Growing Pains & Gains: UA vs. GA4
It's GA4 time.
Google announced in 2022 that Universal Analytics (UA) will stop processing new data in July of this year to make room for GA4. This has been a huge change for most in the industry, meaning it’ll have its growing pains. Since most Analytics users are moving to GA4, we thought it could be helpful to compare the two platforms.
First, let’s take a closer look at the evolution of the Google Analytics platform.
How has Google Analytics changed over time?
In previous Google Analytics updates, the way tracking was implemented changed, the interface changed a little bit, and new reports were added gradually. The rollout of GA4 is much heftier than that. It’s not just the latest update of this platform, but a whole new way of thinking about, measuring, and ensuring user privacy of marketing data.
There are major differences in underlying data structures between UA and GA4, which impacts the metrics we’re able to get out. Essentially, this means that the information between UA and GA4 are not 1:1. We’ll be losing access to some data and features, gaining access to others, and still more features fall into the messy middle where data is still available, but will look a bit different.
What are we losing?
We’re forced to say goodbye to a variety of default reports, familiar naming conventions, and some access to certain data segmentation and filtering abilities. A lot is changing with GA4, but these are the things affecting our clients most often.
UA was released to the public in 2013, becoming the industry standard for tracking users across devices. It has close to 100 default reports whereas GA4’s out-of-the-box solution comes with about 20 default reports. Now, for any given Analytics property, many of those 100 reports in UA might not see much use. The ones we frequent in UA are not available automatically in GA4. And, unfortunately, the ones we do get in GA4 are less manipulatable.
We’re losing some naming conventions we’re all used to:
- “Bounce rate” still exists, but much of GA4 speaks in terms of “engagement rate”, which is essentially the inverse of the former term. In addition to engagement rate, you’ll also hear us talking about new GA4 metrics, such as “average engagement time per session” and “engaged sessions.”
- “Default channel grouping” is now something we have to choose between. It’s either “session default channel grouping” or “user default channel grouping”—not both.
- “Events” is another familiar term we know from UA as custom events that we use to set up and track specific actions, including a file download, a form submission, or a button click. In GA4, that goes out the window; events are literally everything! An event on this platform is captured for every session start, page view, click, video view, download, scroll, and more. (This means looking at total events is an entirely different metric than it once was.)
Data segmentation access
There is also limited access to data segmentation and filtering in reports or in Looker Studio (formerly Data Studio). In order to conduct detailed data segmentation, you must either create segments in individual exploration reports within GA4 or you have to export and query your data with BigQuery.
Okay, so what are we gaining?
I know what you’re thinking: That seems like we’re taking a lot of Ls. Don’t panic—we’ve also gained a lot with this platform. Here’s what we’re getting with GA4 compared to UA: enhanced measurement events, more reporting customization, “explorations” (more on that in a minute), attribution model options, and a greater ability to track custom events.
Enhanced measurement events
Think of these like out-of-the-box events that GA4 can track without needing custom setup. In UA, tracking these kinds of events usually had to be configured via Google Tag Manager. Now, they’re available with the click of a toggle! These events include video engagements, outbound link clicks, file downloads, and more.
While UA had the ability to create custom reports, they were fairly clunky and rarely utilized. Our go-to for custom reporting is Looker Studio (now and in the future), but it’s worth noting that GA4 brings custom reporting to the forefront of their platform… so much so that it feels like it demands you use it. With the ability to define your dimensions and metrics as well as apply basic filters, custom reports give you easy access to the data you want to check often.
Explorations (aka, detailed reports)
With explorations, you can look at just about any metric by any dimension, then filter and chop it up a number of ways. We could do this in UA, but it would always be a one-time investigation and then the reports wouldn’t “stick.” With GA4 explorations, we can dig into the data to solve a specific problem, and then the report remains available for future use and manipulation. Plus, as mentioned before, explorations are the only place that complex segmentation is offered in GA4.
Attribution model options
In UA, the only attribution model available for standard reports was last-touch. With GA4, we have choices! Depending on our client needs, our own strategic recommendations, and usually lots of philosophical debates, we’re able to set different attribution models, including GA4’s new “Data-Driven” model.
Greater ability to track custom events and tie more parameters to them
GA4 gives us more flexibility in tracking event data and the parameters collected with them. Using GA4 and Tag Manager, we can track custom events from a specific button click to a form submission, and then we can see a handful of both standard and any custom-created parameters tied to each event. We could track events in UA, too, so what’s different?
Let’s take a form submission for example. In UA, we were limited to essentially 3 event parameters, one being the event “name.” For a form fill, we’d have the event category as “Form Submission,” the action would usually be the page the event happened on and then the label could be the page that the user came from before that. Not very intuitive, is it?
Now, with GA4, we can have an event called form_submission, and then we can see or set up the following parameters to pass data along with the event: page location, page referrer, form ID, and form name. We can then also build sub-events based on rules; for example, if form ID = 215 then create a new event called demo_form_submission since that’s the ID of our (fictional) demo form, and mark that event as a conversion.
At the end of the day, GA4 is a totally different platform for the industry to contend with. We’ve been familiarizing ourselves with it over the last year and transitioning our clients steadily while helping them understand the implications and opportunities this change brings.
This is how Search & Digital Media Manager, Jessie Sublette, described her experience with GA4 from when we first started to now:
Sure, we’re sad to see some of UA’s features go away for now, but we’re making the most of what GA4 has to offer. We’re also drawing on other tools, such as Looker Studio and Funnel.io, to give our clients the best insight into their web performance as possible.