Aha, the age-old question of why Google’s Universal Analytics (UA) and Google Analytics 4 (GA4) don’t see eye-to-eye. At the end of the day, it’s like trying to compare apples to oranges, or cats to dogs, or SEOs to social media managers (not that we would know anything about that). They may seem similar on the surface, but when you really start digging into the details, you realize they’re fundamentally different. And because your metrics aren’t matching up between UA and GA4, it may lead you to dislike/distrust GA4. But fear not, my newly-deputized data analyst, for we are here to shed some light on the matter and hopefully instill some faith in the future of web analytics along the way.
Really there are two main things to consider here when talking about data discrepancies between UA and GA4 – data modeling and the metrics themselves. Sadly, it does complicate things a bit when both of these are vastly different from platform to platform.
While UA uses session based modeling, GA4 has since switched to the new and improved Event based modeling. Event based modeling allows for almost unlimited customization in terms of event parameters, which in turn allows for businesses to collect data in more specific and meaningful ways.
So, let’s take it metric by metric, piece by piece. How do they vary from platform to platform, and why is your data not matching up?
While UA analytics processes many different types of hits (page view, event, social, ecom, user timing, app/screen view), GA4 is all event based. In UA, events had their own Categories, Action, and Label. This is all gone in GA4 and replaced with Event Parameters. This means that if we’re smart with how we tag events, GA4 can provide far more insight into how a user interacts with our site than UA ever could.
You know when two people use the same word, but mean entirely different things? Yeah, that’s GA4 and UA when it comes to Users, specifically. In UA, users were more closely tied to sessions – really counting how many total people were interacting with the site within a specific time frame. Meanwhile, GA4 counts only active users as “Users”. And active users are those with engaged sessions only, which Google defines as “The number of sessions that lasted 10 seconds or longer, or had 1 or more conversion events or 2 or more page or screen views.” So, completely different metrics from platform to platform. However, if you want to see the UA definition of users, that now comes in on GA4 as “total users.”
Dovetailing quite nicely with Users, we now get to bounce rate. In UA, bounce rate is the lack of interaction after a person lands on your website. Even if they stay and read an entire blog for 18 minutes (we promise we won’t keep you that long), then they leave, that counts as a bounce. In GA4, this is slightly different. A bounce is the lack of an engaged session. So going back to the previous example, this user would not factor into bounce rate since they spent longer than 10 seconds on the page.But in GA4, most of the time you won’t even see bounce rate unless you build custom reports. This is replaced by “engagement rate” which is the inverse of bounce rate. Along with engagement rate comes several other metrics like engaged sessions and engagement time, which leads to a better overall understanding of your users.
According to Google, Page Views should be the most similar from platform to platform, usually within a few percentage points of each other. Especially since, in the case of page views (hit vs event), the Gtag fires on every page and is measured much in the same way. However, due to filters implemented on GA4 and UA, they could still be off slightly from one another. One additional exception is if you’re tracking web and app data, as this is reported in the same data stream on GA4, but needs a completely separate property in UA.
Sessions, also called the session_start event in GA4, are where things start to get a little bit different. One callout here is that there is a session timeout setting that can be changed in each platform. These default to 30 minutes, but could be part of the reason why metrics seem so different.
Additionally, they’re measured very differently. In UA, sessions timeout immediately at midnight (which inherently starts a new session) and, if someone picks up another UTM parameter along the way, it starts a new session. Both of these are solved in GA4. Neither additional UTM parameters nor the clock striking midnight inherently reset sessions. Because of this, you may see much lower session_starts on GA4.
These should be roughly the same according to Google, but can still be off slightly due to filters. They also give themselves some leeway in terms of just how ‘off’ they can be from platform to platform.
These should also be fairly close, though the way they are measured by default is slightly different. While UA will only measure one conversion per session for each goal, GA4 will measure every conversion per session. So, if someone hits the “contact us” button 45 times within a single session – yep, those are all counted as conversions.
In conclusion, my fellow data analyst, while UA will always have a special place in our hearts (after all, it served us well for many years), it’s time to move on to bigger and better things. GA4’s flexible event tracking, customizability, and future proofing make it the clear winner in the analytics game. So, what are you waiting for? Make the switch to GA4 today, and never look back!