Why are impressions up but clicks down in Search Console?

Search Console impressions can rise while clicks fall because clicks equal impressions times click-through rate (CTR), so a drop in CTR can outweigh the added impressions. The rate can fall because added impressions arrive on queries or positions that click less, because the same queries click less, or because Google changed how it counts.

• Google defines click-through rate as the click count divided by the impression count, and in general counts an impression whenever an item appears in the current page of results (Google, Search Console Help, read 30 October 2026).
• Google's data anomalies page says impressions were not accurately reported from 13 May 2025 until 27 April 2026 and that clicks were not affected (Google, Search Console Help, read 30 October 2026).
• Google says sites that appear in AI features such as AI Overviews and AI Mode are reported in the Performance report within the Web search type (Google, Search Central documentation, read 30 October 2026).
• Penlid's CTR gap tool lists a query only with at least 100 impressions, a CTR at most 75% of your own curve, and the top of its Wilson interval still below the curve (Penlid, CTR gap tool, read 30 October 2026).

Why can impressions rise while clicks fall at the same time?

Impressions and clicks can move apart because they count different events, and the click-through rate between them is free to change. Google in general counts an impression when an item appears in the current page of results, a click when someone follows a link from Google to your site, and defines CTR as The click count divided by the impression count.

Summed over your queries, clicks equal each query's impressions times its CTR (our arithmetic), so clicks fall when lost CTR outweighs the clicks the added impressions bring. That happens in three ways: a counting change in Search Console, a mix change (added impressions on queries or positions that click less), or a rate change (queries you already had click less).

Is the rise real, or did Search Console change what it counts?

The rise can be partly a counting effect, so rule out an uneven comparison, a documented data issue and a change in the search types included before you look at queries. Compare windows made of whole weeks so every weekday appears equally often (our advice), and leave out the newest days, which Google says can be preliminary (see why the last days in Search Console change).

We found no Google statement of how far off the reported impressions were in that period, or of whether the multimodal rollout, which adds a multimodal search type filter, changed the totals of existing Web views. Penlid's anomaly check takes the change in impressions minus the change in clicks, in percentage points, over 28 days on each side of 27 April 2026; the guide to telling a real traffic drop from a data anomaly covers the rest.

Did the extra impressions land on queries that click poorly?

The extra impressions may have landed on queries that click poorly, and the blended CTR then falls even when no single query's CTR changes. Branded search is the clearest case: Google says Branded queries typically lead to higher-ranking pages from your site and result in higher click-through rates, and Penlid's CTR gap tool page notes that branded queries usually have a higher CTR than the rest. If branded impressions shrink while non-branded ones grow, the blend drops with every segment's own rate unchanged. Split the two first; the guide to splitting branded from non-branded traffic shows how.

Position works the same way. Google notes that a position number can mean different things in different situations, and the position Search Console reports is an average, so a page that starts appearing for many more queries at weak positions adds impressions faster than clicks (our reasoning). Cut the data by page, country and device as well as query type, and find the segment where impressions rose most and the rate is lowest.

Did the click rate fall on the queries you already had?

If the click rate fell on queries you already had, the change is in the result, its surroundings or the searchers, and Search Console cannot say which. For the related case of impressions holding steady while clicks drop, Google's traffic-drop guide says you might not be generating the best page title and snippet that you could, or perhaps other sites had a more appealing rich result. Google's page on dimensions lists a common use of query data: to find queries with high impressions and low CTR for opportunities.

AI features sit in the same Web totals, so test them too. Google notes that An AI Overview occupies a single position in search results, so every link in it shares that position, and that Clicking a link to an external page in AI Mode counts as a click. The file cannot separate queries whose results pages have an AI answer from those without (Penlid's CTR gap tool page). We found no Google figure for how AI features change clicks, so we give none. The guide to the Generative AI performance report covers the report Google offers on these impressions; Google's help page for it describes impressions only and mentions no clicks.

How do you tell a mix change from a rate change in your own data?

You tell a mix change from a rate change by building your own CTR-by-position curve for each window and comparing the curves and the queries below them. The free CTR gap tool builds the curve and the gap list in your browser from a Queries export; run it on one window's file at a time, so the comparison is yours to make.

1. Export two windows of equal length with the same filters and search type: in Performance, Search results, open the Queries tab with Clicks, Impressions, CTR and Average position switched on, then choose Export and Download CSV.
2. Enter your brand terms, so branded queries, whose CTR runs far above the rest, leave the curve and the list.
3. Compare the curves. Each is weighted by your impressions and smoothed so it never rises with position. A lower curve at most positions in the later window points to a rate change; a curve that has barely moved points to a mix change (our reading).
4. Compare the gap lists. A query's click gap is its impressions multiplied by the difference between the curve's CTR at its position and its own CTR. Queries new on the later list are where to look first.

The tool draws a curve point at a position only with at least 1,000 impressions from at least 5 queries of 100 impressions or more there, and it needs at least four points, so a small site may get no curve. A Wilson interval is a range of plausible values for a query's true rate, and Penlid's methodology page explains why it is used instead of the textbook normal interval.