How to improve CTR: find the pages that get impressions but few clicks with Google Search Console

What a “low CTR” means, and what it does not

CTR depends first on position: a result at position 3 is clicked far more often than one at position 10. So 1% can be a good score at position 12 and a poor one at position 3.

It also depends on the query and on the results page. Brand searches get high CTRs. Some queries are answered directly on the results page (a definition, a calculator, a map), which leaves little to click. Ads and other result blocks push organic links down.

Google does not publish a CTR curve you can rely on, and third-party studies disagree. Your own site is the best baseline: compare each page with the pages of yours that sit at a similar position.

Step 1: set up the report

1. Open Search Console, then Performance and Search results. Set the period to the last 3 months, so that a few unusual days do not decide the result.
2. Check that Average CTR and Average position are switched on at the top of the report. The table only shows the columns you enabled.
3. Open the Pages tab and sort by impressions, highest first.

To keep brand searches from hiding the picture, add a Query filter, choose “Queries not containing” and type your brand name. Brand searches have a high CTR for reasons that have nothing to do with your titles.

Step 2: find the pages clicked less than their neighbours

1. In the table, read the Position and CTR columns together. Group the pages by position range (1 to 3, 4 to 6, and so on) and by type, comparing product pages with product pages and articles with articles.
2. Within a range, look for a page whose CTR is much lower than the others, and whose impressions are high enough to matter.
3. As a rule of thumb (not a Google threshold), ignore pages that have only a few dozen impressions: one click more or less moves their CTR a lot.

A page with many impressions and a CTR well under its neighbours is a candidate. At position 15, a low CTR is not the first thing to look at: that is a ranking question first.

Step 3: check which queries pull the CTR down

A page with a low CTR does not always have a weak title. Click the page to filter on it, then open the Queries tab.

• One query dominates and fits badly. The page shows up for a query it does not really answer, so the people who see it do not click. The title is not at fault: either the page should not target this query, or it lacks content that answers it.
• The queries are the right ones. Compare the CTR of each query with what you would expect at its position. If your main queries are all clicked less than your other pages at the same position, the snippet (title and description) is the first suspect.

Step 4: look at what the searcher sees

Search the query yourself, ideally in a private window, and read the page as a searcher would: the title, the description, the address, and the results next to yours. Results vary by country, device and personalisation, so treat this as a sample, not as the truth.

• Does the title say what the page offers and for whom, in the words people use for this query?
• Does it bring something the neighbouring results do not: a price, a free trial, a date, a figure, a format? Only if it is true.
• Is it cut off? Google shortens titles that are too wide for the space available (the cut depends on the width, not on a fixed number of characters), so put the important words first.
• Is the title Google shows different from yours? Google may replace a title that is empty, outdated, duplicated or that does not describe the page. A replacement is a clue to compare with what the page really says.
• Is the description the one you wrote? Google builds the snippet mainly from the content of the page, and uses your description only when it describes the page better, so it often differs. If it does, check that the page states the point clearly near the top.

Write what is true. A title that promises what the page does not deliver may win a click and lose the visitor.

Step 5: change one thing, then measure

Rewrite the title first, then the description, one page at a time or in small groups, so that you can tell what worked. Before changing anything, write down the date, the query, and the page’s CTR and average position.

Google has to recrawl the page and update what it displays; the URL Inspection tool shows the date of the last crawl. Then compare equal periods, for the same queries and at a similar average position: a CTR that rises because the page moved up proves nothing about the title.

Nothing guarantees a gain. A better snippet gives searchers a better reason to click; it does not decide what Google shows.

A worked example (fictional data)

The site app.example sells project management software. Three of its pages sit at a similar average position over 3 months. These numbers are invented to illustrate the method; they do not come from a real site.

Reading: the three pages hold the same range of positions, but /pricing/ has roughly a quarter to a third of the CTR of the other two, although it is the page shown most often. It is the page to look at.

Next step: the Queries tab shows that most of its impressions come from “project management software pricing”, so the match is right. The title, “Pricing | App”, says neither what the plans cost nor that there is a free trial, as the results around it do. Decision: rewrite the title to state the starting price and the trial, if both are true, and leave the description for a second step. The number to watch afterwards is the CTR of /pricing/ at a similar position.

Related guides

• SEO quick wins: keywords stuck on page 2If the page sits at positions 11 to 20, it is a ranking question first: this guide covers what to check.
• Keyword cannibalization: how to find and fix itIf two of your pages show up for the same query, they may be splitting its clicks.