Reports of the Death of Websites Are Greatly Exaggerated: Understanding Zero-Click Search

Every SEO consultant knows this phone call. “My traffic is down. Did Google penalize me?”

A drop in website traffic deserves investigation. One possibility worth examining is what happens on the search results page before anyone reaches your website — and zero-click research provides context for that investigation, not a diagnosis of your site. This article explains what “zero-click search” is, how common it really is, what the data can and cannot tell you, and what to check before assuming the worst.

What “zero-click search” means

A zero-click search is one that ends without a recorded click. What counts as a click depends on the study.

The idea is simple. The measurement is not — and that’s where most of the scary statistics you see online go wrong. The two most credible studies measure it differently:

Why does this matter to you? Because percentages from these studies measure different things, and articles that blend them produce numbers that sound precise and mean nothing. When you read a zero-click statistic, ask: measured how, over what period, and by whom?

How common is it, really?

Two studies illustrate different aspects of non-click behavior.

68.01% of U.S. Google searches ended without a click between January and April 2026 — a browser-panel estimate from Similarweb data, not a universal rate (SparkToro, June 2026). Pew’s panel, measuring a slightly different thing, found that in about two-thirds of searches, the user did not follow a detected result link as their next action — they kept searching or left Google (Pew, July 2025).

And this is not a new development. SparkToro’s retrospective account of the trend:

One honest caveat: those four points come from three different data panels, reported retrospectively by one research team, so treat the trend as a direction rather than a verified continuous series.

Notice what this timeline means: by SparkToro’s account, zero-click search was already nearly half of all searches in 2019 — years before AI Overviews existed (SparkToro, June 2026). The majority of searches going clickless is not an AI-era shock. It’s a pattern that predates AI Overviews.

Why this is happening (the short version)

The results page is no longer just a list of links. Depending on the query, Google inserts local map results, answer boxes, expandable questions, and — since 2024 — AI-generated summaries into the results page; Google says its AI Overviews appear only where its systems determine they add value beyond classic Search (Google Search Central).

Pew’s data puts numbers on the newest of these. In its sample, Google showed an AI summary on about 18% of searches (from the April 2025 results-page reconstruction), and when one appeared, users followed a traditional result link on 8% of those visits — versus 15% when no AI summary was present. The Pew researchers note this is an observational association, not a causal finding: the study cannot prove the AI summary itself caused the difference (Pew, July 2025; Chapekis, Lieb, Shah & Smith, Pew preprint). We’ll go deep on AI Overviews in the next article in this series.

What to check if clicks fall

(Recommendations, not findings.)

The measurement that matters lives in Google Search Console’s Performance report, which shows impressions, clicks, and click-through rate by query and page.

Stable impressions with falling click-through rate means fewer clicks per recorded impression. That is a reason to investigate — not, by itself, proof that zero-click behavior caused the decline, and not proof that your website has no problems.

A sensible investigation has four steps:

1. Compare click-through rates like with like. In Search Console, look at impressions, clicks, and CTR by query over the last 12 months — similar queries, similar pages, similar devices, similar periods. A falling CTR on one query type while another holds steady tells you more than a site-wide average ever will.
2. Inspect the actual results page. For your important queries, search Google the way a customer would — phone in hand, from your usual location — and see what actually appears: AI summaries, local map packs, review sites, competitors. Record the date, device, and location, and treat what you find as a snapshot — customers’ results can vary.
3. Compare conversions and leads, not just sessions. Pull your business-outcome records — phone calls, form fills, sales — alongside your traffic. If visits are down but leads are steady, that’s important context. If leads are down too, that’s the more urgent signal. One caution: steady sales overall don’t by themselves prove organic search is healthy — check where those leads came from.
4. Segment informational, transactional, and local queries. Quick-answer informational queries, local searches, and purchase-intent searches behave differently on the results page, and a site-wide average conceals exactly those differences. Separate them before drawing conclusions.

When a missing click may matter — and when it may not

(This section is our interpretation. The studies measure large populations and cannot tell you what any individual searcher intended.)

The research can’t distinguish why a particular search ended without a click. The plausible explanations include:

• The searcher got their answer. Sometimes that’s genuinely fine. Google’s own example of a query that may not need a click is “when is the next full moon” (Google, August 2025).
• Someone else occupies the answer. A local pack full of rivals, a review site ranking above you, an AI summary citing a competitor. That may represent reduced visibility for you — though whether it means lost business is something only your own leads and sales data can indicate.
• The searcher ran another query. A searcher may run another query to refine, expand, or redirect their search; that does not by itself establish frustration or a visibility problem.

These are possibilities, not diagnoses. A dated review of your search results and your business’s own outcomes can provide useful context for the queries you care about. Neither one reveals why an individual non-visitor did not click.

Next in this series: why ranking #1 doesn’t mean what it used to.

References

1. SparkToro (June 8, 2026). “In 2026, Less than One Third of Google Searches Still Send a Click.” — 68.01% U.S. zero-click figure, Jan–Apr 2026; retrospective historical trend across Jumpshot, Datos, and Similarweb panels. sparktoro.com
2. Pew Research Center (July 22, 2025). “Google users are less likely to click on links when an AI summary appears in the results.” — 900 U.S. adults; browsing tracked March 2025; results pages reconstructed April 2025; ~two-thirds of searches with no result-link click as next action; 8% vs. 15% click rates; ~18% AI Overview prevalence. pewresearch.org
3. Chapekis, A., Lieb, A., Shah, S., & Smith, A. (Pew Research Center; arXiv preprint 2608.04831). “Investigating Click Behaviors On Google Search Result Pages That Produce an AI Overview.” — methodology paper for the Pew analysis; observational design, explicitly not causal. arxiv.org
4. Google — Liz Reid, VP & Head of Google Search (August 6, 2025). “AI in Search is driving more queries and higher quality clicks.” — Google’s official position: stable total organic click volume, higher click quality, “next full moon” example. blog.google
5. Google Search Central. “AI Features and Your Website.” — AI Overviews appear where Google’s systems determine they add value beyond classic Search; no additional technical requirements beyond standard Search eligibility. developers.google.com