Matthew Howman is Marketing Strategy Director at Affinity with over 7 years of cross-channel experience in the digital realm, contributing to the blog on a multitude of topics from analytics to practical marketing advice – here he discusses GA4 in more detail.
The transition date for website owners migrating from Universal Analytics to GA4 is almost upon us; from July 1st we’ll no longer be able to collect data from UA and my early assessment is this will be a painful experience for many marketers and business owners out there.
GA4 has been presented by Google as their future-proofed measurement solution, using event-based data rather than session-based and providing privacy controls for cookieless measurement as we traverse through an online era of extensive cookie banners and data opt-in agreements.
If you’re new to using GA4, you might have taken a cursory look through the platform, or be looking at getting an account set up ahead of July, one burning question will be – is this a step forward from the existing platform?
GA4 – The Good
As reticent as I am to embrace change when the existing platform is so familiar, there are some good features.
- The event-based tracking is extremely detailed. Using the debugview you can monitor the events that GA4 tracks as you navigate through the site. Clicks, scroll, dwell time, entrances, exits and engagements, it does track in excellent detail which provides an ideal dataset for UX analysis.
- GA4 is moving away from highlighting metrics such as Bounce Rate which, arguably, are thoroughly outdated, in favour of engagement-based measurement. Bounce Rate still exists, however, the main reporting now contains engaged sessions and average engagement time by default.
An engaged session is defined as a visit that lasts longer than 10 seconds, or a user visits two or more pages – this is straight away a good quality filter you could use to create a segment of data to hone in on the user journey for your most engaged traffic.
Let’s say you had 10,000 sessions and of those, 8,000 were engaged, if you’re looking to gain a more detailed understanding of which pages were influential in the user journey, or which pages provided a pain point, you are in a much better position to isolate and analyse the pool of 8,000 sessions, as opposed to anyone and everyone who found their way to the site.
GA4 – The Bad
Unfortunately, GA4 has caused a lot of stir and debate amongst the industry as many who use the platform day in, and day out, (myself included) have found it to be an incredibly frustrating platform to use
- Exploration reports
Exploration reports represent the best and worst of GA4’s capabilities, however, once you’ve cracked it and have created a nice, shiny report that has all the data you need, you can’t just easily share it with colleagues or clients. You can share the framework for the report, then separately other users need to duplicate the report to change simple things like the date range.
- Event reporting pinned to sales & CR%
One of the key reporting processes I’ve used over the past few years is to place custom events on a website and use Google Analytics to determine the actions that influence users’ converting. That might be engagement with certain features on the site, buttons, videos, content blocks etc.
In GA4, being event-based measurement, the purchase/conversion event operates independently to other events on the site.
There’s no longer a way to easily break down actions on the site and compare their effect on users converting. It’s possible, everything is arguably possible in GA4, but the platform has turned a five-minute task into an arduous endeavour to decipher the data through an exploration.
- Cross-device tracking
A key benefit professed by Google for using GA4 is a full unified view of device-overlap data and cross-device conversion paths. This can be done through USER-ID tracking, which is to create a user profile when someone logs into the site or via Google Signals, which associates visits with a user when they’re logged into their Google account.
As it currently stands, you can collect all of this data, however, there is no way to report on it. Google Signals won’t collect cross-device tracking as iOS14+ isn’t supported, User-ID tracking does collect the data, but there is no reporting available to understand an important user path, very frustrating.
GA4 – The Ugly
In the standard reports, there are a lot of line graphs and small data tables which don’t hold a lot of use when looked at in isolation. The general reporting structure and UI for GA4 looks and feels like it was designed to look like an in-depth data platform, without proper scrutiny on whether each graph or scorecard of data shown would be useful for the general user.
With many clients, we’ve found the feedback to be that GA4 is confusing and, at times, overwhelming, so whilst marketing teams and business owners get used to the transition, I’ve regularly found us creating dashboards and reports that replicate the previous iteration of analytics as closely as possible.
The hope for a progressive, improved analytics platform lies in the continual updates which are being rolled out across the platform. Once you start to get used to GA4, you will begin to find a level of familiarity. Over the past few months, we’ve seen an introduction of conversion rate metrics, custom channel groupings, improved eCommerce measurement.
There are updates which are gradually improving the platform and hopefully plenty more to come. As a decade-long user of universal analytics, change was always going to be a challenging prospect.
However, as we venture into a period of time for the internet where privacy, tracking and data are under continual and increasing scrutiny, the adjustment for the marketing industry will be further geared towards how we use sampled and aggregated data to make smart decisions.