The End Of Universal Analytics

From July 2023, all Google Analytics Universal Analytics properties will stop collecting and processing data. This means that any reports you have using this data will stop working and you’ll need to transition to Google Analytics version 4 (a.k.a. GA4). In this article, I’m going to explain what are the main differences between the 2 versions and give you tips on how to make the switch from UA to GA4 easier. Let’s dive in.

If you’ve logged in to your Google Analytics recently, you may have seen a message explaining that, from the first of July 2023, all Universal Analytics properties will stop processing your website data and you’ll need to switch over to Google Analytics 4.

Differences Between GA4 and UA

If you’re not an analytics expert and the thought of switching and migrating your existing reports is making you nervous, don’t worry. I’m going to break down the key differences for you and by the end of this video, you’ll hopefully feel ready to make the switch and migrate your existing UA reports to GA4.

So I’ve been working with Google Analytics since 2011 and have seen it incrementally evolve over the years. But GA4 actually feels more like a revolution due to the more fundamental changes that have been made to the tool.

You may have created all of your existing reports, either inside Analytics or externally in a BI tool like Looker Studio, using a Universal Analytics view. And, the bad news is, you’ll need to recreate them using GA4.

This is down to changes in the way that data is now collected by GA4. Which, in turn, means that the metrics and dimensions you use in your UA-based reports will, for the most part, be different. So what is this change in data collection?

Data Collection Changes GA4 vs. UA

Well, to start with, GA4 gets rid of the notion of property-based views.

So, in UA your reports are built using data from views that are created by applying filters to different properties (i.e, websites and apps) within an analytics account. Those are the 3 levels of the hierarchy.

Whereas in GA4, data is collected from what are called streams that are now aggregated at a property level. What this means in practical terms is that you can now combine website and app data into the same GA4 property to then analyse them together.

So that’s the first main change that’s been made in terms of data collection. The next relates to the actual data model.

Data Model Changes GA4 vs. UA

In Universal Analytics, the data model is mainly based around the idea of hits, sessions and pageviews. Whereas in GA4, data is now mainly collected as events. And the Total events metric has been replaced by the event count metric that you use in conjunction with the event name dimension.

Events in UA were things you set up in Google Tag Manager that had a category, label, name and optional value. And if you didn’t know how to set them up, you didn’t collect the data.

Well, the good news is that, in GA4, some of these events that you would have had to set up in Tag Manager are now available “out of the box” as standard and called “Automatically collected events” and some other events you just need to toggle on in the Google Analytics interface to start collecting the data. These are called “Enhanced measurement events”.

To activate them you just need to go to the admin section of the property, select data streams then the relevant stream and then toggle them on and configure the ones you want to start collecting data for.

There are 2 other categories of events, “Recommended events” and “Custom events” that will still need setting up in Google Tag Manager.

Conversions

So, some of these events can be easily also turned into conversions. The notion of conversion events is important because, for example, an Engaged Session in GA4 is one that has at least one conversion event. And engaged sessions, as I’ll explain later, are used to calculate a new important metric.

To turn events into conversions, just go to the configure section in Google Analytics and, under events, you can toggle them on and off individually. Simple.

For conversions in UA, there are 4 different types of goals that you set up to convert when something happens. So this new way of working with events and conversions in GA4 actually makes life easier and gives you simpler tracking capabilities.

Reports

Ok, so now that we’ve talked about how the data collection method and model are different between UA and GA4, let’s start talking about how these changes are going to affect your analytics reports and what you’re going to need to do to migrate them.

First off, let’s go into Google Analytics and compare the report sections. This will give you a better understanding of the different models. You can see that, in UA, there are reports based around 4 main categories. Audience, Acquisition, Behaviour and Conversions. In GA4, you have Acquisition, Engagement, Monetisation and Retention under a global “Life cycle” category. And Demographics and Tech under a User category.

The biggest difference you’ll notice, other than the fact that the majority of reports present in UA have been removed from GA4, is that the Behaviour reports have been replaced by Engagement reports.

Page analysis

Whereas before you had the pageviews metric, you now have the page view and screen view events. In your reports, you can use the views metric which is the number of times either of these 2 events is triggered (page views plus screen views).

One thing to bear in mind though is that the notion of unique page views has disappeared

In GA4 you not only analyse pages by the number of views but also the number of different events that have been triggered on different pages. Things like link clicks, page scrolling etc. So you use the event count metric to measure the number of these events.

Sessions, users & bounce rate

Two of the most commonly used metrics from Universal Analytics are sessions and users and there are changes to these that you need to understand.

Firstly, sessions. Google describes sessions as “a group of user interactions with your website that takes place within a given time frame”.

Whereas in UA a session could contain a combination of multiple page views, events, social interactions and ecommerce transactions, in GA4 a session is started when someone loads a page on a website. Which triggers the session start event.

The duration of a session is calculated between the first and last events triggered by a user. This means that Session duration is now no longer a thing and has been replaced by the User Engagement metric which is a sum of all user time spent on a website.

Bounce rate has also been impacted by the new data model and mainly been replaced by Engagement rate.

Whereas Bounce rate in UA was calculated using the number of single page sessions. Engagement rate uses the number of engaged sessions, calculated as a session that lasts longer than 10 seconds, has a conversion event, or has at least 2 pageviews.

Bounce rate is still a thing but it’s now calculated as the inverse of engagement rate. So 40 percent engagement rate means a 60 percent bounce rate.

When it comes to Users, there are differences here too. As well as Total Users and New Users, GA4 now has a third metric, Active Users. And, you guessed it, events are involved.

Total users is the number of unique users who logged an event, say the session start event. New users are those who triggered the first_open or first_visit event. And an Active user is any user who visits your website and logs an engaged session (like we talked about earlier).

So which should you use? Well whereas UA reports focus on the total users metric, GA4 focuses on Active users. And Google says,

Depending on how frequently your users return to your website, the Total Users metric in UA and the Active Users metric in GA4 may be more or less similar.

So there you go.

Feeling better?

Hopefully, so far, this has all made sense and you now feel more relaxed about the prospect of migrating your existing reports. But, if not, I’ll be making a video (with accompanying article) where I go step by step through how to build a custom GA4 dashboard using Looker Studio very soon.