High Impressions, Low Clicks? How to Diagnose CTR Decay

The short version: high impressions with low clicks means Google is showing your page and searchers are choosing someone else. If your average position held steady, that's a click-through problem, not a ranking problem — and the fix is the title and meta description, the cheapest refresh there is. Before you rewrite anything, rule out the two cases where the title isn't the culprit: the page sits at position 8–10 (low CTR is normal there), or an AI Overview is eating the click above you.

That distinction — did the position hold, or did it slip? — is the whole diagnosis. Get it right and you spend twenty minutes on a title instead of two days rewriting a page that didn't need it.

I run this check on my own sites by hand, so this is the manual version — free Search Console, no paid tool.

Why you get impressions but no clicks

Two metrics, and it helps to be precise about them. An impression is counted every time your page shows up in someone's results. A click is when they actually pick you. CTR is just clicks ÷ impressions — the share of people who saw you and chose you.

High impressions, low clicks means the first number is healthy and the second isn't. You're winning the visibility and losing the choice. Something about how your listing looks in the results — the title, the description, the URL, or what's sitting around it — isn't convincing people to click.

But "low" only means something relative to where you rank. CTR falls off a cliff as you go down the page, and that's normal, not decay:

Across the published CTR studies, the top three organic results capture roughly half of all clicks between them, and by the bottom of page one CTR drops under about 3%. The exact numbers vary a lot study to study — position-1 CTR is reported anywhere from the high single digits to ~40% depending on the dataset and the SERP — so treat them as a range, not gospel. (Figures aggregated across Backlinko, SISTRIX, seoClarity, First Page Sage, and GrowthSRC; see navboost's roundup.)

The takeaway from that spread: judge your CTR against your position, never against a flat target. A 4% CTR is a problem at position 2 and completely fine at position 8. Which is exactly why the next step checks position before it touches your title.

Diagnose it in Search Console: the >500 impressions, <2% CTR filter

You want to surface the pages that get seen a lot and clicked rarely — because those are the ones where a title rewrite has the most to work with. Here's the filter I use.

1. Open Search Console → Performance → Search results. You'll see the four metric cards: Clicks, Impressions, Average CTR, Average position. Make sure Average CTR and Average position are both toggled on — you'll need them.
2. Switch to the Pages tab (under the chart). Now each row is one URL with its clicks, impressions, CTR, and position.
3. Filter to pages with real exposure. You want pages with meaningful impressions — I use a floor of roughly 500 impressions over the period, so you're not reacting to a page nobody sees. (Sort the Impressions column descending and work down, or set a date range wide enough that your real pages clear the floor.)
4. Find the low-CTR rows. Among those high-impression pages, flag anything with a CTR under ~2%. Sort the CTR column ascending so the worst offenders rise to the top. High impressions + sub-2% CTR = a page being shown often and skipped often.
5. Read the position on each flagged row. This is the step that decides everything, and it's the one most guides skip. A low CTR at position 3 is a title problem. A low CTR at position 9 is just position 9. Same CTR number, completely different diagnosis — the position column is what tells them apart.

The exact thresholds aren't sacred — 500 and 2% are starting lines, not laws. Loosen them on a small site, tighten them on a big one. The point is to end up with a short list of pages that already earn impressions but don't convert them.

If you're doing this as part of a wider sweep for pages losing traffic, this filter is the CTR branch of that larger check — see how to find declining pages in Google Search Console for the full compare-mode routine that surfaces every kind of decline, not just CTR.

Position steady + CTR down = rewrite the title, not the page

Here's the fork the ranking articles never draw cleanly. Once you've got a page with high impressions and low clicks, look at what its position did over time:

• Position held, CTR fell → this is a click-through problem. You're ranking where you always did; people just stopped choosing you. The fix is the title and meta description — how the listing reads in the results. This is the cheapest refresh there is: you're editing two lines of text, not rewriting an article, and you're not waiting on a ranking change to see the payoff.
• Position dropped, CTR fell with it → this is a ranking problem wearing a CTR costume. Your CTR looks bad partly because you fell to a spot that naturally earns fewer clicks. A better title won't save a page that slipped from position 3 to 9 — you need an actual content refresh to win the ranking back.

Confusing the two is how people waste a weekend. They see "low CTR," rewrite the title, and wonder why nothing recovers — because the real problem was a ranking slide a new title was never going to fix. Check the position first. It tells you which of two very different jobs you're actually looking at.

When it is a title problem, keep the rewrite honest: lead with the searcher's actual query, be specific about what the page delivers, and don't promise something the page doesn't pay off (that just trades a low CTR for a high bounce). Then treat the change as an experiment — and measure whether the clicks actually came back before you call it a win.

When it's not your title

Before you rewrite anything, rule these out — because in every one of them, a better title is either useless or a rounding error:

• You're on page 2 (or the bottom of page 1). Positions 8–20 barely get clicks at any CTR, because almost nobody scrolls that far. The fix isn't a better title, it's a better ranking. Don't sink title effort into a page whose real problem is that it's too far down.
• An AI Overview is eating the click. AI Overviews now appear on roughly a third of result pages, and when one shows up it answers the query in place and pushes the organic links down. Studies put the CTR hit to top-ranking pages at somewhere around 50–70% on those searches (Ahrefs analyzed ~300,000 keywords in early 2026 and found AI Overviews correlated with a ~58% drop in clicks for the top result). If you held your ranking but lost click-through, search the query yourself — if an AI Overview or a pile of features sits above you, no title fully wins that click back.
• Other SERP features are crowding you out. A featured snippet, a People Also Ask box, a map pack, shopping results, or ads can all push the true organic results below the fold. Same effect: high impressions, suppressed clicks, and not your title's fault.
• It's a branded-vs-unbranded mix. If a page ranks for a lot of informational queries where people are just browsing (or where they're actually looking for a competitor's brand), its blended CTR will look low even with a perfectly good title. Filter the Queries tab to see what it's actually being shown for before you conclude anything.

None of these are things a title rewrite fixes. Spotting them is what keeps you from "optimizing" a page that was never the problem. A low CTR is one of five patterns a decaying page can make, and each one points at a different fix — the title rewrite is only the right answer for one of them. The honest answer, sometimes, is: your CTR is low because the SERP changed under you, and that's the game now.

A before-and-after title, and what to expect

Say a page ranks at position 3 for "cold brew coffee ratio" with a title like:

Cold Brew Coffee — Our Complete Guide | Beanmarket

It holds position 3, but CTR is stuck at 1.8% — well under what position 3 usually earns. The title buries the query behind a generic "complete guide." A tighter rewrite that leads with the exact thing people searched:

Cold Brew Coffee Ratio: The 1:8 Rule (and How to Adjust It)

Now the listing answers the query at a glance and adds a concrete detail (the ratio) that signals the page actually delivers. That's the kind of change that lifts CTR on a page whose ranking was never the issue.

I'm using round numbers to show the shape of the fix, not reporting a case study — your lift depends on your query, your competition, and what the rest of the SERP looks like. The way you know it worked is to measure it: compare the four weeks after the change against the four weeks before, on that one page, in Search Console. Same before/after discipline as checking whether a content refresh actually worked.

Where this fits in the bigger picture

A CTR problem is one of the cheapest kinds of decay to fix, but it's easy to miss because the page still ranks — it just quietly stops earning clicks. That's why it's worth a dedicated check rather than lumping it in with "traffic is down."

The full loop looks like this: find the pages losing clicks, sort out which are CTR problems (this article) versus ranking losses, prioritize them by clicks at stake, fix the cheap ones first, and verify the fix worked. CTR decay is where you get the fastest, cheapest wins in that loop — often a twenty-minute title edit instead of a full rewrite.

Catch CTR decay without the manual filter

Running this check by hand — filtering to high-impression pages, spotting the low-CTR ones, and remembering to look at whether position held — is a real routine, and it's the kind that stops happening after a busy month. Miss it, and a page keeps its ranking while quietly bleeding clicks that a two-line title fix would have kept.

That's a job RefreshRadar does for you. It watches every page in your Search Console and flags CTR-decay pages separately from ranking losses — so when the problem is a title, you're told it's a title, not handed a full-refresh to-do you don't need. Each flag comes with the query losing ground and the clicks/month at stake, in plain English.

Connect Search Console read-only and see what's decaying on your site — free — no card, about 30 seconds, and we never post or change anything. If nothing on your site is decaying, we'll tell you that instead.