If there is one thing that is misunderstood more than anything in Google Analytics, it is the definition of Bounce rate.
Bounce rate perhaps intuitively signals that a user has come to a page and left it because they didn’t like what was on the page. However, this is rarely the case.
What is worse, all the explanations I have read about what Bounce rate is, are using a language that doesn’t make it clear to an average user of Google Analytics.
The challenge in most of the definitions that I have
The four concepts that needs to be defined better are:
- Single-page session
- Engagement Hits
- Triggering a request to Analytics
- Session Duration
Simple definition of Bounce rate in Google Analytics
So. For those of you with little time. To explain Bounce rate with the most simple words I can use, the definition would probably be:
A user who visits a page on your
website, stays on it for as long as they like, but do not visit another page before they leave. They are a bounce.
(Except… and here it begins… if you have told your Google Analytics account to register certain activities like scrolling, is using or simply registering that a user has stayed on a page for let’s say 30 seconds… and by doing so, shouldn’t be counted as a bounce. By adding such rules you create an Adjusted Bounce Rate, meaning you adjust it to better tell you the story of how many people have actually didn’t get what they wanted from the content.)
The “storytelling” definition of Bounce Rate in Google Analytics
In order to fully understand what Bounce rate is, we need to dig a bit deeper into how the Google script that made Google Analytics work on your website, actually listens in to user activity on a page.
Consider a situation where you go back in time and you’re a teen with a really annoying little sibling. One of those who really
In this story, this little sibling represents the Google Analytics tracking code.
There are some things that make your sibling listen and register everything you do. For example, they know that when you close the door to your room or when you use a curse word or lower your voice, that you are probably talking about something interesting enough to listen into.
They know this as there have been historic events where they have registered that these moments are worth paying attention to. Actually, you are the one who
Equally. Google Analytics will only care about the moments you have previously defined as important by defining it, or only listen in on the normal signals it already knows are important.
These situations are called Engagement Hits in Google Analytics lingo.
If there is no engagement hit, then the interaction will count as a bounce. Just as anything you try to tell your annoying sibling, which they are not wired to listen in on, will bounce of their attention disorder like flat stones bounce on water.
That is to say, if a user comes to your website and only visits one page, without anything happening whilst they are there that could trigger the tracking script to react, then Google Analytics will count this session as a bounced session.
The Accurate definition of Bounce Rate in Google Analytics
Google writes: A bounce is a single-page session on your site. In Analytics, a bounce is calculated specifically as a session that triggers only a single request to the Analytics server, such as when a user opens a single page on your site and then exits without triggering any other requests to the Analytics server during that session.
What this means is that the tracking script you have installed on your website only listens to the things you have told it to listen to. They will register a page view by default, but they will count it as a bounce unless the user does something that you have defined that they should listen to.
This is called a single-page session as it does not send any other information to the Google Analytics script. Or in other words, it does not trigger the analytics script to listen to any of the idle time, scrolling or cursor interaction that the user might have with the page.
The only engagement they listen to automatically is if the user has another page view of another page on your website, within a 30 minute period (the time a session is counted for).
This means that a user can end up on a web page like a long article and spend an hour on that page, and then leave it, yet count as a bounce. Even though they read the full article. Because there was no other action (engagement hit) that you had told it to register.
In fact. All sessions that don’t include at least two engagement hits – actions – such as viewing another page or starting a video (if you have defined that to be an engagement hit/event) are counted as a zero time visit.
It is not as though they were never there at all. They are simply counted as bounces.
What you Should use the Bounce Rate Metric for?
As you might understand from this rather long explanation, it is not unproblematic to use Bounce rate as a measurement of content quality by itself.
You know that a user who reads your full article might have really liked it, they might have even shared the link by copying it and pasting it onto some social media platform, yet, they are discounted as a bounce.
Equally, a person who visits your website where you have a video that goes on auto play and you record every video start as an event, might leave the website instantaneously yet will not be counted as a bounce.
In order to do a proper analysis where you use bounce rate as a metric, you need to either use it in combination with other metrics, OR split your website into segments where content that is likely to generate high bounce rate such as blog posts, are separated from content that shouldn’t generate high bounce rate, such as a product page.
How to Create a more Accurate Adjusted Bounce Rate metric
To help us out, there are things we can do to our Google Analytics tracking that helps us out when trying to improve the bounce rate metric.
One of the things we can do is to install a plugin such as Scroll Depth that helps us track how far a user has scrolled on a page.
Or we can tell Google Tag Manager to register visits that are longer than a certain period of time, or where scroll has been activated, or where some other thing such as clicks in the text have happened more than once.
We will get back to how to use Google Tag Manager in later posts and if you want me to prioritize that, you can comment below or DM me on LinkedIn.
But for now though. This is probably the most well rounded explanation of bounce rate I can give you.