ToolsThatRank / measured data
First party census
Zero clicks is usually an impressions problem
We counted every URL across 34 Search Console properties that Google showed to somebody over 90 days. 5,130 appeared. Only 467 earned a click. But the reason is not that the other pages are bad. It is that almost none of them were shown enough times to find out.
Verified 2026-08-15 against the Search Console searchAnalytics endpoint. Window 2026-05-15 to 2026-08-12. Fleet click rate 0.2942%.
Check your own page
How many impressions before zero means something
Type the impressions your page has. You'll get the measured share of URLs at that volume that earned a click, and the odds of getting zero purely by chance at our click rate.
The odds column is a Poisson calculation at our own fleet click rate, not a prediction for your site. It answers one question. Could this page have got zero by luck.
The measurement
Click coverage rises with impressions, steeply
Group every URL by how many times Google showed it, then count how many in each group earned at least one click. The pattern is not subtle. Below ten impressions almost nothing converts. Above five hundred, most things do.
| Impressions | URLs | Earned a click | Coverage |
|---|---|---|---|
| 1 | 705 | 5 | 0.71% |
| 2 | 486 | 5 | 1.03% |
| 3 to 9 | 1,584 | 13 | 0.82% |
| 10 to 49 | 1,339 | 46 | 3.44% |
| 50 to 99 | 262 | 41 | 15.65% |
| 100 to 499 | 424 | 126 | 29.72% |
| 500 or more | 330 | 231 | 70% |
Highlighted rows cleared 25% coverage.
The correction
Most zero click pages did not fail
4,663 URLs took impressions and never converted one. The tempting read is that all of them need better titles. Run the arithmetic and that read falls apart.
The median zero click URL was shown 7 times in 90 days. At our fleet click rate of 0.2942%, its expected click count was about 0.02. Zero was always the likely result. Model each one as a Poisson draw at that rate and 4,456 of the 4,663 zero click URLs, 95.6% of them, were more likely than not to record zero clicks by chance alone.
Only 39 URLs, 0.76% of the whole census, earned meaningfully fewer clicks than their impressions predicted. Those are the pages where a title rewrite is worth your afternoon. The other 4,456 have an impressions problem, and no amount of copy editing fixes that.
So the headline number, 9.1% of URLs earning a click, measures impression scarcity across this portfolio. It is not a quality score, and it should not be read as one.
Not the same statistic
This is not the zero click searches number
There is a well known figure that around two thirds of Google searches end without a click. Search Engine Land put it at 68% in early 2026, and earlier work put it near 60% in 2024. That counts SEARCHES where the user clicked nothing at all.
This page counts something else. It counts URLs, and asks how many of them earned a click from somebody over 90 days. A single search that ends without a click affects every result on that page. A URL that earns no clicks may simply have been shown four times. Do not stack the two numbers or read one as evidence for the other.
Every property
How the 34 properties differ
Across all 34 properties the share of URLs earning a click runs from 0% to 50%, with a median of 10.53%. Those extremes come from tiny properties where a single page swings the percentage, so the table marks them. Keep only the 23 properties with at least 20 URLs and the range tightens to 0.55% through 31.51%, median 8.62%. Both readings ship because picking one would be picking a story.
| Property | URLs | Earned a click | Share |
|---|---|---|---|
| belikenative.com | 2,233 | 304 | 13.61% |
| zovo.one | 981 | 14 | 1.43% |
| bestchromeextensions.com | 909 | 5 | 0.55% |
| heytensor.com | 104 | 23 | 22.12% |
| claudhq.com | 101 | 17 | 16.83% |
| lakelevelnow.com | 73 | 23 | 31.51% |
| b2berp.com | 65 | 2 | 3.08% |
| lochbot.com | 64 | 4 | 6.25% |
| kickllm.com | 59 | 10 | 16.95% |
| epochpilot.com | 58 | 5 | 8.62% |
| ml0x.com | 39 | 9 | 23.08% |
| abwex.com | 38 | 4 | 10.53% |
| ingredientcalculator.com | 34 | 9 | 26.47% |
| enhio.com | 34 | 4 | 11.76% |
| recipetool.net | 33 | 2 | 6.06% |
| claudflow.com | 32 | 6 | 18.75% |
| gpt0x.com | 31 | 2 | 6.45% |
| claudkit.com | 30 | 2 | 6.67% |
| ml3x.com | 29 | 2 | 6.9% |
| lockml.com | 26 | 2 | 7.69% |
| invokebot.com | 25 | 1 | 4% |
| gen8x.com | 25 | 4 | 16% |
| kappafy.com | 22 | 1 | 4.55% |
| krzen.com (too few pages to read) | 18 | 3 | 16.67% |
| neovistainc.com (too few pages to read) | 14 | 2 | 14.29% |
| dismissticket.com (too few pages to read) | 13 | 0 | 0% |
| viryd.com (too few pages to read) | 8 | 0 | 0% |
| englishlearner.org (too few pages to read) | 8 | 3 | 37.5% |
| grammarhelper.org (too few pages to read) | 7 | 0 | 0% |
| pictureeditor.net (too few pages to read) | 5 | 1 | 20% |
| bugbountyreality.com (too few pages to read) | 4 | 2 | 50% |
| writingassistant.org (too few pages to read) | 3 | 0 | 0% |
| aiwebsitepipeline.com (too few pages to read) | 3 | 0 | 0% |
| paraphrasetool.org (too few pages to read) | 2 | 1 | 50% |
Method
How we measured this
We measured this on 2026-08-15, over 2026-05-15 to 2026-08-12, against the Google Search Console Search Analytics API on 34 verified properties the operator owns. We pulled the page dimension for every property and paged through every result set to completion, so nothing was truncated at the row cap.
URLs, not pages
Search Console reports URL strings. 5,130 came back, and normalising for scheme, www, trailing slashes and fragments collapses them to 5,009. So 121 are alternate spellings of a page counted elsewhere.
What counts as earning a click
One click or more across the whole 90 days. It is the lowest bar available, which is what makes the coverage numbers stark.
The odds calculation
Poisson, with the rate set by our own fleet click rate of 0.2942% applied to each URL's impressions. It is a sanity check on whether zero is surprising, not a model of your site.
What we did not do
No sampling, no estimating, no modelling of the coverage numbers themselves. Those are counts of rows the API returned, checked twice.
Limits
What this census cannot tell you
The denominator only contains URLs Google actually showed. A page published and never surfaced never appears in Search Console at all, and it would convert at zero by definition. So 9.1% is an upper bound on the share of published pages that earn clicks. The true figure for everything an operator publishes is lower, and this data cannot say by how much.
A URL first published near the end of the window had less time to earn a click. Fleet traffic was falling rather than ramping across these 90 days, which argues against recency explaining much, but we cannot rule it out per URL because Search Console does not report a first seen date in this export.
These are 34 small sites in mixed niches run by one operator, and one property supplies most of the impressions. It is a real census of a real portfolio, not a sample of the web.
No ranking, traffic, or revenue outcome is promised.
The operator sells the software described here and owns every site in the dataset.
Search Console is authenticated, so no third party can verify these readings from outside.
Figures are read on the stamped date and change as Search Console reprocesses data.
Related reading. What pages below position 30 actually earn, measured on the same 34 properties.