As you may be aware, Google is shutting down their Universal Analytics tool in favour of Google Analytics 4 in July 2023. After many years of working with UA, many people are understandably confused and maybe even intimidated about the shift to GA4. Here are some implementation tips and lessons we have learned from helping our clients move into the new age of analytics. See this not only as a challenge, but an opportunity to freshen up your in-house analytics skills.
It’s not as easy as you might think
We are not going to lie, it’s not always a walk in the park to switch your tracking to GA4.
Even though the actual implementation part is relatively easy thanks to Google’s handy installation wizard that plugs GA4 onto your existing UA property, there are many changes that can make the transition challenging.
Things such as landing page reports, bounce rates, and the ways sessions are counted are fundamentally different/no longer exist in their current format. This can lead to broken reports and confusion over site performance.
Our recommendation is that you start collecting GA4 data as soon as possible, as you are going to want time to investigate how it collects data and familiarise yourself with what GA4 reports look like.
Tip: in GA4, your bounce rate is the inverse of the reported engagement rate. This is a good example of how GA4 is changing the way in which you track and report on data.
GA4 features are still coming out, some are still missing
Google Analytics 4 is still finding its feet. It is not uncommon to still be finding (as we write this in early 2023) missing connectors to Looker Studio, or parts of GA4 reports that seem unfinished.
While this can be frustrating, there are situations where it is simply a case of waiting it out.
The other challenge with implementation is the lack of documentation, blogs and tutorials.
So far YouTube has been one of the best places to find up-to-date and detailed information about GA4. This is one of the best YouTube channels for delving into GA4 implementation: https://www.youtube.com/@AnalyticsMania
Year-on-year analysis will change
GA4 only holds data for a maximum of 14 months, so you will not be able to compare year-on-year data in the same way in the future.
One way to get around this is to export your data from GA4 into BigQuery using the APIs something that not only offers you a more long-term data repository, but is also a strategic option when it comes to building more complex reports and automation.
Tip: Ensure you set the data collection period to the maximum (14 months) when installing GA4 in the installation wizard. You can go back and change it later, but is an easy one to miss, and is best done at the point of installation.
You’ll need to rebuild your Looker Studio (ex-Data Studio) reports
Even though there are many elements of your Looker reports that will remain the same, a significant part of your old reports will cease to function.
It is not simple the case of switching over the data streams and watching the reports fill up, as some of the data is collected in fundamentally different ways.
Reports on bounce rates, landing pages, and page paths are examples of ones that will not necessarily be straightforward to transfer over.
Tip: Check out some of the free Looker Studio templates to get you going, though many of the ones on offer are still the old UA ones.
You might get hit by Looker Studio API quotas
Recently there has been a lot of reports of people coming up against the GA4/Looker Studio quota, with Looker Studio reports refusing to work.
The quota takes a few things into account, such as report usage and data tokens. Data tokens are influenced by things like filters and report complexity. Usually, quotas refresh quite quickly, often after one hour. Google recommends sharing the report with less people and splitting multiple tables into multiple pages.
It is hard to test for quotas, and some people are opting for the BigQuery export model instead to combat this.
It’s time to switch to Google Tag Manager (really)
GA4 is a step up in terms of analytics tracking, so it is no surprise that it comes with upgrades to how tracking should be implemented.
This is a great time to switch to using Google Tag Manager. Google Tag Manager is an easy way to install scripts of different kinds (not just analytics) to your website. Using GTM means that you can insert multiple tracking codes to your website via one line of code. GTM scripts are managed through the GTM interface, and it is recommended that you use one container for all your analytics tracking.
Why is GTM a good idea?
- It is a more flexible way of adding and removing scripts to your website
- It makes it easier to manage multiple data streams and properties
- GTM makes creating custom tags and triggers a lot easier.
Key differences between UA & GA4
One of the major ways in which the two systems differ in a major way, is how user interactions are measured.
In GA4, there is a shift towards event-based tracking, wherein user interactions are measured on a granular level, broken down into clicks, scrolls etc.
This is actually a vast improvement from the rather simplified session-based tracking in UA, but it can take a while to get your head round what all the events mean. What do all those event counts tell us about user journeys?
There is a lot of great potential in the new GA4 reports: the user flow ones are especially interesting. Now, you can really drill down on individual user journeys.
With time, you will see vast improvements in terms of how granular and detailed your tracking will become, but there is a learning curve.
See this is an analytics opportunity
It is easy to get frustrated with all the changes you are being asked to make but try to see this as an opportunity to maximise the ways in which you are measuring data in your organisation:
- This is a great moment to invest in analytics training
- Re-build reports in conjunction with marketing to get exactly what you want out of them
- Could you even use this as an opportunity to go even deeper with your data analysis and build a custom connector for Looker Studio? This could allow you to enrich analytics reports with other user data
How can we help?
We can help you make the switch, re-build reports, and offer custom training for your organisation. We can also help you investigate your current analytics landscape and make recommendations. Feel free to get in touch with us for a no-obligations chat.