Metrics Matter: Understanding differences between Universal Analytics & GA4

Google Analytics 4 brings not just a new tracking infrastructure, and a new interface. It also introduces a couple of new metrics and defines existing ones differently. It’s critical to understand the key differences in particular as BAU reporting will see dips and spikes. 

Looking at the new definitions of old metrics in GA4 

As we said, there’s a change to the way some everyday metrics are defined in GA4, compared to Universal Analytics (known as UA or GA3). Some of the changes are quite subtle and won’t massively shift the needle on metrics. Others fundamentally change what is meant by the metric. It’s like the Lilt scenario, but in web analytics form. 

Google has a great help page here , but I’ve outlined the ones that it’s worth knowing about below: 

• Sessions is defined only a little differently - you're more likely to see a small difference in your pre- and post- GA4 charts. The actual definition of what a session is (a user engaging with your site) hasn’t changed and some of the key ways a new session starts, will continue to exist. For example, if a user has 30 mins of inactivity, returning to the site will cause a new session. Likewise if they timeout on the site, as soon as they do something else – view another page, trigger an event – it will start a new session. However the two buig differences are midnight. No longer causes a new session in GA4, as it did in GA3 – which was always odd. And the presence of new campaign information – again, this used to trigger a new session in GA3, but won’t do in GA4.
• Users is a bit more of a whopper. In UA, the Users metric referred to simply the deduplicated number of users that Google has seen. In GA4, they throw you a real doozy because although the metric is labelled as "Users" it actually refers to "Active Users". This metric counts users who come to the site / app for the first time, or who trigger an engagement (in milliseconds) event or who were “engaged” (i.e. spent more than 10 secs on site or viewed 2 or more pages). If you want to compare or plot a trend for Users as it was in Universal Analytics and as it is in GA4, you need to use the metric “Total Users”
• Likewise, bounce rate is defined differently too. In UA, a bounced session was any session where only one event was fired. In GA4, a bounced session is a session that was not an engaged session. Critically an engaged session is, a user who has been on site for 10 seconds or more, triggered at least one conversion event or at least two page views. Depending on your UA set-up, this could see a real shift in your data. 

Introducing the new metrics in GA4 

There’s reasonably few new metrics in GA4 that you’d write home about. The most fundament are a set of metrics around Events, as this is the new cornerstone of tracking. So you have: 

• Event count – the number of times a particular event was triggered
• Event count per user – you can probably guess that one….
• Event per session – likewise, although the inconsistency in naming is frustrating!
• Event value – a sum of any value parameters sent with events, for example time (for video) or distance (scrolled) 

The most excited one are those I’ve written about before and can be found here . Predictive Metrics! These include: 

• Churn probability – refers to the likelihood of a user who’s been to the site in the past 7 days, NOT doing so in the next 7
• In-app purchase probability – refers specifically to users that have been active in the last 28 days, making a purchase in the next 7
• Purchase probability – same, except for website
• Revenue probability – another simple one except, this is predicted revenue in the next 28 days, from users that have been active in the past 28 days 

Google also dips into the world of product analytics and has started to include metrics such as: 

• ARPU (Active Revenue Per User)
• ARPPU (Active Revenue Per Paying User) 

Measure what matters 

Never let it be said that Google don't give with one hand and take with the other. They keep you on your toes with changes to metrics like the above...but they provide a handy Help article to help you navigate it. 

If you’ve got any issues with your GA4 data, navigating the interface or building dashboards that work, please do give us a shout.