Scroll Tracking Analysis using Google Analytics

Scroll tracking enables website operators to measure user scroll behavior. Scroll tracking shows how users act on a website and how they deal with its content. Data such as bounce rate, average time on page, or pages per session reflect general user behavior on a website, but they don't give an insight into the user behavior on a specific page.

Why is Scroll Tracking Important for Website Analysis?

Measuring and analyzing the scroll depth of a website’s visitors has many benefits. It helps understand how users behave on a page as well as how they interact with the website.

Figure 1: Scroll tracking analysis

Example: A user lands on a website from the Google search results and stays there for three minutes without navigating to another page on this website. In this case, Google Analytics will record a 100% bounce rate. If the user navigates to another page, the time on page of 3 minutes will also be recorded. From the point of view of the website operator, there is no information on whether or not the user actually read the entire contents of that particular page, and it is therefore difficult to recognize whether the user was satisfied with the information provided on the website.

This is why scroll tracking can be very helpful, particularly for long articles. The consumption of information is highly dependent on its placement on the website. Measuring the scroll depth makes it possible to recognize which sections the user read, and you can therefore see if the user missed any important information.

Figure 2: Scroll tracking events in Google Analytics

With this data, you can carry out a funnel analysis to distinguish the actual bounce rate from visits by users who found your website interesting. Scroll tracking in Google Analytics is done based on the scroll depth and the respective event. One event per 25% or per 10% scroll depth is recommended.

When displaying the events in this way, it is important to differentiate between the unique and total events. The unique events show the number of individual events per category, action, or label at a session level. The total events show all the results from the entire period.

Figure 3: Unique events in the scroll behavior

This example shows that 401 users scrolled through less than half of the entire scroll depth. 222 users (~25%) left even before they had seen 30% of the website.

Set up scroll tracking

A script on how to integrate scroll tracking using Google Analytics is available on github.com. When setting up scroll tracking, it is important to pre-specify the pages that should be taken into account. Ppages on which scrolling is not necessary due to layout or amount of content can be excluded. It is also recommended that you set a minimum height for which the script should be activated.

When tracking the scrolling behavior, you can generally decide whether you want to calculate the page height or scroll depth in percent, or whether you want to include certain elements, such as the page footer, as events. The WP Scroll Depth plugin is also available for integration of scroll tracking in WordPress websites.

Track the actual bounce rate through the scroll behavior

In Google Analytics, the bounce rate shows the percentage of visitors who only viewed a single page, i.e. visitors who left the website without any further interaction. A high bounce rate would therefore indicate that users have little interest in the website's content. However, this isn't always the case - a user may find the content of a page interesting but still do not want to view another page (this could for example often be the case for magazine articles). In this case, a high bounce rate wouldn't necessarily be a bad thing.

If "firing-off" an event while scrolling through a website should not have an effect on the bounce rate, this can be avoided with parameters. Events that are triggered in Analytics through scroll tracking are passive events. For this reason, you should activate non-interaction events to prevent such events from having an effect on the bounce rate.

Measure the “True Reader Bounce Rate” based on time

It is advisable to measure the so-called "true reader bounce rate". Here, a user leaving the page will add to the bounce rate they spend a specified amount of time on the page, even if they only view a single page. Therefore, if the specified time is three minutes, and the user stays on the homepage for longer than three minutes before leaving the website, it won't be counted as a bounce.

Figure 4: Measure the actual bounce rate based on the average time on page

Here, in order to optimize the standard tracking in Google Analytics, you simply have to add the setTimeout line to your existing universal tracking code. You should then define the amount of time after which the "No Bounce" event should be recorded. In the above example, a user’s departure is no longer considered a bounce if they have been on a page for more than 45 seconds. Combined with the measurement of the scroll depth, this data provides you with credible information on the actual user behavior on your website, and allows you to carry out more targeted analyses in Google Analytics.

Analyze the User Engagement Using Micro Conversions

If you want to measures how many visitors on your website pay attention to your website content, and how they read the text provided, you should define goals. Configuring target actions for your website enables you to evaluate the user engagement as a micro-conversion in Google Analytics.

Figure 5: Specify scroll conversions in Google Analytics

In the above example, events triggered when scrolling are defined as goals.
You can specify the appropriate scroll depth in your Google Analytics account. To do this, go to Admin > View > Goals:

1. Goal setup > user-defined

2. Goal description > event

3. Goal details > define the category, action, and label of the event

Every session where the user has at least scrolled through half the page is recorded as a conversion. The corresponding conversion rate therefore shows you the actual number of "engaged users".

Conclusion

For a successful website and great website content, it is important to understand user behavior and scroll trolling is an essential way of achieving this user analysis. Data from the scroll tracking analysis should help you identify content that visitors find interesting, providing you with useful tips for where you should place important elements such as calls-to-actions. The data will help you improve your user experience, and thus your overall website performance.