Google Zero-Click Searches Hit 68 Percent as AI Mode Moves Search From Clicks to Citations
SparkToro's new zero-click data shows why AI Mode makes citation presence more important than traffic.
Google search is becoming a citation economy faster than a traffic economy. SparkToro's June 8 analysis of Similarweb clickstream data found that 68.01% of U.S. Google searches in January-April 2026 ended without a click, while Google says AI Mode has already passed one billion monthly users.
Google zero-click search is now the baseline, not the exception
The fresh signal is not that zero-click search exists. The signal is that it is accelerating while Google is moving AI Mode closer to the default search surface. SparkToro's 2026 study reports that fewer than one-third of Google searches now send a click, using Similarweb desktop and mobile panel data for U.S. searches from January through April 2026.
The directional change matters more than the exact decimal. SparkToro put U.S. zero-click searches at 60.45% in 2024 and 68.01% in early 2026, a 7.56-point increase in two years. It also notes a methodological caveat: 2024 and 2026 use different panels, so the clean read is acceleration, not lab-grade continuity. The business read is still blunt: search visibility is no longer synonymous with site visits.
Google is not hiding the product direction. In its May 19 I/O Search update, Google said AI Mode had surpassed one billion monthly users and that queries were more than doubling every quarter. It also announced an AI-native Search box, follow-up questions from AI Overviews into AI Mode, information agents, and agentic booking.
That is a product roadmap for keeping the user inside the answer surface longer.
AI summaries cut clicks even when sources are visible
Citation is not the same as traffic. Pew Research Center's 2025 analysis found that users who saw a Google AI summary clicked a traditional search result in 8% of visits, compared with 15% of visits when no AI summary appeared. Users clicked a link inside the AI summary in only 1% of visits.
That does not make citations irrelevant. It makes them a different asset. A citation inside an AI answer can shape memory, trust, and brand comparison even when it does not produce an immediate referral session. The old bargain was ranking for traffic. The new bargain is being selected, named, and trusted inside the answer.
This is why the common "AI search kills SEO" framing is lazy. SEO still feeds retrieval. The problem is that retrieval is no longer the final surface. The machine retrieves a broad field, compresses it, and shows a narrow answer with a smaller citation set. A page can be eligible without being credited.
The measurement surface is moving from rankings to citation share
The useful KPI shift is from "did the page rank?" to "did the machine select the source?" Traditional search measured position, impressions, clicks, CTR, and conversions. AI search adds a harder question: when an answer engine compresses the source field, which entities and URLs survive into the visible answer?
The Machine Relations research on AI search click collapse and citation concentration frames this as compression: more answers happen inside AI interfaces, the visible answer layer is narrower than the open web, and systems may consume more sources than they publicly credit. That distinction is now operational. Marketers need to know not only whether Google crawled a page, but whether ChatGPT, Perplexity, Gemini, Claude, and Google AI Mode use the brand as answer material.
AuthorityTech's AI Publication Intelligence Index is one example of that measurement shift. It ranks publications by answer-engine signal across engines, queries, average position, days cited, and current cites. The point is not that every brand needs the same index. The point is that the unit of visibility is moving from a blue-link visit to a cited source event.
| Old search metric | AI search replacement | Why it matters |
|---|---|---|
| Organic rank | Citation presence | The answer may cite only a few sources |
| Click-through rate | Share of citation | Brand credit can happen without a visit |
| Page traffic | Answer inclusion | The buyer may decide before clicking |
| Keyword volume | Query cluster coverage | AI Mode handles longer, conversational prompts |
Machine Relations is the right frame for zero-click search
Zero-click search turns AI visibility into an entity-selection problem. The Machine Relations frame is useful here because it separates the layers: earned authority, entity clarity, citation architecture, distribution across answer surfaces, and measurement. GEO and AEO are not wrong. They are just too narrow when clicks stop being the clean success signal.
That category history matters because terminology shapes what teams measure. Machine Relations was coined by Jaxon Parrott in 2024 to name the shift from human-mediated discovery to machine-mediated discovery. In a zero-click environment, the question is not simply "how do we get traffic?" It is "why would the machine trust this brand enough to use it in an answer?"
The operational answer is evidence, not formatting tricks. Brands need original data, consistent entity records, third-party corroboration, answer-first pages, and earned authority from publications that engines already cite. Otherwise, the answer engine has no reason to select them when it compresses ten possible sources into three visible citations.
What teams should change this quarter
The immediate move is to instrument citation presence before traffic loss becomes the only visible metric. A team that only watches Search Console will see impressions, clicks, and CTR. It will not see when a model names the brand, omits the brand, cites a competitor, or uses a third-party publication as the proof source.
Three changes matter now:
- Track branded and category prompts across ChatGPT, Perplexity, Gemini, Claude, and Google AI Mode.
- Separate retrieval from citation by logging whether the brand appears, whether a source URL appears, and which third-party domains are credited.
- Repair the pages and earned media assets that already have authority but lack extractable definitions, current data, or clean entity attribution.
This is not a call to abandon SEO. It is a call to stop pretending clicks are the only proof that search visibility worked. In AI search, the first win may be getting named. The second is getting cited. The click, if it comes, is downstream.
FAQ
What did SparkToro's 2026 zero-click study find?
SparkToro reported that 68.01% of U.S. Google searches from January through April 2026 ended without a click, based on Similarweb clickstream data. The same analysis put U.S. zero-click searches at 60.45% in 2024 and described the two-year increase as the fastest acceleration in the past decade.
Why does Google AI Mode matter for zero-click search?
Google says AI Mode has passed one billion monthly users and that AI Mode queries have more than doubled every quarter since launch. If AI Mode becomes a normal continuation path from AI Overviews, more searches will be resolved inside Google's synthesized answer surface before a user reaches an external page.
Is citation visibility more important than website traffic?
It depends on the query, but citation visibility is becoming a separate metric from website traffic. Pew found that users clicked traditional results less often when AI summaries appeared, which means a brand can influence an answer without receiving a visit. That makes citation tracking necessary, not optional.
How should a brand test whether AI search is using it?
Run the same high-intent category prompts across multiple answer engines, record whether the brand appears, capture which sources are cited, and compare those sources against the brand's earned media and owned pages. A free starting point is an AI visibility audit that checks whether machines can find and cite the brand at all.