← Blog
SEO · November 12, 2026 · 9 min read

Low Click-Through Rate in Search Console: When a Title Needs Rewriting, When It Doesn't, and How to Test One Without Losing the Position

D
Low Click-Through Rate in Search Console: When a Title Needs Rewriting, When It Doesn't, and How to Test One Without Losing the Position

Click-through rate is the Search Console number people act on fastest and understand least. A page at 4 percent looks like a problem until you notice it sits at position 7, where 4 percent is good; a page at 12 percent looks fine until you notice it's at position 1, where it's losing two-thirds of the clicks it should have.

And of the pages that really do have a low CTR for their position, most aren't the title's fault, which doesn't stop the title being the thing everyone rewrites.

This post covers what CTR is normal at each position so the comparison is fair, the four reasons CTR is low that a title can't fix and how to recognise each, the two it can, how to find the pages worth testing (high impressions, positions 3 to 10, CTR below the position's norm), the one-title-at-a-time test with a four-week read, and what never to change in the same edit.

The titles post covers how to write a title; this is how to know whether a title is the problem.

The short answer

Compare CTR with the norm for the page's position, not with a fixed number. Rule out the results page first: AI Overviews, shopping units, ads, intent mismatch and brand results all lower CTR without the title being wrong.

For the pages left, rewrite one title at a time to say the query's words plainly in 50 to 60 characters, leave everything else alone, wait four weeks, and keep it only if CTR rose and position held.

What's normal at each position

Average positionTypical organic CTROn a query with ads, shopping or an AI Overview above
125 to 35%10 to 20%
212 to 18%6 to 10%
38 to 12%4 to 7%
4 to 55 to 8%3 to 5%
6 to 102 to 4%1 to 3%
11 to 20Under 1%Under 1%

The ranges come from large studies of organic click distribution and vary by query type; buying queries with shopping units run at the low end.

Search Console's "average position" is an average across every impression, so a page at 4.2 is sometimes 2 and sometimes 7, which blurs the comparison; it's still the right comparison, and the rank tracking post covers why the 28-day average is the number to use.

Four reasons a title can't fix

  1. Something answers the question above you. An AI Overview, a featured snippet, a People Also Ask box. The searcher got their answer without clicking anyone. Check the query in a private window; if there's an answer box, your CTR is low because the results page is, and the AI Overviews post covers what still gets the click.
  2. Shopping units and ads push you down the screen. On "best X" queries a shopping carousel and three ads can put position 1 below the fold on a phone. Position 1 at 12 percent on such a query is normal.
  3. Intent mismatch. The page ranks for a query it half-fits: an informational post ranking for a buying query, or a guide for one product ranking for a different model's name. Searchers skim past because the snippet is for something else. The fix is a different page, not a different title.
  4. Brand results. If the query includes a brand or product name, the brand's own site, Amazon's listing and the manufacturer take most clicks whatever position you hold. A review at position 3 on "[product name]" with 5 percent is doing fine.

Two reasons it can

Finding the candidates

  1. Search Console → Performance → Search results, last 3 months, Pages tab, with impressions, CTR and position columns on.
  2. Filter to position between 3 and 10 (below 3 the title matters less than the result features; past 10 nobody sees it).
  3. Sort by impressions. The top twenty pages with CTR below the norm for their position are the list.
  4. For each, open the Queries tab for that page: if the top query's words aren't in the title, it's a candidate. If they are, check the results page for the four reasons above before touching anything.

Most sites end up with five to ten real candidates. That's a season's worth of tests, which is the right pace.

The test

  1. One page, one title. Not five pages at once; you won't know which change did what, and a site-wide title rewrite looks like churn.
  2. Rewrite to the query's words, in the searcher's order, saying what the page delivers, in 50 to 60 characters. Keep the keyword at the front.
  3. Change nothing else. Not the H1 (unless it also lacks the keyword, in which case align it and note that you changed two things), not the URL, not the meta description in the same week, not the content.
  4. Note the date and the previous four weeks' CTR and position for the page.
  5. Wait four weeks. Two is weather. Compare the four weeks after with the four before, for both numbers.
  6. Keep it if CTR rose and position held or improved. Revert it if position fell, whatever CTR did. Then the next page.

What never to change in the same edit

Doing it with AffSERP

Guides written by AffSERP's X Best Review tool put the target keyword at the front of the title in 50 to 60 characters with the H1 and slug aligned, which removes the two fixable causes before the page is published; what's left to test is the wording after the keyword, and that's an edit in WordPress that doesn't touch the rest of the page.

When the Refresh option regenerates a guide, it keeps the URL and the title unless you change them, so a title you've tested and kept survives the refresh. The Search Console reading is yours, monthly, and the list of candidates is usually short.

FAQ

What is a good click-through rate in Google Search Console?

It depends almost entirely on position. Across studies of organic results, position one takes roughly 25 to 35 percent of clicks, position two 12 to 18, position three 8 to 12, positions four and five 5 to 8, and positions six to ten 2 to 4 percent each, with everything lower on queries that show ads, shopping units or AI Overviews above the results. So a 3 percent CTR is excellent at position 9 and a disaster at position 1. Judge a page's CTR against the norm for its average position, never against a single number.

Why is my CTR low even though I rank well?

Usually something on the results page that isn't your snippet: an AI Overview or featured snippet answering the question above you, a shopping unit or ads pushing you down the screen, a query whose intent your page half-matches so searchers skip it, or a brand's own site taking the clicks for its own name. None of those is fixed by a title. Once those are ruled out, the two causes a title can fix are a title that doesn't say what the searcher typed, and a title Google has rewritten into something blander because yours was too long or didn't match the page.

How do I improve CTR in Google search?

One title at a time. Pick a page with high impressions, average position between 3 and 10, and CTR below the norm for that position. Rewrite the title so it contains the query's words in the searcher's order, says what the page delivers, and fits in 50 to 60 characters; leave the H1, the URL and the content alone. Wait four weeks, compare CTR and position against the previous four, and keep the change only if CTR rose and position held. Then do the next one.

Does changing a title affect rankings?

It can, in both directions, which is why the test changes only the title and reads position alongside CTR. A title that matches the query better often lifts position slightly as well as CTR; a title that drops the keyword or wanders from the page's content can lose position, and a title rewritten every fortnight never settles. Change one thing, wait a month, read both numbers, and never change the URL or the H1 in the same edit.

Keyword at the front, title and H1 aligned, from the first publish

AffSERP guides carry the target keyword first in a 50-to-60-character title with the H1 and slug aligned, and Refresh keeps a title you've tested.

Start with free words
D
Written by Daniel R. SEO Lead, AffSERP

Daniel leads SEO at AffSERP. He writes about keyword research, search intent and site structure, favouring methods you can run with Google and Search Console before reaching for a paid tool.

More about Daniel →