I gave a talk this evening on tactics and techniques to improve organic click through rate at the wonderful MKGO event in Milton Keynes. You can get the slides here. But one thing I mentioned as being critical to improving click through rate is a true understanding of what your click through rate actually is to start with.
CTR studies are insightful in many ways. A couple definitely worth checking out are this one from Advanced Web Ranking and this one from Moz. But you can’t take a study as a true reflection of what YOUR click through rate should be for any given query. There are so many variables now – much more so than a few years ago.
We’ve got featured snippets of different types, more visual organic listings (video thumbnails, for example), different devices to consider, some queries return maps, some return answers from Google above the results… the list is seemingly endless.
So while a study can give you a loose idea of click through rates on the whole or by category, if you’re serious about optimising yours, you’ll need to look at your own data.
CTRs at a Page Level
Using Search Console (and I’ll go into a bit more detail around this later), you can find click through rate for any given period of time (going back a maximum of 16 months in the new Search Console). You can see the click through rate on average for your whole website:
But really, that’s not an overly useful figure. In this case, there are thousands upon thousands of different queries/keywords generating those clicks, each with different positions, types of results pages and similar.
You can break it down by page by clicking on the “pages” tab as below:
From here, you’ll see a list like this:
It’s a list of pages on your site along with the clicks, impressions, position and click through rate. If you click on any one of those URLs, you can see a page like this:
But while this is a little more useful than the sitewide average, it’s still not really granular enough. If we click on “queries,” we can see a tab that shows all the queries that have driven impressions and clicks to this page:
At the bottom, we can see “1-10 of 999.” 999 is the limit for queries that Search Console will show. It means that there were at least 999 queries sending impressions or clicks to this page. And even from the ones we see here, we can see click through rates and positions are hugely variable.
So even a page level report on click through rate doesn’t necessarily yield the granular level of data you need to get a true measure on click through rate.
That’s why I measure click through rate at a query level. And I take a lot of data to also work out what click through rate is per query at various different positions.
Query Level CTR Info From Search Console
So, let’s rewind a touch. And instead of going into “Pages” as we did above, we’re going into “Queries.”
Then we’ll see this:
From there, you can click through to the query you’re looking for data on (choose high priority queries to run this assessment) and you’ll see this:
The position there shows 4.6. But this is a global average. And this site targets a UK audience. If it shows up outside the UK, the position is generally low (skewing our overall averages) and that audience isn’t relevant for us. So we now filter to show only UK impressions and clicks.
Head down to “Country,” here and then select UK and “apply,” as below:
Then we see updated figures:
The average position in the UK (which we are most concerned with) is 1.9 over the timeframe we’re looking at.
This is a useful figure.
But it’s still not enough information…
What this tells us is that from an average position of 1.9 click through rate is 12.2%. But I need to know how that click through varies with variances in the position.
Why?
Because if I want to test things that influence click through rate, I have to allow for the fact that other things might change too. What if I make a page title change and then my ranking drops a place? Obviously, my CTR will drop too. But if I have data about historic click through rates at multiple positions, I can benchmark my “new” click through rate and assess whether it is better or worse than CTR at that position before my changes.
It’s not perfect. And you might not have position data for loads of positions. But where you have the data, extracting it at least gives you something to benchmark any future performance against.
So let’s export that information.
Switching Back to Old Search Console
All the screenshots so far have been taken from the new Search Console. This is an improved version of Google Search Console but is still very much in development, with some features and data not yet available. Within the new Search Console is a prominent option to”switch back to the old version,” and to export the report we want to export, we’ll have to head back. 🙁
You’re going back the same way you did in the new version, “queries,” report and then filtering by country.
Then you need to select the maximum date range (a measly 90 days in the old version) and select to view by date:
You’ll then get a report filtered for that one query you selected, filtered for UK only and showing the clicks, impressions, CTR and position data by day.
Make sure you’re showing all rows and then download that report:
Below I’ve shown you (using an export on another query, which has more varied position data) what I then do with those exports.
The export looks like this.
I then remove all columns apart from position and click through rate, to end up with something that looks like this:
I then remove the decimal places from positions to get something like this:
Order the columns by position, so all the 2s are together, all the 3s are together etc. Add a row below each group and get an average of all CTRs at that position so you get something like the below:
Delete everything but those averages and you eventually get a list of average CTR by position for that query:
So what do you have now?
You’ve got CTR by position for a specific query on your website. That’s much more insightful than a study for you. It means that you can analyse with a little more accuracy the impact of any CTR optimisation work you do even if things also fluctuate by a position or two.