What Google's Open Web Click Share Really Shows

Google searches, 232 clicks reach what SparkToro calls the open web, according to new data drawn from Similarweb's clickstream panel. The report, published by SparkToro co founder Rand Fishkin, found 68% of U.S. The same pattern also shows up in GBP Comes to GA4, where the practical question is how the signal becomes visible.

searches ended without any click from January through April. Its 2024 report, based on Datos data, separately reported 360 open web clicks per 1,000 U.S.

What The Numbers Show

Data shows post search behavior going three ways: 39% of searches end with no further action. 29% lead to a new query in Google's search bar.

Of those clicks, 66% lead searchers to pages on the open web, while 27% go to Alphabet owned properties and Google surfaces, including YouTube, Maps, and AI Mode. The remaining 6% goes to paid ads. Compared with 2024 data, the share of searches that produced at least one click fell from 41% to 32%.

What The Numbers Show: the traffic story should be read as a planning signal, not a reason to abandon search. If fewer searches produce open web clicks, owned pages need stronger intent fit, clearer brand demand, and better measurement of influence beyond the last session.

What The Numbers Show: the decision layer should separate what is directly measurable from what is directional. Clean referrals, branded demand, assisted conversions, citations, and sales notes all matter, but they should not be blended into one overconfident number.

Paid Clicks Take A Larger Share

Paid's share of all clicks rose from 1% in the 2024 data to 6% in 2026. Fishkin cautions against reading too much into the jump.

The 2024 panel from Datos had a higher than average share of users running ad blockers, which hid or minimized search ads. The true 2024 paid figure was likely higher, making the increase look steeper than it was.

Paid Clicks Take A Larger Share: open web clicks are now a scarcer outcome, which makes page quality and brand recall more important. The visit has to earn more because fewer searches may create one.

Paid Clicks Take A Larger Share: the report should show confidence levels. A tracked click is different from a brand lift signal, and both are different from a sales note that suggests AI influenced the conversation.

How This Squares With Google's Statements

Google has spent the past year arguing that AI features aren't draining useful traffic from websites. VP of Search Liz Reid has said organic click volume is "relatively stable" and that AI Overviews mostly remove "bounce clicks," visits where users grab a fact and leave.

Google hasn't published data supporting either claim. SparkToro's data tracks clicks per search. Reid's statements concern total click volume, which can hold steady if query growth offsets a falling click rate.

How This Squares With Google's Statements: the more useful metric is not only traffic volume. It is whether search still creates qualified demand, branded follow up, subscriptions, assisted conversions, or stronger consideration.

How This Squares With Google's Statements: a useful dashboard should explain what changed and what action follows. Otherwise it becomes another view that looks impressive but does not improve the next decision.

About The Data

The analysis uses Similarweb's U.S. desktop and mobile panel (January April).

SparkToro weighted results as two thirds mobile, one third desktop. Fishkin says zero click behavior in the app is likely higher than in browsers, but searches in the Google mobile app aren't included The 2016 and 2019 figures came from the now defunct Jumpshot panel, the 2024 figures from Datos, and the 2026 figures from Similarweb. Fishkin calls the cross year chart "a bit of apples and oranges." For transparency, SparkToro sells audience research software, and Fishkin is publishing a book on zero click marketing with co author Amanda Natividad.

About The Data: this should also change forecasting. Models built on older click curves can overstate reachable traffic if zero click behavior and AI answers keep absorbing more of the journey. This connects with structured data when the same signal needs a clearer operating decision.

Why this changes the operating question

This report is a sign that traffic forecasts built on older click rates need revisiting, and the 232 per-1,000 figure gives you a concrete number for those conversations. Fishkin writes that SEO matters as much as ever, "it just won't earn you traffic the way it once did." He points instead to categories that still benefit from SEO, such as branded searches, local businesses, and high intent transactional queries, which he credits to Cyrus Shepard's analysis.

This report is a sign that traffic forecasts built on older click rates need revisiting, and the 232 per-1,000 figure gives you a concrete number for those conversations. Fishkin writes that SEO matters as much as ever, "it just won't earn you traffic the way it once did." He points instead to... The useful way to handle this is to connect the observation to a clear signal, then decide whether it changes content quality, crawlability, measurement, brand evidence, or the user's decision path.

Why this changes the operating question: for local SEO, this moves profile data closer to daily decision making. Reviews, services, categories, photos, and website content need to support the same answer because Gemini can turn weak profile hygiene into a visible customer experience problem. A useful companion note is What Gemini Business Profile Tools Mean, because it looks at a nearby part of the same system.

Why this changes the operating question: the report should show confidence levels. A tracked click is different from a brand lift signal, and both are different from a sales note that suggests AI influenced the conversation.

What to watch next

AI Mode is the variable to watch. It accounted for 0.34% of searches in this dataset, but Google says usage has passed 1 billion monthly users, and queries are more than doubling each quarter.

SparkToro will publish zero click figures for Europe, the UK, and Canada in the days ahead. He plans to repeat the analysis within 6 to 12 months.

What to watch next: open web clicks are now a scarcer outcome, which makes page quality and brand recall more important. The visit has to earn more because fewer searches may create one.

What to watch next: a useful dashboard should explain what changed and what action follows. Otherwise it becomes another view that looks impressive but does not improve the next decision.

What What The Numbers Show changes in practice

What What The Numbers Show changes in practice should be checked against the page and the wider search system, not treated as an isolated note. The goal is to find the weakest missing proof point and improve that before expanding the topic further.

What should be checked before changing the strategy

What should be checked before changing the strategy: The first check is whether the page already answers the question with enough specificity. If the visible content, internal links, and supporting proof are thin, publishing more around the same topic will only spread the weakness across more URLs.

What Google's Open Web Click Share Really Shows should therefore be treated as part of an operating system, not a one off news item. The useful outcome is a clearer page, a cleaner signal, and a better decision about what deserves attention next.

Where the signal needs stronger proof

Where the signal needs stronger proof: The evidence layer matters because search systems need corroboration. A strong page should connect the entity, the topic, the offer, and the user's intent without forcing a crawler or an assistant to infer the missing pieces.