Zero-Click Search Hit 68% in 2026. Industrial Websites Must Win the Answer, Not the Click.
The short answer: in early 2026, 68% of US Google searches ended without a click — and in a July 2025 Pew study, users who saw an AI summary clicked a traditional result 8% of the time instead of 15%. For an industrial company, the website's job is changing: from winning the visit to winning the answer.
In short
- SparkToro's June 2026 study (Similarweb US panel, January–April 2026): 68.01% of Google searches ended without a click, up from 60.45% in 2024. The panels differ across years, so read it as direction, not a precise index.
- Pew Research Center, July 2025 (900 US adults, 68,879 Google searches, March 2025): with an AI summary on the page, users clicked a traditional result in 8% of visits vs 15% without; just 1% clicked a link inside the summary; 26% ended their session vs 16%.
- AI Overviews now appear on more than 20% of searches, and searches showing one have roughly 60% lower click-through than comparable searches without one (Ahrefs, 2026, via SparkToro).
- The industrial version: 48% of industrial buyers already use AI to research suppliers but trust it only 2.7 out of 5. The same study found 83% use vendor websites at some point in the process — which is why your site is the verification layer.
- Five moves: write pages that win the answer, get named for use cases, defend the answer's accuracy, treat the rare click as gold, and measure influence instead of sessions.
Your traffic report is telling you a story about a world that is ending. For twenty years the deal was simple: rank well, earn the click, convert the visitor. In the first four months of 2026, 68.01% of Google searches ended without a click at all — no visit for anyone to convert. The click isn't the prize anymore. The answer is.
The numbers
SparkToro has tracked this for a decade. In 2019, 49% of Google searches ended without a click (Jumpshot panel). In 2024, 60.45% (Datos panel). In the first four months of 2026, 68.01% (Similarweb's US clickstream panel, January–April 2026). Rand Fishkin's own warning travels with the trendline: these are different panels, different users, different devices — “a bit of apples and oranges.” The levels are the fact; the precise slope is not. But the direction has never reversed, and AI Overviews are the most visible current contributor: they now appear on more than 20% of searches; searches showing one have roughly 60% lower click-through than comparable searches without one (Ahrefs, 2026, observational); and a 2026 preregistered field experiment (Wang et al., N=1,100) found that removing AI Overviews and AI Mode increased click-through to publishers.
The Pew half
The most careful measurement of the AI piece comes from Pew Research Center. In July 2025, Pew published a study of 900 US adults' actual browsing — 68,879 Google searches from March 2025, collected via KnowledgePanel Digital. When a search page showed an AI summary, users clicked a traditional result in 8% of visits, against 15% when no summary appeared. Links inside the summary itself were clicked in just 1% of visits. And 26% of visits to pages with a summary ended the browsing session entirely, versus 16% without one.
Two honest caveats, both built into the study's shape: it's observational — it shows the summary and the missing clicks arriving together, not proof the summary caused each one — and it's one panel, one month, Google only. Google disputes the framing; its head of Search, Liz Reid, said in August 2025 that Google continues to send “billions of clicks to websites every day.” Both things can be true: the pie of attention is being re-sliced, and Google is still the bakery.
Why this hits industrial sites differently
Industrial was never playing the traffic game. Your buyers don't impulse-click — but they do ask AI. A 2026 B2B Manufacturing Buying Journey study found 48% of industrial buyers used AI to research suppliers in the prior 90 days, while rating their trust in AI answers at only about 2.7 out of 5. The same study found 83% use vendor websites at some point in the process, and 37% start there.
Read those two facts together and the new funnel appears: the AI answer is the first impression, and your website is the verification layer. The buyer asks the AI who can solve the problem; the AI names names from what it can quote — and in industrial and manufacturing, 77.4% of AI citations point to company websites (OtterlyAI's 2026 index). Then the skeptical buyer — trust 2.7 out of 5 — clicks through to check. If your pages give the AI nothing quotable, you're absent from the first impression. If they give it something wrong, the verification visit goes to a competitor.
Move 1: Win the answer, not the click
The unit of competition changed from the ranking to the quoted paragraph. When 68% of US searches end without a click and summary links get clicked 1% of the time, “position one” is a vanity metric — being the sentence the AI reads aloud is the asset. That means answer-first pages: the direct answer in the first hundred words, specific numbers instead of adjectives, plain sentences that survive being lifted out of context. It also means the technical layer: FAQ and product schema, an llms.txt pointing AI systems at your canonical pages, and a robots.txt that doesn't block AI crawlers — quotable content still has to be machine-readable. Most industrial pages were written for a machine that ranks; they need rewriting for one that summarizes. The playbook for quotable pages.
Move 2: Get named for the use case
Being quotable isn't the same as being recommended. Semrush's 2026 survey of 600+ B2B professionals found that 53% notice vendors in AI answers for use-case fit — and only 7% for brand recognition. The AI names the company that fits the described problem, not the one with the biggest logo. That's a page-structure problem: use-case pages (“conveyor drive retrofit for food plants”) beat capability pages (“Power Transmission Solutions”) far more often. The playbook for getting named.
Move 3: Defend the answer
Winning the answer is the offense; accuracy is the defense. A 2026 WashU audit of 55,000+ Google searches found roughly 89% of AI Overview claims supported by their cited sources — which leaves up to about 1 in 9 not supported. Run the accuracy check monthly: ask the AI what it says about your company, compare it against your own pages, and fix the source page wherever the AI is quoting stale or third-party text. The correction playbook.
Move 4: Treat the rare click as gold
The visitors who still click — the shrinking share that still clicks — are the most pre-briefed, most problem-focused visitors your site will meet. They arrive mid-problem, pre-briefed by an AI, and skeptical. Don't greet them with a homepage carousel; greet them with an answer. That's the lobby-ambassador job: a first-time visitor types a problem, and somebody knowledgeable responds in seconds — where site search helps and where it doesn't. See the pre-sales ambassador on your own content.
Move 5: Measure influence, not sessions
SparkToro's own advice for the zero-click era: replace traffic as a KPI and build a correlation dashboard instead. For an industrial site that means watching search impressions, tracking AI referral traffic separately from organic, monitoring branded search, and asking sales what prospects mention on calls. Google's Search Console Generative AI reports — live since June 3, 2026 — give you the baseline for how often your pages surface inside AI answers. The dashboard that only counts sessions is now measuring the shrinking slice.
Start with the free grader
Run your site through the free Website Grader — it scores how easily a first-time visitor can find answers on your site, in about a minute. For the longer read, the 14-page 2026 Findability Benchmark is free in the Growth Kit.
Related: 77% of industrial AI citations go to company websites, why AI doesn't recommend your company to B2B buyers, and the industrial buyers' AI trust gap.
What this post doesn't prove
Worth being straight:
- The decade trendline mixes panels — Jumpshot in 2019, Datos in 2024, Similarweb in 2026. Rand Fishkin's own caveat: “a bit of apples and oranges.” Direction, not a precise index.
- Pew's study is observational, from March 2025, on Google, with US adults. It doesn't prove the summary caused every missing click, and Google disputes the framing.
- None of these studies measured industrial buyers specifically. The 68% is Google-wide and US-only; what your buyers do is your own Search Console's story.
- Zero-click isn't all lost demand. Some searches are satisfied on the page, some get asked again differently. The threat is the specific slice where your company should have been the answer and wasn't.
Method
Zero-click figures — 68.01% of US Google searches with no click, January–April 2026; 60.45% in 2024; ~49% in 2019 and ~45% in 2016 (Jumpshot panel era); AI Overviews on 20%+ of searches; searches showing one have ~60% lower CTR than comparable searches without (Ahrefs, 2026, observational); AI Mode 0.34% of searches — from SparkToro, June 2026 (Similarweb US clickstream panel; SparkToro sells audience research; panels differ across years — see the post's own caveat). Pew figures — 900 US adults, 68,879 Google searches, March 2025; AI summary on 18% of searches; traditional-result clicks 8% of visits with a summary vs 15% without; 1% clicked a link inside the summary; 26% ended the session vs 16% — from Pew Research Center, July 2025 (observational; KnowledgePanel Digital). Causal direction — removing AI Overviews and AI Mode increased publisher click-through in a 2026 preregistered field experiment (Wang et al., N=1,100) — via arXiv:2608.18352 (directional finding from the abstract). Google's response (“billions of clicks,” Liz Reid, August 2025) via NorthJersey.com, October 2026. Industrial-buyer figures — 48% used AI to research suppliers in the prior 90 days; trust ~2.7/5; 83% use vendor sites during the process; 37% start there — from the 2026 B2B Manufacturing Buying Journey & Thought Leadership study, via our trust-gap post. Semrush 2026 survey figures (600+ B2B professionals; 53% use-case fit vs 7% brand recognition) via our recommendation post. WashU audit figures via our correction playbook. All figures attributed inline with years.