Bounce, exits, and average visit time
One thing to know when analyzing the average time spent on your website or specific pages is that time spent on exit pages is not included in the analytics. All exit pages, including bounces, have 0 seconds visit duration time. However, it is possible to measure the time more accurately, and you can see how to do that below.
Before we get to how to measure more accurately, let’s look at how 0 seconds on exit pages affects your average time spent on pages.
How the average time is calculated
As mentioned, the time spent on all exit pages is by default 0 seconds. It doesn’t matter how long the visitors actually stay, the analytics still count exits as 0 seconds. That means that bounces also stay on your website for 0 seconds, according to the analytics.
Say the average time on a page is 1 minute (including exits) and you have 50% exits (including bounce). The analytics is going to tell you that the average time on that page is 30 seconds. If you instead have 75% exits, the average time is going to be just 15 seconds.
This means that you need to take the exits into consideration when evaluating the average time on each page, and the website in general.
Filter out bounce
One way of compensating for exits counting as 0 seconds is to create a new segment to filter out the bounces. However, all exists cannot be excluded. Since the definition of a bounce is that they make no further actions after entering the website, you can filter out bounces by creating a filter for number of interactions being at least 2.
With this segment, you will see the average time for all visits except bounces, but as mentioned, this time will still include other exists.
Manually correct the average time
You can also manually correct the average time per page. Let’s use this as our example:
The average time is 44 seconds and the exit rate is 53%. That means that we only have the time for 47% of the visits. To get the average for those 47%, first change the time to decimals by dividing 44 over 60, which equals 0.73 minutes (only do this for the seconds. 1:45 becomes 1.75). Then, take that times 100, which equals 73.3 minutes. Then, divide that by 47, which equals 1.56 minutes. Lastly, take 0.56 times 60, which equals 34 seconds. The average time spent on the page for the 47% who didn’t leave was 1 minute and 34 seconds (approximately).
Here’s the full formula, once you have changed the time to decimals.
([average time]*100)/(100 – [exit rate])
Change the time format back if you wish to.
How to accurately measure time on website
If you want to include time spent on exit pages in your analytics, you need to add the following to your JavaScript code:
extellio_actions.push(['enableHeartBeatTimer']);
With this bit of code, all time spent on the website (as long as the tab is active and in focus) will be included in your analytics. The heartbeat request is executed when visitors:
- switch to another browser tab after the current tab was active for at least 15 seconds*,
- navigate to another page within the same tab, or
- close the tab.
* If you want a different minimal visit duration time before executing the heartbeat timer, use this code instead:
extellio_actions.push(['enableHeartBeatTimer', 30]);
