Impressions but no clicks: what it means (186 sites)

Impressions with no clicks are the times Google shows one of your pages in the results and nobody comes in, and if you open the performance report in Google Search Console you probably have a pretty long list of those pages. The question that shows up right after is whether that is a problem and what needs fixing.

So we went and counted. We reviewed 186 company websites in the US and Latin America, from October 10 to November 6, 2026: at the median, 24% of the pages Google showed 100 times or more did not get a single click. And close to half of those pages show up, on average, inside the top 10.

My read is that the list has to be split by position before you touch anything: outside the top 10, zero clicks is usually plain arithmetic, and the work on titles, intent and format pays off in the half that does sit on the first page.

What are impressions without clicks?

An impression is every time a link to your website appears on a results page a user saw, even if they don't click. A click is when that user comes in.

Impressions without clicks are the gap between the two: your result was on the screen and people let it go by. All of this is measured in the Search Console performance report.

The relationship between both metrics is CTR, calculated by dividing clicks by impressions and multiplying by 100. A page with 2,000 impressions and 20 visits has a 1% CTR.

One with 300 impressions and no visits has a 0% CTR, and if you don't want to do the math by hand, the CTR calculator does it for you.

One important detail: the position Search Console shows is the page's average position, an average across every search where it appeared.

How does Google count an impression?

This explains a good share of the impressions in Google that were never going to bring a click. According to Search Console's official help, these are the report rules that matter most:

• A regular result counts when it appears on the page that loaded, even if the user never scrolls down to it.

• AI Overviews take up a single position, and every link in the summary shares it. If your page is cited there, it adds an impression even if the user reads the answer and never clicks.

• In AI Mode, every follow-up question counts as a new query.

• The featured snippet counts in position 1.

• Sitelinks (the links that show up under a result) add impressions for other URLs on your site, and the image pack can also add to web search.

In other words, an impression means your link was in the search results, or could have been.

How many pages get impressions without clicks?

At the median per site, 24% of the pages with 100 or more impressions in 28 days had zero clicks (p25 14%, p75 37%; 182 sites with at least 10 of those pages).

That does not equal a quarter of your visibility: the pages with no visits took only 9% of each site's impressions (median; p25 3%, p75 19%).

The first thing that jumps out is the staircase: 8% of pages with no clicks in spots 1 to 3, 16% from 4 to 10, 30% from 11 to 20 and 47% from 21 down. It is a correlation, but a fairly clear one.

The second is that the top 10 doesn't get a pass. Of each site's total pages with no clicks, 47% show up on average between spots 1 and 10 (median; p25 22%, p75 75%; 175 sites), 23% between 11 and 20 and 20% further down.

And among top 10 pages with 100 or more impressions, 18% did not get a single visit (median; 166 sites).

Why does a page get impressions but no clicks?

The causes every guide repeats are real, but they weigh differently depending on where the page shows up.

Outside the top 10: it's the position

In our CTR by position study across 153 sites, non-branded searches had a median CTR of 0.25% in spot 10 and in the 11 to 20 range.

At that rate, a page needs about 400 impressions to expect a single click. One in spot 15 with 100 impressions and zero visits has a low CTR and is doing exactly what you would expect (nothing broken in the title, people just don't scroll that far).

The same thing happens with a new website: Google tests your pages on lots of searches, low on the list, and impressions arrive weeks before the traffic does.

In that range I wouldn't start with the title. I would start by moving the page up, with better content for that intent and internal links from pages that already get visits.

In the top 10: title, intent, or the answer is already on Google

A page on the first page of results with 100 or more impressions and zero clicks does deserve a review. There are three suspects, and the data can't tell which one weighs more.

The first is the title and the meta description, your shop window on Google. If the title doesn't say what the user is looking for, or the description is generic, they pick the neighbor (in SEO title tag there is how to write it and measure the change).

The second is search intent. If a user is looking for a definition and your page is a sales landing page, or the other way around, it can rank high and still convince nobody.

The third is that the answer is already on Google's page: a featured snippet, an AI summary, a map or a carousel. In my analysis of 2,998 popular searches across five countries for Ahrefs in Spanish, 81% had an AI Overview (79% in Chile).

On top of that, Ahrefs associated the summary with a 58% lower CTR for the first organic result, in a study of 300,000 keywords.

A homegrown example: the Spanish version of my article on Google's AI Mode had lots of impressions (125,441 in the same period) and few clicks, 36, at an average spot of 6.3.

I suspect a good share of that demand wants to use AI Mode and lands on an article that explains it, so that's where I would check intent first.

URLs with #: probably not a problem

If you see URLs with # in your list (a link to a section inside the page), don't panic. Across the 25 sites with 10 or more of these URLs, 96% had impressions and no visits (p25 91%, p75 100%), at an average spot of 6.

Most likely they reflect how Search Console records links to sections: the click can end up on the main URL (Google assigns data to each page's canonical URL). That's why we took them out of the main calculation.

Why do impressions go up and clicks don't?

When you see a spike in impressions with no visits to match, the usual story is that the page started showing up for new searches, almost always low on the list, or it got into an AI summary or an image pack.

The other frequent explanation is the season. Demand goes up and down, so it's worth comparing with the same month last year and checking the query behind the spike before changing anything.

What kind of pages end up with no clicks?

The blog contributes more: 27% of its pages with impressions got no visits (p25 15%, p75 44%; 154 sites) versus 21% for the rest of the site (p25 11%, p75 35%; 171 sites), and it does worse in 87 of the 145 sites that have both.

But if you look at the top 10 only, the difference almost disappears: 15% on the blog against 14% elsewhere. The gap comes mostly from the blog having lots of pages low on the list.

By country, the US had 24% (104 sites) and Latin America 18% (39 sites). By size, sites with fewer than 300 pages with impressions reached 29% and those with 3,000 or more, 18%. Part of these differences may come from the industry mix and from size, which this data can't separate.

How do you find your pages with impressions and no clicks in Search Console?

With the same measurement as the study:

1. In Google Search Console, open the web search performance report, turn on clicks, impressions, CTR and position, and pick the last 28 days.

2. Go to the Pages tab and export the table. Search Console cuts off at 1,000 rows, so on a large site the Search Console API or the Search Analytics for Sheets extension is the better choice, since they pull many more rows.

3. Filter pages with 100 or more impressions and 0 clicks. With fewer impressions, the number depends on a couple of visits.

4. Remove # URLs and those with parameters, which tend to be noise.

5. In Google Sheets, build a pivot table with the average position rounded into ranges (1 to 3, 4 to 10, 11 to 20 and 21 or more) and count how many pages fall into each one.

6. For the top 10 ones, filter by page, check which queries each one shows up for and search those queries on Google to see what sits above you.

I ran the exercise on my own site for the same period: 23 of the 53 pages with 100 or more impressions got no clicks, 43%, above the study's p75 (yes, I landed on the ugly side of the table, though it fits a small site).

Only 4 of those 23 were in the top 10, so my pending work is mostly about moving up.

What should you do with each page depending on its position?

With the list split by range, the decision gets a lot more orderly:

For the lower ones, content pruning and a content audit help decide what gets improved, what gets merged and what stays.

Study methodology and limits

For the analysis we used Search Console (web search) from company websites in the US and Latin America that I have access to. Out of 358 properties we dropped unverified and subfolder ones, and kept one per domain: 232 candidates.

With criteria set before looking at results (1,000 or more impressions and 100 or more clicks in the period) we were left with 203 sites. On 17 very large sites the API returned a truncated list, sorted by clicks, which leaves out precisely the pages with no clicks, so we excluded them (with them the median is 25%). The main sample ended at 186.

A page with no clicks is one with 100 or more impressions and no visits in 28 days. The percentage was calculated per site, across the 182 with at least 10 of those pages, and we publish medians with p25 and p75. We used each page's average position, in ranges. Page type came from the URL and the country is the one that brings the most visits. # URLs were left out of the main percentage (with them it's 25.6%).

The limits: it is a convenience sample, mostly from the US, that doesn't represent the whole web. Average position blends many searches, and links inside AI summaries and AI Mode are counted inside web search with no way to separate them. A third of the pages with no clicks had no clear type. It covers a single month, with its seasonality, and the Latin America cut (39 sites) clears the minimum but is small. It is descriptive: correlation, with no proven causality.

What I would do with this data

If I had an afternoon, I would pull the list of pages with 100 or more impressions and zero clicks, split it by range and keep the top 10 ones, looking at their results page calmly before rewriting anything.

Pages from the second page onward would go into the content plan as a ranking issue, which is where the visibility that later turns into business is won. And beyond the 24% itself, I would look at how many of those pages should be bringing visits to the business.