Bounce rates are the most popular bounce rate when it comes to web analytics. Bounce Rates are the quintessential engagement metrics, but what are bounce rates? Well, they measure interest for the web page. They can help you determine if a page is a “good” page or a page that needs improvement. That being said, due to the nature of bounce rates context is very important. Part of that context is understanding how a bounce rate is calculated and the potential pitfalls that accompany it.
Engagement Metrics
This is the first engagement metric mentioned in my Google Analytics 101 series. Previously I talked about several dimensions in Google Analytics. So I thought it’d be a good opportunity to define engagement metrics. Engagement metrics are metrics that help you understand user interest, or engagement, in the page they are on. The three main engagement metrics are bounce rates, avg. session duration, and pages per session. All three reside in the acquisition report in Google Analytics next to the traffic metrics of sessions, users, and new users.
Generally speaking each engagement metric captures something slightly different about user behavior. So that when looking at all three you can form a picture on how the average user uses your site. Look for a post about that in the future. In the mean time here’s a general rule of thumb:
You want Bounce Rates to be low, and you want Avg. Sessions Duration and Pages per Session to be high.
Bounce Rate Calculation
Here’s the bounce rate calculation:
Bounce Rate = ( Total Bounces ) / ( Total Sessions )
So what’s a bounce? A bounce is a single hit session on a website. You’ll often see a bounce defined as a single page visit session. That’s not quite right. If a page has events on it and those events are interacted with, it is not a single hit session. And therefore, not a bounce.
So for example:
A user lands on a web page, reads the first paragraph then closes the browser = Bounce
A user lands on a web page, clicks the play button on a video and watches it for 30 seconds = Not a Bounce
A user lands on a web page, scrolls all the way to the bottom and then goes to a different URL = Bounce
A user lands on a web page, reads a paragraph then goes to another page on the same site = Not a Bounce
It all depends on hits. To learn more about hits I recommend the Google Analytics documentation about hits.
There are two ways to analyze a bounce rate fro the page perspective or from the traffic perspective.
Interpreting Bounce Rates – Page
What’s a good Bounce Rate? The immortal question. and the Answer is it kind of depends.
Your bounce rate depends on what you want people to do on that web page. If you have a lot of reasons for users to enage you would expect a lower bounce rate. If you had a page with just a pargraph or tutorials you would expect a higher bounce rate.
For example this website has a slightly higher bounce rate. I publish tutorials that are often not related. So I would expect users to come read the tutorial, and then leave the site. And that’s not really a knock on anything just the nature of how web pages are used. So keep that in mind when analyzing a page’s bounce rate.
Here is a link to some bounce rate benchmarks. These can help while you build up your intuition.
General Rule of thumb:
40-85% is normal
>85% is not good
<40% is good
These are all context specific.
Interpreting Bounce Rates – Traffic
Bounce rates analyze the quality of traffic from paid media. Maybe you want to compare the types of traffic two different ads are bringing in. Using bounce rates would be one way to gauge the quality of that traffic.
Analyzing based on traffic is easy. The higher the bounce rate the worse it is. that being said tactics themselves differ. Paid search traffic is almost always more engaged than paid social which is more engaged than display. bounce rates generally align with the position in the funnel of the tactic.
If you want to learn how to get up and running with Google Analytics quickly, I recommend my free Google Analytics Quick Start Guide.