Zero-Click Hits 68%: Inside the 2026 SparkToro Study

Zero-click search reached a new high in 2026: SparkToro and Similarweb report that 68.01% of US Google searches ended without a click in the first four months of the year. That is up from 60.45% in 2024 — a 7.56-point jump in two years, which SparkToro describes as the fastest acceleration since it began tracking the metric.

What makes this study unusually credible for us to write about: its methodology cites Digital Applied's 2026 Google Ads benchmark panel for one of its inputs — the desktop-versus-mobile paid-click ratio used to estimate ad clicks. We are not commentators on someone else's data here; one slice of the calculation runs on numbers we publish. That is also why we are scoping this post tightly to the June 2026 study rather than re-litigating earlier zero-click coverage, which we link throughout as context.

This analysis walks through what the study actually found, the more visceral "276 per 1,000" framing, what is genuinely driving the surge versus what is being misattributed to it, the methodological caveats SparkToro itself flags, the honest counter-case that AI traffic converts better, and the strategic pivot that follows. Every figure below is sourced to the study or its cited references; where a number is vendor-stated, we say so.

The top-line finding is blunt: in the first four months of 2026, 68.01% of US Google searches ended without a click. In 2024 the figure was 60.45%. The 7.56-percentage-point increase over two years is, per SparkToro, the steepest two-year move since it started measuring the metric. The 2026 number is powered by Similarweb's clickstream panel; the 2024 baseline came from Datos, now a Semrush company.

The longer arc, as SparkToro presents it, runs from roughly 45% in 2016 to about 49% in 2019, to 60.45% in 2024, to 68.01% in 2026 — roughly a 23-point rise across a decade. We cover the framing of that longer trend, and the earlier data points, in our prior zero-click statistics deep-dive and, for the moment the metric first crossed 60%, in our 2024 60%-threshold coverage. Those earlier numbers are anchors, not new findings — the new finding is the 68.01% mark and the acceleration behind it.

A percentage is easy to wave away. A ratio is not. SparkToro's most concrete number is this: for every 1,000 US Google searches, only 276 clicks now reach the open web. In the 2024 Datos study, that figure was 374. So in two years, the open web lost roughly a quarter of its clicks per 1,000 searches — from 374 to 276, a 26% reduction.

Where do the other clicks go? Barry Schwartz at SE Roundtable broke down the destination split for the searches that do produce a click: roughly 66% land on open-web destinations, about 27% stay inside Google properties (AI Mode, YouTube, Maps, Images), and around 6% go to paid ads. The dominant outcome, though, is no click at all — about 680 of every 1,000 searches end in the results page itself.

Two mechanisms do most of the work. The first is AI Overviews. According to Ahrefs trigger data cited in the study, AI Overviews now appear on more than 20% of all Google searches, and — per Ahrefs' CTR analysis, again cited by SparkToro — when an AI Overview is present, click-through rate drops by roughly 60%. We covered that specific Ahrefs dataset in our earlier AI Overviews data update; the SparkToro study leans on the same source for its attribution.

The second mechanism is behavioral, and it is the one most coverage misses. SparkToro found that searches leading to another Google search rose 7.2 percentage points from 2024 to 2026 — Google is getting better at persuading people to refine and re-query inside the results rather than click out. The largest single behavioral shift in the study, the "Clicks 1X+" metric (all non-search clicks combined), fell 9.51 points, a 22.9% relative decline. The walled garden is being built out of interface design, not just AI answers.

There is a piece of context that often gets lost in the alarm: Google's ad revenue and stock performance grew over this same window even as organic clicks fell. Fishkin points to Google's Q1 2026 earnings as evidence that the walled-garden strategy is, for Google, economically rational — it can retain searchers, serve more answers in-SERP, and still grow revenue. With the US antitrust case resolved without significant business impact, there is little external pressure pushing the interface back toward sending clicks out. Read forward, that means the 276-per-1,000 figure is more likely to keep falling than to rebound, which is the trajectory our Q3 2026 zero-click acceleration projection models out.

A lot of the 2026 anxiety has been aimed at the wrong target. The study found that only 0.34% of US Google searches were routed to AI Mode in January–April 2026. That is a rounding error against the 68% zero-click rate. AI Mode is not what is eating the open web's clicks today — AI Overviews are. Conflating the two leads teams to prepare for the wrong threat.

That said, AI Mode is plausibly the future pressure rather than the present one. Google reported at I/O 2026 that AI Mode had surpassed 1 billion monthly users and that query volume was more than doubling every quarter — both figures vendor-stated by Google, so treat them as directional claims rather than independently verified facts. The honest read: AI Overviews are the live mechanism behind the zero-click acceleration; AI Mode is the one to watch on a 12- to 18-month horizon, not the one to panic over this quarter.

Most coverage graphs the long-run zero-click trend as a single smooth line. It is not one line — it is three different measurement panels stitched together, and SparkToro flags this itself. The 2016 and 2019 figures came from Jumpshot, a panel that has since shut down. The 2024 figure came from Datos (now a Semrush company). The 2026 figure comes from Similarweb. Different panels, different methodologies, different sampling.

This does not invalidate the trend — the direction is consistent across every independent panel that has measured it, and Search Engine Land independently reported the same 68.01% and 276-per-1,000 figures. But it is worth stating plainly: the year-over-year deltas inside one panel (60.45% → 68.01%, both anchored to comparable recent methodology) are more reliable than the decade-long line that spans three vendors. We treat the recent two-year move as the robust finding and the longer arc as directional.

The honest version of this story includes a counter-argument. Adobe's Q2 2026 AI Traffic Report — vendor-stated, since Adobe sells AI-optimization products alongside the research — found that AI tool referral traffic to US retail grew 393% year-over-year in Q1 2026, and that in March 2026 AI-referred traffic converted about 42% better than non-AI traffic. That is a sign flip from a year earlier, when AI visitors converted at roughly half the rate of other channels.

Treat those figures with the appropriate skepticism — the 393% growth is retail-sector-specific, not all-website, and the conversion claims come from a company with a commercial interest in the narrative. But the directional point holds and is worth taking seriously: the traffic that does arrive via AI referrals appears to be higher-intent. If volume is falling while quality is rising, the rational response is to stop optimizing for raw sessions and start optimizing for visibility, citations, and qualified arrivals — which is exactly the KPI shift the loss data already implies.

Zero-click is not uniform across query types. SparkToro's companion analysis on what still works points to branded, local, and transactional queries as the categories that continue to generate clicks even in a 68%-zero-click environment — informational and how-to queries are the ones AI Overviews most readily absorb. Cyrus Shepard's analysis of winning Google sites (via the Zyppy Signal newsletter) found five shared traits among winners: real products or services, proprietary assets, niche focus, recognizable brands, and task completion — present in 84% of winners versus 50% of losers.

The strategic conclusion the data forces is not "abandon SEO." It is "stop measuring SEO primarily by traffic." Fishkin's own framing is that SEO matters as much as ever for brand influence and for the citation signals that feed AI answers — it simply will not deliver the click volume it once did. The practical response has three moves: re-instrument measurement, invest off-site, and treat citations as the new ranking. We lay out the full operating model in our zero-click strategy guide.

First, re-instrument. A traffic dashboard alone will read like a slow-motion failure even when the business is healthy; pair it with correlation dashboards that tie branded-search volume, impressions, and AI citations to revenue. Second, invest where discovery now happens — unowned platforms. Even off Google the link-out economics are hostile: Meta's spring 2025 Widely Viewed Content Report found 97.3% of US Facebook post views went to updates without an external link, which is precisely why platform-native, link-light content out-reaches link-bait. Third, optimize to be the cited answer, which is the heart of generative engine optimization — the discipline our agentic SEO service is built around.

One realistic caveat on the AI-referral upside: ChatGPT currently drives only about 0.2% of referral traffic and Perplexity under 2%, so the AI channels are not yet replacing the lost Google clicks at volume. The pivot is about positioning for where attention is going, not harvesting traffic that already exists. For teams that want this instrumented end-to-end — measurement, off-site presence, and citation optimization run as one program — that is the kind of build our content engine is designed to deliver.