What Is a Good Click-Through Rate in Search Console?

There is no single good number, because CTR is almost entirely determined by ranking position. A page averaging position 3 might see 10% and be underperforming; a page at position 9 seeing 3% is doing well. The only benchmark worth using is your own site, compared position band against position band.

What Search Console's CTR actually measures

The formula is unremarkable: clicks ÷ impressions. The trouble is what Google counts as an impression.

An impression is recorded whenever your URL appears in a result set the user loaded — including results far down page two, page three, or inside a collapsed "more results" block the user never scrolled to. You were technically shown. Nobody technically looked.

This is why a raw sitewide CTR figure is close to meaningless. A site with thousands of page-four impressions and a handful of page-one rankings will show a dismal average, and that average says nothing about whether the page-one listings are working.

Three properties follow from this, and they explain most confusing CTR numbers:

Position dominates everything else. Title and description quality matter at the margins. Position sets the ceiling.
Impressions are cheap and clicks are not. Ranking for more queries badly will lower your CTR while raising your traffic.
The average hides the distribution. One branded query can carry the whole site's number.

The shape of the CTR curve

Published studies disagree on the exact figures — they use different samples, countries, devices and years — but they agree on the shape. Position one takes a large share, the drop to position two is steep, and by the bottom of page one you are into low single digits.

The most widely cited recent figure comes from Backlinko's 2023 analysis of roughly four million Google search results, which put the average CTR for the first organic result at 27.6%, with a sharp decline after it. Advanced Web Ranking publishes updated curves broken out by position, device and query type, which is more useful than any single headline number.

Position band	Roughly what to expect	What it tells you
1	High teens to high twenties, sometimes more	Anything well below this on a non-branded query means the snippet is losing clicks
2–3	Roughly halving from position 1	Normal range is wide; judge against your own pages
4–10	Low single digits	Title and description work has the least leverage here
11+	Fractions of a percent	Impressions are near-worthless; the fix is ranking, not copy

Treat that table as the shape of the curve, not a target. Two things make any published average unreliable for your site: query type and SERP features. An informational query answered by an AI Overview or a featured snippet behaves nothing like a navigational query where the user is looking for a specific brand.

The three distortions that ruin the number

Before you conclude anything from a CTR figure, strip these out.

1. Branded queries. People searching your company name click your listing at rates that dwarf everything else — often several times the site average. Leave them in and your sitewide CTR looks healthy while every informational page underperforms. In Search Console, add a Query filter set to doesn't contain your brand name and any common misspellings, then read the number again. That figure is the one that reflects your SEO.

2. Average position is an average. Search Console's position column is averaged across every impression for that row. A page ranking 2nd for a low-volume query and 40th for a high-volume one shows an "average position" somewhere in the twenties that describes neither reality. Filter to a single query before comparing CTR to position, or you are comparing to a number that does not exist.

3. SERP features sitting above you. An AI Overview, a featured snippet you do not own, a shopping carousel, or four ads push the first organic result well down the page. Your position has not changed; the amount of screen above you has. This is the most common cause of a CTR decline with flat rankings — see is Google search traffic declining for the wider trend, and how do I show up in Google AI Overviews for the AI-specific half.

The only benchmark that means anything

Compare your pages to your own pages, matched on position band and query type.

The method takes about fifteen minutes:

Open Performance → Search results and set the date range to the last three months.
Exclude branded queries with a doesn't contain filter.
Switch to the Pages tab and enable the Position column.
Export to a spreadsheet and group pages into position bands — 1–3, 4–6, 7–10.
Take the median CTR within each band. That is your baseline.

Now the useful question becomes answerable: which pages sit well below the median for their own band? Those are the ones where the listing itself, not the ranking, is losing clicks. Everything else is a ranking problem wearing a CTR costume.

This distinction matters because the two problems have completely different fixes, and rewriting titles on a page that ranks 14th is wasted effort.

What to actually do about a low CTR

Only once you have confirmed the page ranks well and still underperforms its band.

Rewrite the title tag first. It is the largest text in the result and carries most of the decision. Lead with what the searcher asked for, be specific, and drop generic padding. What makes a good title tag covers the patterns that work.

Rewrite the description as ad copy, not a summary. It is not a ranking factor, but it is the sales line under the headline. Google rewrites it often — that is fine, a good one still gets used most of the time. See do meta descriptions affect SEO.

Check what Google is actually displaying. Search the query yourself in an incognito window. A truncated title, a date stamp making the page look stale, or a rewritten description you did not authorise all cost clicks invisibly.

Match the intent more obviously. If the query is a question and your title is a product page name, the click goes to whoever phrased it as an answer. This is often the real cause.

Go after the snippet. Winning the answer box changes the geometry of the whole result — how to get a featured snippet is the companion move.

And accept the ceiling: on a query where an AI Overview answers the question completely, a lower CTR is the new normal rather than a failure. Your job there is to be the source it cites.

One caveat about the data itself

Search Console does not show you everything. Rare queries are filtered out for privacy reasons, which means the query-level rows will not sum to your totals — sometimes not even close. The Pages tab and the Queries tab are aggregated differently and will not reconcile either.

This is expected behaviour rather than a bug, and it is one of several reasons the tool disagrees with your analytics platform. That whole mess is untangled in why Search Console and Google Analytics don't match.

Where CTR sits in the bigger picture

CTR is a diagnostic, not a goal. Optimising it in isolation gets you clickbait titles that win the click and lose the visit. The metric only earns its place when it is read alongside rankings, conversions and revenue — the full chain is in how do I measure SEO ROI.

And the honest constraint underneath all of it: position sets the ceiling, and position is earned with relevance, content quality and links. If your pages sit outside the top handful of results, no amount of title tuning will fix the number. That is the point at which authority work matters, and Backlinkster exists for exactly that — one-for-one in-content link swaps between real site owners, verified live by code. Five a month free, with plans from $19.

Frequently asked questions

What is a good CTR in Google Search Console? There is no universal figure, because position drives most of it. A useful rule is to compare each page against the median CTR of your own pages in the same position band, with branded queries excluded. Beating your own median is the meaningful benchmark.

Is a 2% CTR bad? Not necessarily. At an average position of 8 or 9 that is roughly normal. At position 1 on a non-branded query it is a clear signal that the title and description are losing clicks that the ranking already earned.

Why did my CTR drop when my rankings stayed the same? Usually something new appeared above you — an AI Overview, a featured snippet, ads, or a shopping block. Your position is unchanged but there is more between the top of the page and your listing. A sudden rise in low-quality impressions does the same thing.

Does CTR affect rankings? Google has consistently said click data is not used as a direct ranking signal in the way SEOs often assume, and attempts to manipulate it do not work. Improve CTR because it brings visitors, not because you expect a ranking lift.

Why don't my Search Console query numbers add up to the totals? Google filters out rare queries to protect user privacy, so the query rows are an incomplete subset. The Pages and Queries tabs are also aggregated differently, so they will not reconcile with each other.

Should I exclude branded queries when measuring CTR? Yes, for any assessment of SEO performance. Branded searches convert at far higher rates and will mask underperformance everywhere else. Use a doesn't contain query filter on your brand name.

How long should I wait before judging a new page's CTR? Give it at least four to six weeks of stable impressions. Early data is dominated by a handful of exploratory rankings that move around before settling.

The bottom line

Stop looking for the number. CTR without a position band and a branded-query filter attached is not a measurement, it is noise — and most of the "average CTR" figures circulating were produced on samples that look nothing like your site. Build the fifteen-minute baseline from your own data, find the pages that underperform their own band, and fix those. For everything else, the lever is not the snippet. It is the ranking.