to its own sample and its own definition of "zero-click" or CTR.

The honest reading of the table:

  • The direction is consistent. When a direct answer occupies the top of the results page, fewer people click on traditional links. Pew, Ahrefs and Seer agree on that from different angles.
  • The size is not. Pew’s 8% versus 15% measures real people’s visits over a month; Ahrefs’ 58% compares aggregated position 1 CTRs; SparkToro’s 68.01% measures searches that end without a click of any kind: to the open web, to ads or to Google’s own properties. These are not the same phenomenon.
  • The sources are neither neutral nor definitive. Ahrefs sells SEO tools, Seer is a consultancy and SparkToro sells audience research software. That doesn’t invalidate their data, but read their conclusions with their stated limits in hand. The authors do so themselves: SparkToro acknowledges its series mix different sources, and Seer says it cannot claim causation.
  • The data moves. Seer observed a partial CTR recovery with AI Overviews between December 2025 and February 2026. Every figure in this article is a snapshot of its period.

That’s why we recommend not importing any percentage into your business plan. The relevant figure is the one for your own queries, which you can get in Search Console by comparing impressions, position and CTR by type of query.

A counterpoint from Google belongs in the conversation: the company says that when people click through from results with AI Overviews, those clicks are higher quality (users spend more time on the site), and it recommends looking at the overall value of visits using conversion indicators. That is Google’s claim about its own data. Treat it as a hypothesis to check against your analytics, not as a fact about your site.


Not All Zero-Click Searches Cost the Same

The most common mistake in talking about zero-click is treating it as one block. A zero-click search where someone checks a bank’s opening hours didn’t cost you a customer. One where a buyer compares three providers in your category and never even names you did.

Sort your demand by the value of the click: how much business you lose if the answer is consumed without visiting you.

Four types of demand and the role of the brand in each (illustrative examples)
Type of demand Illustrative example (B2B services) What you lose without a click Role to pursue
Simple answer "What is a CRM," "what is the filing deadline for X" Little direct value: the reader wants a fact, not a vendor Source: be the cited reference; the click is not the goal
Exploration "How to improve customer acquisition at a logistics company" Early influence: who shapes the buying criteria Brand: be remembered and searched for by name later
Comparison and evaluation "Best agency for…," "X vs. Y," "how much does… cost" High: this is the stage where the buyer picks whom to contact Destination: content with the depth a summary can't replace
Brand and navigational Searches with your company or service name Very high: they already express interest in you Protect: make sure the listing, results and links reflect your offer

Editorial classification by Maccam Network with illustrative examples, not market data. The boundaries between types are blurry and depend on your industry.

This classification prevents two symmetrical mistakes. The first is fighting for clicks where it isn’t worth it: investing to win back traffic from simple-answer queries that, even if recovered, doesn’t turn into opportunities. The second is resigning yourself where it does matter: assuming the decline is inevitable on evaluation queries, when that’s where in-depth content is still the differentiating asset.

One data point supports prioritizing comparison queries. In Seer Interactive’s sample (April 2026; 53 brands), 95.4% of informational comparison queries of the “X vs. Y” type showed an AI Overview, versus under 5% of transactional queries. It is one specific sample, not a universal rate, but it suggests that on comparison queries the summary competes with you more often.

A first practical exercise, using Search Console data: export your highest-impression queries, sort them into the four types and compare CTR and average position trends for each. If CTR falls on simple-answer queries while position holds, that’s the expected pattern and not necessarily a problem. If it falls on evaluation queries, you have a real priority.


Three Roles: The Source, the Brand, the Destination

Being the Source

For simple-answer demand, useful visibility means the answer is built with your information. Google says AI Overviews and AI Mode show supporting links, and that to be one of them a page must be indexed and eligible to be shown with a snippet. The operating steps are in SEO in the AI Era: What to Do for Google and ChatGPT.

A methodological caution applies here. Seer found that brands cited in AI Overviews had a higher organic CTR, but it warns that higher-authority brands may also be the ones cited most. Working on citability makes sense on its own terms; it is not a guarantee of more clicks.

Being the Brand

In exploration, the click doesn’t arrive, but the influence can remain. This is the territory of brand demand: your company’s name becoming associated with a problem, so that people search for you directly when they’re ready. SparkToro makes the point in its 2026 study: even as traffic falls, the influence your content has on AI answers and zero-click features remains, and there are still reasons to publish the support article, the video, or the reply in the communities where opinion forms.

What builds that demand is brand strategy, not an SEO trick: clear positioning, consistent messaging and presence where your customer gets informed. We develop this in The Visibility System: SEO, Content, and Brand and in B2B Positioning for Mid-Sized Companies.

One concrete consequence: look after the consistency of your external listings. Google recommends keeping Business Profile and, if you sell products, Merchant Center information up to date for your appearances in AI features. For businesses with a local presence, local business marketing goes deeper.

Being the Destination

In comparison and evaluation, you win if the buyer needs to leave the summary to decide. That demands content a summary can’t replace without loss: original data with methodology, decision criteria with their exceptions, calculators and templates, documented cases with permission, and honest comparisons that admit when you are not the best option. A practical test: can a good three-line summary replace this page? If so, that content belongs to the first role (source), not the third. The structure that supports it is covered in content architecture for visibility.


A Six-Move Plan

Editorial framework · Maccam Network

  1. Map your demand by the value of the click

    Export your highest-impression queries from Search Console and sort them into simple answer, exploration, evaluation and brand. For each group, note CTR, position and the last 12 months' trend.

  2. Protect brand demand

    Use Search Console's branded queries filter (with Google's caveat: classification is AI-assisted and can misfire). Check that your brand searches return an accurate listing and that your service pages match the intent.

  3. Decide where you want to be source, brand or destination

    Assign a role to each priority topic. A page can't be both the answer that gets summarized and the destination that doesn't: decide what each piece is working for.

  4. Rebuild destinations with something that can't be summarized

    On evaluation topics, prioritize original data, criteria with exceptions, tools and verifiable cases. Retire or merge pages that only repeat what any summary already says.

  5. Make citable what should be the source

    On simple-answer topics, answer precisely and with a date, cite primary sources and maintain technical eligibility. The goal is for the right information to carry your name, not the click.

  6. Measure in units that don't depend only on the click

    Impressions and position as exposure, branded queries and direct traffic as demand, self-reported source and opportunity quality as outcomes. Disclose each source's limits.

Order matters: first understand where the value is (01 to 03), then build (04 and 05), and only then measure (06) with units consistent with the role you gave each piece of content.


How to Measure When the Click Is Not the Unit

There are four families of signals, none conclusive on its own:

  1. Exposure. Impressions and position in Search Console. Google reports that clicks on external links in AI Overviews and AI Mode count as clicks, and that since 2026 there is a dedicated report of impressions in generative AI features. An impression means the link was shown, not that it was seen or remembered.
  2. Brand demand. The trend in branded queries and direct traffic. If a visibility strategy works, demand for your name is one of the first things that should move, even if it takes a while.
  3. Self-reported attribution. A source question on forms and in the first sales conversation, fed into the CRM, captures influences analytics can’t see.
  4. Outcome quality. Qualified inquiry rate, sales cycle and opportunity value, not just session volume.

The detail on what each tool (Search Console, GA4, Bing Webmaster Tools) does and doesn’t measure, and on how to sample AI answers without falling for fake rankings, is in how to measure brand visibility in AI answers. To carry these signals into a leadership conversation, how to measure digital marketing ROI helps.


Common Mistakes

  • Importing a global figure into your plan. “68% of searches are zero-click” describes a US panel, not your industry or your market.
  • Treating all lost traffic as recoverable. Some of the click volume lost on simple-answer queries won’t come back, and may never have been worth what was assumed.
  • Abandoning in-depth content out of fear of the summary. It is exactly what differentiates the destination. What should be abandoned is commodity content.
  • Optimizing to be cited without watching what is said about you. Appearing with incorrect information has a cost.
  • Measuring only clicks and concluding it isn’t working. With the wrong units, any visibility strategy looks like a failure.
  • Confusing diagnosis with strategy. Understanding why a page fell is one job; deciding where to invest is another. This article is the second, and it builds on the first.

Next Step

If your team sees CTR falling and doesn’t know which queries matter, which content deserves rebuilding as a destination and which should aim only to be the source, the starting point is a demand map by value of the click. It’s the kind of work we do within our strategic SEO service, supported by Maccam Network’s SEO methodology. If you’d like to discuss it, get in touch.


Sources

Sources verified as of October 9, 2026.

Preguntas frecuentes

It depends on the study and its method. SparkToro, using Datos data, estimated in 2024 that 58.5% of US searches and 59.7% of EU searches did not end in a click to the open web (September 2022 to May 2024). In 2026, SparkToro using Similarweb panel data put the US figure at 68.01% for January through April 2026. The authors themselves caution that comparing different sources is mixing apples and oranges, and that the 2026 figure excludes the Google mobile app, where it is likely higher.

The studies agree on direction, not on size. Pew Research Center (July 2025), using the real browsing of 900 US adults, found clicks on traditional results in 8% of visits with an AI summary versus 15% without one, and clicks on links inside the summary in 1%. Ahrefs (February 2026) estimated that for position 1, the presence of an AI Overview is associated with a 58% lower CTR. Seer Interactive (April 2026) observed a CTR decline through December 2025 and a partial recovery by February 2026. These are correlations on specific samples, not laws.

No. It means the click stops being the only unit of value. A brand can gain influence by being the source of an answer, by being remembered and searched for by name, or by being the destination a summary can't replace. Google recommends looking at the overall value of visits, using conversion indicators, and not just clicks. What changes is what you measure and which content deserves to exist.

There is an association, but causation is not established. Seer Interactive (April 2026) found that on informational queries in 2025, brands cited in an AI Overview averaged a 2.07% organic CTR versus 0.94% for brands not cited. Its authors say they cannot claim causation, because higher-authority brands may also be the ones cited most, and they note that a single account weighs heavily on the not-cited figure (about 1.61% without it). It is a reason to work on citability, not a guarantee of results.

There is no direct conversion. Combine signals: the trend in branded queries (Search Console offers a branded and non-branded query filter), direct visits, referral sessions from AI assistants, and a self-reported source field in forms and the CRM. None of them alone proves that an impression produced a sale; together they form a reasonable indication. Always disclose each source's limits.

Sort your queries by the value of the click. Simple-answer queries (definitions, facts) tend to have little commercial value per click, so treat them as an opportunity to be the source or to reinforce the brand. Comparison and evaluation queries, where the buyer needs depth, are the ones that justify content that works as a destination. Branded queries are the most worth protecting, because they already express interest in you.

Newsletter

New ideas, analysis and research — directly to your inbox.

Subscribe to receive new Insights publications and other selected content from Maccam Network. No spam. Unsubscribe at any time.

Shall we talk about your business?

Let's talk about what your business needs.

A 30-minute conversation is enough to understand the context, identify the problem and see if we are the right team to help you.