There is a lot of confusion about the meaning of bounce rate in Google Analytics. I’ve heard a few theories – from the correct to the ludicrous. The biggest learning curve when using Google Analytics is getting to grips with the terminology so I want to help you sift through the misinformation about bounce rate before it gets out of hand.
So what is bounce rate?
According to Google, bounce rate is “the percentage of single-page sessions (i.e. sessions in which the person left your site from the entrance page without interacting with the page)”.
So, a bounce is where someone has landed on a page and left without navigating to another page. The bounce rate is the number of bounces divided by the number of entrances (when someone landed on the page and did go on to visit another page on the site or interacted with something on your page which can be registered in Google Analytics).
How long have they stayed on the page?
This is one of the most common misconceptions I’ve heard about bounce rate. I hear ‘bounce rate is when someone lands on your website and immediately leaves’ quite a lot.
This isn’t technically true. There’s no data available for how long a user spends on a landing page if they haven’t travelled to another page on the website or haven’t triggered any events. In fact, Google can’t calculate the time on the last page of any visit if the user hasn’t triggered any events.
Google Analytics works out time on a landing page by taking the start time of page one and the time the user moved to page two and page three and so on and subtracts them from each other to get the individual time on page.
For example,
Page one – start: 14:00
Page two – start: 14:03
Page three – start: 14:07
Leave site: ???
Google Analytics can deduce from this information that the user spent 3 minutes on page one and four minutes on page two. As the user left the site after page three and didn’t trigger any events, they don’t have a time to reference to find out how long they spent on that last page.
With a bounce, the user doesn’t move on to another page so Google Analytics has no reference with which to figure out the time on page.
Can events give more accurate time on page?
You will have noticed my mention of events above. If a user bounces (leaves on the page they entered the website on without going to any other pages) but triggers an event while on that page, Google Analytics will receive the time of the interaction and use that to estimate the time on page.
To trigger events, you will have to set up events in Google Analytics. An example of an event could be a video play or a time on site of 10 minutes.
Using the example above,
Page one – start: 14:00
Page two – start: 14:03
Page three – start: 14:07
Triggers ‘time on site >10 minutes’ event
Leave site: ???
In this scenario, Google Analytics can deduce that the user spent at least 3 minutes on the final page and will factor this into the average time on page for the final page.
Is increased bounce rate bad?
We’ve already clarified that unless you have events or other available interactions on an entrance page, Google Analytics is unable to deduce the amount of time a user has spent on an entrance page if they left before navigating to a second page.
Whether a high bounce rate is bad (or, indeed, good) completely depends on your aims for the page in question. If it is a simple question-answering page that you expect users to access from search engines to answer a quick question, for example, then you might be relieved to find out that someone Googled their question, found your page, found the answer quickly and went about their day better informed. And you might be disappointed to hear that someone in a similar situation Googled a question, found your page, read it and didn’t find the answer they were looking for, so continued to different pages on your website looking for the simple answer they were after, eventually giving up.
The first situation would qualify as a bounce and the second as a session with 3 or 4 pageviews.
Many people would say a bounce is ‘bad’ and a session with a few pageviews is ‘good’. But the example above seems to strongly contradict that.
The ‘goodness’ or ‘badness’ of a bounce rate is entirely dependent on the aims and purpose of the page. For more on bounce rate, read more about how to tell whether a high bounce rate is good or bad.
