What a good CTR looks like by position

Sooner or later everyone who opens Google Search Console asks the same question: is my click-through rate any good?

It's a fair question, and there are public studies that give a rough idea of what a good CTR by position looks like. But the honest answer is "it depends", and the more useful answer is "compare against yourself". This post covers both: what the big studies found, why those averages can mislead you, and how to build your own benchmark in Search Console in about fifteen minutes.

A quick definition

Click-through rate (CTR) is clicks divided by impressions. If your page appeared 1,000 times in search results and got 40 clicks, your CTR is 4%.

Search Console shows CTR per query and per page in Performance, then Search results. Google's help page on the metrics explains exactly how impressions, clicks and position are counted.

What the public studies found

Two large, public studies are worth knowing about. Both measured how CTR changes as you move down the first page of Google.

• Sistrix analysed over 80 million keywords. They found position 1 averaged about 28.5% CTR, position 2 about 15.7%, position 3 about 11%, falling to around 2.5% at position 10.
• Backlinko, using data from Semrush, analysed around 4 million search results. They found position 1 averaged about 27.6%, the top three results took just over half of all clicks, and the first result was roughly ten times more likely to be clicked than the tenth.

So, as a very rough guide:

The pattern matters more than the exact numbers. Clicks drop steeply from position 1 to 3, then keep falling more gently to the bottom of page one. Page two gets very little.

There's also an ongoing benchmark worth bookmarking: Advanced Web Ranking's organic CTR study is updated monthly from anonymised Search Console data, and lets you filter by device and search intent.

Why these averages mislead

If you compare your pages to the table above, you'll probably conclude something is wrong. Often it isn't. Here's why.

The layout of the results page changes everything

The same Sistrix study split its data by what else appeared on the page. Position 1 CTR ranged from roughly 14% on pages with shopping results to around 47% on results showing sitelinks. On a plain page of ten organic links it was about 34%, and with a featured snippet above it, about 23%.

In other words, "position 1" isn't one thing. It depends on how many ads, boxes and answers Google puts above you. Newer features like AI Overviews add another layer, which is why ongoing studies now track them separately.

Your queries aren't average queries

Brand searches (people typing your business name) often get very high CTR, because the searcher already wants you. Broad informational questions often get low CTR, because Google answers them on the page. If most of your traffic is one or the other, your numbers will sit well away from any industry average.

Search Console's position is an average too

Search Console reports the highest position you held for each search, averaged across all searches. A query showing position 4 might really alternate between 2 and 7. Comparing a blended number to a study's precise position is always going to be fuzzy.

Mobile and desktop behave differently

Phones show fewer results on screen and more features above them. If most of your visitors search on mobile, your CTR at a given position can look quite different from a desktop-heavy site.

How to benchmark your own pages instead

The most useful comparison is your site against itself. You want to find the pages that underperform compared to your other pages at similar positions.

1. Open Performance, then Search results. Set the date range to the last 3 months.
2. Turn on Average CTR and Average position.
3. Go to the Queries tab. Use Filter rows to show Impressions greater than 100, so you're only looking at queries with enough data.
4. Click Export (top right) and choose Google Sheets or CSV.
5. In the sheet, add a column that groups each query by position: 1 to 3, 4 to 6, 7 to 10, and 11 to 20.
6. For each group, calculate your median CTR. The median is less distorted by a few unusual queries than the average.
7. Now look for queries well below the median for their group. Those are your underperformers.

You've just built a benchmark that reflects your own audience, your own search features and your own mix of brand and non-brand searches.

One tip: separate brand queries first. Filter out queries containing your business name, or put them on their own tab. Otherwise they'll inflate your top-position median and make everything else look worse.

A worked example

Say you run an online store selling handmade ceramics. After exporting and grouping your queries, your medians look like this:

Your median at positions 1 to 3 is below the public studies' averages. That's expected: many of your searches trigger shopping results, which the Sistrix data shows pull CTR down.

Now you look for outliers. One query stands out:

That's far below your own 18% median for positions 1 to 3. You search it yourself. The ranking page's title is "Shop | Clay and Co", which says nothing about mugs. Rewriting it to "Handmade ceramic mugs, made in the UK | Clay and Co" and adding a meta description that mentions delivery and price range is a small change with a clear reason behind it.

Compare the query again in four to six weeks using the date filter's Compare tab.

When low CTR is fine

Not every low number needs fixing.

• Queries on page two. Low CTR is normal there. The fix is ranking higher, as covered in how to find keywords you almost rank for.
• Queries Google answers directly. If the searcher gets a full answer from a snippet or box, a lower CTR may be the ceiling.
• Very small queries. A handful of impressions tells you almost nothing.

For the specific case of lots of impressions and almost no clicks, there's a step-by-step guide in impressions but no clicks: why it happens and how to fix it.

The short version

Use public studies to understand the shape of the curve. Use your own data to decide what to fix. The pages worth your time are the ones that underperform your own site at the same position, and the fix is usually a title and description that match what people searched.

If you'd rather not rebuild the spreadsheet every month, Sense Console checks your Search Console data for queries whose CTR sits well below what their position would normally earn, and suggests what to change.