Understanding the difference between Google's Universal Analytics and GA4
How to compare metrics between the two analytics platforms
As of the summer of 2023, Google’s Universal Analytics (UA) is no longer collecting data and has been replaced by Google Analytics 4 (GA4). Pugpig clients are automatically set up to collect data via GA4, but there are some significant differences between the two platforms that it's important to understand.
This article will explain some of those differences, focusing on how you should map your metrics from Universal Analytics to GA4.
If you would like to discuss in more detail or have any questions or feedback please get in touch with us at Pugpig Consulting at firstname.lastname@example.org.
So, what's changed?
The key points:
- GA4 combines both app and web data in the same property, whereas in Universal Analytics they were kept separate.
- It does this with a data model built on events and parameters whereas UA relied on page views and sessions.
- This means that unlike in UA where there were many different hit types - such as page views, transactions and social interactions - every interaction in GA4 is captured as an event.
- It also results in GA4 sessions working slightly differently. They are no longer reset when a session goes across midnight or from new campaign (UTM) parameters.
- GA4 has also increased it's privacy settings. UA allowed you to set user-specific data retention to 14, 26, 38, 50 months or to never automatically expire. In GA4 there are only two choices: 2 or 14 months.
In GA4 Users is now Total Users. It is effectively the same metric as UA but the definition is slightly different: It has changed to the number of unique users who logged an event, rather than a pageview.
New Users continues to mean a user who has interacted with the site or app for the first time.
Active Users is a new dimension that didn’t appear in UA. An Active User is someone who has an engaged session or has a first visit or first open or an engagement event. Whenever GA4 refers to Users in any of the summary or overview charts it is referring to Active Users.
Since pageviews can now be combined across web and app, UA’s Pageview metric is called Views in GA4.
The Unique Pageview metric no longer exists. However, it is possible to to view the number of Users for each page, which is similar.
In UA you would define a Goal to indicate an conversion but in GA4 you should mark an event as an Conversion in the Events UI. GA4 will normally count every instance of a conversion event, even if it happens several times during one session. To make it only count once per session, as was the case with UA, you need to specify this in the counting method setting.
Bounce rate and engagement rate
Bounce rate has been quite significantly changed as Google has made a decision to focus on a more “positive” metric: Engagement rate. This is measured as the number of engaged sessions as a percentage of overall sessions. An engaged session is defined as a session that lasts 10 seconds or longer, or has 2 or more views or has 1 or more conversion events.
Bounce rate still exists but it is the inverse of this metric. It won't exactly match UA, as the calculation is different. IN UA it was calculated as the percentage of single-page sessions in which there was no interaction with the page.
Event category, action and label
In Universal Analytics, category, action and label was sent through with the event and made available automatically. This allowed for you to group together different events and count them.
There is no such functionality within GA4. Instead, to add detail to events, values can be passed through as parameters. Google use the example of a sign-up. You might have an event name of sign_up with parameters page_location, product, form_id, etc. The same event name could (and should) be used on every sign up button across the site (whereas in UA, you would want to use unique event naming for each button).
Although at first the changes between the two platforms are significant, once you've adapted to the new data model then GA4 has the potential to be much more powerful than Universal Analytics. At Pugpig Consulting we have experience with assessing analytics requirements, designing custom events and building dashboards. If you'd like some help, please get in touch with Pugpig Consulting at email@example.com.