Google Search Console gives you four headline numbers for every page and query: impressions, clicks, CTR, and average position. Each is simple on its own. The confusion starts when they move together, because they do not always move the way you expect. This guide carries one page through two months and shows how the four interact, including the case that fools almost everyone.
The four metrics in one line each
The beginner's guide to Search Console covers setup and navigation. Here is the short recap:
- Impressions: how many times your page appeared in someone's search results, whether or not they scrolled to see it.
- Clicks: how many times someone actually clicked through to your page from those results.
- CTR (click-through rate): CTR = clicks ÷ impressions. The share of appearances that turned into a visit, shown as a percentage.
- Average position: the mean ranking of your page in results, weighted by impressions. Position 1 is the top of the page. Lower numbers are better.
That last point trips people up: with average position, the number going down is good news, and going up is bad news.
The intuitive case: a page climbs the rankings
Picture a single page, say a guide on "utm parameters", that you improved between month 1 and month 2. It moved up the results, and everything responded as you would hope.
| Metric | Month 1 | Month 2 |
|---|---|---|
| Impressions | 10,000 | 14,000 |
| Clicks | 300 | 980 |
| CTR (clicks ÷ impressions) | 3.0% | 7.0% |
| Average position | 8.0 | 3.0 |
The page moved from position 8.0 to 3.0, from the bottom of the first page towards the top. Higher visibility means more impressions (10,000 to 14,000). A better position also earns a far better click-through rate, because results near the top get clicked much more often. CTR jumps from 3.0% to 7.0%, and clicks more than triple, from 300 to 980. Every metric agrees.
CTR depends heavily on position
CTR is tightly linked to where you rank. Position 1 typically earns several times the CTR of position 5, and positions past the first page earn almost nothing. So CTR is only meaningful next to position.
If a page sits at a strong position (say 2 or 3) but has a low CTR, look at the result itself. The title tag and meta description (the headline and grey summary text shown in search results) are not earning the click. That is often the fastest win in Search Console.
Quick rule: a low CTR at a high position is a copy problem, not a ranking problem. Rewrite the title and meta description before you touch anything else.
The counterintuitive case: position gets worse while clicks rise
Now the case that fools nearly everyone. Same page, a different two months. Clicks go up, but average position gets worse, the number rises. People panic, or assume the data is broken. It is neither.
| Metric | Month 1 | Month 2 |
|---|---|---|
| Impressions | 14,000 | 40,000 |
| Clicks | 980 | 1,300 |
| CTR (clicks ÷ impressions) | 7.0% | 3.25% |
| Average position | 3.0 | 6.5 |
The page kept ranking at position 3 for its original term, which still drives strong clicks. But it also started appearing for many new long-tail queries (longer, more specific searches like "how to add utm parameters to a mailchimp link"), and for those it ranks low, around position 15 to 20. Each is an impression at a poor position.
Because average position is weighted by impressions, those new low-position appearances drag the average down (the number up), from 3.0 to 6.5. Impressions nearly triple. CTR falls, not because your main result got worse, but because the denominator is now full of low-intent, low-position appearances that rarely get clicked. Total clicks still rose, because the new queries each contribute a few. The page is healthier, not worse.
A worsening average position with rising clicks usually means you are ranking for more things, not falling for the old ones.
This is the most misread pattern in Search Console. Before you treat a rising average position as a problem, segment by query and check whether your core terms held their rank. If they did, the average moved because your reach grew. This impression-weighting is one reason Search Console and analytics tools count differently, as the guide on why Search Console and GA4 do not match explains.
Queries and pages split the same clicks differently
The same clicks look different depending on which dimension you view. A dimension is the angle you slice by, usually Queries or Pages.
Say your page got 1,300 clicks in month 2. View it by page and you see one row: 1,300 clicks. View it by query and those same 1,300 clicks scatter across dozens of search terms, maybe 400 on the main term and the rest spread thin across long-tail queries. The total is the same, but the shape is completely different.
The trap is assuming the numbers add up across dimensions. They do not. A click is one click whether you slice it by query, page, country, or device. Search Console also withholds some low-volume queries for privacy, so the query rows often sum to slightly less than the page total. Treat each dimension as a different lens on the same clicks, never as numbers to add together.
In Clearly
These mistakes come from reading one metric in isolation, or comparing two periods in your head. The fix is all four numbers side by side, over time. Connect your Search Console property and the report shows impressions, clicks, CTR, and average position together, with the prior period beside each so the direction of every change is clear at a glance.
With position beside CTR and clicks, the counterintuitive case becomes readable instead of alarming: you can see reach expand, watch the impression-weighted position drift, and confirm your core queries held. You can then share the report with a client without narrating every nuance.
The takeaway
Read the four metrics together, never alone. Impressions and clicks tell you reach and demand, CTR = clicks ÷ impressions tells you how compelling your result is at the position it holds, and average position (where lower is better) only makes sense once you know whether your reach is widening. When clicks rise and position worsens, suspect new long-tail queries before a decline. That habit saves you from most of the wrong conclusions Search Console invites.