With the introduction of GA4 (Google Analytics 4), there have also been some changes in the measurement of dwell time. The function included in GA4 by default only measures pages >90% view! While Universal Analytics (GA3) was based on sessions, GA4 relies on events. However, Google has announced that it will expand GA4 in the future to focus on "session scope". [toc]
Own success control: Is my content interesting to readers?
Anyone who creates content or has content created often (and should) want to know whether the content has reached the reader or interested them. This is the only way to make adjustments for the future.
Detailed tracking is required for in-depth analyses
The simple content metrics:
Direct views: Your article is accessed directly, e.g. because someone has bookmarked it. This is likely to happen rather rarely on an article basis and is usually of little relevance in the wider context. It also says nothing about the quality or acceptance of the content. Calls on the basis of a search (SEO): A user is interested in the topic "xyz" and searches for it. If the text is SEO-optimized, there is a greater chance that it will be found and viewed more often. This says very little about the popularity, quality or acceptance of the content, at least initially. Only over time will Google recognize whether the text is "valuable". Incidentally, Google classifies this article as valuable. Views based on a promotion: A provider draws the attention of users who are searching for a topic (i.e. "active searchers" e.g. SEA with Google Ads) or those who are obviously interested in certain content due to their other interests and/or visits to special websites (i.e. "latent interested parties" e.g. via online PR or social media ads etc.). This says little about the popularity, quality or acceptance of the content. Views based on text links (referrals): Another author found your article so interesting that they linked to it. Only between 0.0003% and max.(!) 2% of readers will click through to your article. However, the fact that this one author liked your article does not allow any conclusions to be drawn about the popularity, quality or acceptance of your content by the rest of the audience.
The more complex content metrics:
Viewtime on a video: You know the situation - you watch a video on YouTube, for example. And the operator of the YouTube channel can see exactly how many users have viewed at least 25% of the video content, how many 50%, 75% or even 100%. Depending on the "dramaturgy" and the goals of the video creator, even 25% or less than 50% can be a success. In this case, the important content must be placed at the beginning - also a topic for video marketing. Dwell time on a page (one article per page): The length of time spent on an article is often used to determine its success. This is a good approximation. So what is the problem? You know that the reader needs about 10 minutes to read 100% of this one long article. The analysis software (e.g. Google Analytics 4) shows you that the readers stayed an average of 8 minutes. This begs the question: Is the number of readers large enough to determine an average value? If it is, then 8 minutes would be extremely good. Or not? Well, you know that the readers were there for an average of 8 minutes, but whether they were dreamily looking out of the window, talking on the phone, etc., etc., you can't see. - Unfortunately, you can't see that. This is where scroll depth and other tools such as Hotjar or MS Clarity come into play. Scroll depth on a page: A scroll tracking analysis makes sense, especially for longer articles. The absorption of information by content depends, among other things, on its placement on a page. By measuring the scroll depth, you can determine whether individual sections and important information are being read or not. By determining the scroll depth, you can now also see how many readers (in %) have viewed at least 25% or 50% or even 100%, in addition to the dwell time. Admittedly, users could also just scroll and not read your content at all. However, in combination with the dwell time, this gives a good picture of what percentage of the content was actually consumed on average. Instead of 25% increments, other increments could also be recorded, e.g. 10% increments. But exactly: where do you find the value for the scroll depth? Exactly - nowhere by default: But there are great tools and methods for many things. In our case, we use the Google Tag Manager (GTM) and in the Google Tag Manager we create a tag to determine the scrolling behavior. This video shows you how to do this in detail:
PS: Unfortunately, the video, although created by the German Julian Juenemann (already from 2015) is only available in English. You can currently find it at MeasureSchool. For interested parties who find the process too complex, we integrate this feature on his website - contact us. Ratings, social sharing & comments: These three methods are certainly other good indicators of the success of your web content. Unfortunately, some topics are unsuitable for sharing (e.g. "diseases" - who wants to tell everyone that they are interested in psychosis or are researching it - others might get "stupid" ideas). Also, although some topics are extremely interesting for readers, it may be "politically incorrect" to comment on them - keyword "do-gooder" & "asylum crisis". And even if this is not the case - only a few users rate content, even if they (technically) could. Note: Topics such as measuring interactions and conversions are not necessarily content-dependent and will be covered later.