In the last couple of years the team at DataTribe has planned, collaborated, and worked hard to complete the transition to GA4 for a large number of organisations. We’ve also been busy delivering GA4 training. What a couple of months it has been!
GA4 is a different beast and does have a great number of new features, offering deeper insights, enhanced tracking capabilities, and a much better data model. While it’s true that in some ways it compares negatively to Universal Analytics, the fact remains that it is the new reality and we all need to take advantage of those new positives.
OK. So now that you have GA4 tracking in place, is it “job done”? Unlikely. Here are some next steps.
#1 Export historical data from Universal Analytics
Now is a good time to start thinking about what data you want to export from your Universal Analytics property before it is lost forever. Exporting historical data is essential for preserving past performance, identifying trends, and making informed decisions.
While your Universal Analytics property may have stopped processing data (or will do very soon), the historical data is still accessible in your Universal Analytics account. However, it will only be accessible until 1 July 2024. Because the data model of Universal Analytics is different to GA4, the historical data can’t be moved into your GA4 account. Therefore you will need to export whatever data you want to retain.
Unfortunately, you can’t simply export all the raw data into your own database somewhere. The only way to export the data is through the API. This has limitations. For example, you can only export a table with 7 dimensions and 10 metrics at a time. Sampling may also be applied depending on the volume of data involved. Therefore, you’ll need to be selective about what data you want to retain and export it in chunks that avoid sampling. For example, critical reports such as Traffic Sources, Content, eCommerce, etc.
DataTribe has some prebuilt extraction templates that will help you move your data to the location of your choice. Every business and tracking setup is slightly different, but we can adjust our templates for you based on your particular setup.
#2 GA4 training
Training is vital for maximising analytics capabilities, empowering data-driven decisions, and staying ahead in the evolving landscape with product updates happening constantly. Collecting the data is one thing, but knowing how to access the data you need, when you need it, is imperative to being able to extract value from that data.
The GA4 interface is very different from the Universal Analytics interface and many of the dimensions and metrics in GA4 are different too, so it’s not easy to simply start using the GA4 interface without going through a learning process.
DataTribe can provide GA4 training for your organisation. See more information here.
#3 Dashboards and custom reports
Dashboards streamline data visualisation, driving quick and informed decisions. Customizable and intuitive, they empower teams with actionable insights for better business outcomes.
Your organisation probably has downstream dashboards and custom reports built on Universal Analytics data using tools like Looker Studio, Power BI, etc. The data source for many of these will need to be switched from Universal Analytics to GA4 data. Because the data models and tracking methods are different, you will probably also want to make some changes to take advantage of data that was not available in Universal Analytics. For example, conversion counts for traffic sources based on the data-driven attribution method is a new attribution method available in GA4.
In addition, because GA4 has fewer out of the box reports, it’s likely that you will not see some of the reports you previously had in Universal Analytics. These could be built in your dashboard tool, or, for certain scenarios, in your GA4 interface.
#4 Enable BigQuery link
Linking BigQuery provides you with ownership of a database with all the raw data collected by GA4. This provides endless possibilities in terms of being able to produce the reports and insights that you want. However, it is also a good idea to link BigQuery for a couple of other reasons:
- If you are building reports in Looker Studio, Power BI, etc. or otherwise using the GA4 API, this now has quotas as to how much data you can request from the API in any given hour or day. Although recently increased, these quotas are not that generous. Therefore it’s possible that in order for you to have reports that don’t give you quota usage error messages, you may have to build some parts of your reports using BigQuery data.
- Depending on which part of the GA4 interface you are using, the GA4 reporting interface also has limitations in terms of custom reports that you can build, filters that you can apply, and date ranges that you can apply to the reports (believe me, there are a lot of gotchas when trying to build the exact reports you want!).
Even if you don’t think you need the BigQuery data right now, when you link BigQuery it will only collect data from that point onwards. In other words, it won’t include historical data prior to the date of linking. Therefore if you link BigQuery now, it’s an insurance policy that you can rely on later on.
Most organisations will come under BigQuery data storage free tier, so it’s unlikely to cost you anything until you start querying the data. We highly recommend all organisations take out this insurance policy and set up BigQuery linking today.
#5 Import GA4 conversions into Google Ads
Import GA4 conversions into Google Ads for smarter ad targeting and improved ROI. Unleash the power of data synergy for enhanced campaign success.
I’m sure you have done this already, but if your Google Ads activity has been using conversions imported from Universal Analytics, you will need to switch over to using conversions from GA4 data. However, do note that the conversion counting method could be different (goals imported from Universal Analytics would be using the “once-per-session” counting method, whereas by default, conversions from GA4 are “once-per-event”. Therefore, in order for you to be able to get what you need, you may need a guiding hand to ensure you count what you want.
If you haven’t been using conversions in Google Ads imported from Google Analytics, then maybe now is the time to think about it. Using GA4 conversions you can create a sophisticated balance of micro and macro conversions that will help you optimise your ad spend. This is a bit more fiddly to achieve if you are going about it using Google Ads conversion tags.
#6 Build GA4 audiences (and import them into Google Ads)
Leverage GA4 audiences to identify the behaviours of different user segments and better understand how you might adapt your marketing to improve the user experience for different audiences. Those audiences can then be imported in Google Ads campaign targeting to deliver more relevant ads and apply different bidding strategies.
Building an audience in GA4 simply involves grouping visitors based on conditions and sequences of events using the data collected in GA4. The possibilities are endless. For example, all the following are useful audiences that you could create for insights and targeting:
- New visitors
- Returning visitors
- High value customers
- High engagement visitors
- NZ only visitors
- Abandoned cart visitors
- Users that have logged in
- Users that haven’t logged in the last X months
All of these may have a valuable targeting application depending on what you want to achieve. However, unfortunately, there is no one-size-fits all. The most successful strategies will use audiences that are customised around their business and typically this involves creating audiences around the following subjects:
- Your marketing customer segments and personas
- Marketing funnel stage
#7 Optimise your GA4 implementation
You have a GA4 implementation, but is it customised and optimised to the features of that platform, or is it simply a similar setup to what you had in Universal Analytics? If it’s the latter, then it’s possible that an improved setup might deliver more value in terms of the quality and usability of your data and the insights you can extract from it.
Now that you at least have baseline tracking in GA4 and the urgency of the sunsetting of Universal Analytics is over, take the time to use your existing tracking and see if you are able to get the insights you need from it. If not, then this will be either because you need some training (see #2) or the data you are collecting is not meeting all your needs. What questions are going unanswered? Getting someone to review your implementation and sense check it, and giving them the opportunity to suggest valuable improvements may open up possibilities that you weren’t aware of. This could be either because you didn’t know that it was possible to track those things, or because you weren’t aware of specific features that are available in GA4.
Either way, the key message here is that your GA4 setup should reflect what is best for your organisation based on the GA4 platform, not the Universal Analytics platform.
#8 Put in place an analytics governance plan
You have probably invested time and resources into migrating to GA4. However, your website may go through changes, and any one of those could break your tracking or change what is important to measure. Make sure you have the relevant systems in place to ensure that your tracking does not break and is updated when you make a website change. Typically this will mean:
- Having clear ownership of analytics with the organisation. Who is responsible for its maintenance and accuracy? Whoever that person is, be fair to them. Don’t just load it onto their existing workload but create the space for it.
- Before any changes are pushed live, ensure that analytics is a tick-box that someone has checked to confirm that the tracking still works in the update and that you any new important events. Your analytics should be integrated into your everyday web or app development processes. This tick-box should be checked on the staging site before anything is pushed into production, and then re-checked once your changes go live.
- Ensure that any tracking updates you make conform to your naming conventions and tracking methodologies. This is about having your analytics implementation documented so that you can reference it and understand how to keep it tidy and valuable.
- Consider having some kind of monitoring in your processes. There are various options here and each has their pros and cons. However, having some kind of automated monitoring, testing and/or alerting system can be an extremely valuable last line of defence in maintaining the value of your GA4 implementation.
#9 Ensure you are still meeting all your privacy obligations
New Zealand is a bit more relaxed than other countries in that you don’t have to get consent to track visitors from this country. However, you still need to inform your visitors about what data you collect and how you use it in your privacy policy. It’s worth dotting your i’s and crossing your t’s here as your new GA4 tracking may be collecting new data that isn’t covered by your privacy policy.
Additionally, if you track visitors from other countries, then you will likely have obligations towards their privacy. You generally don’t have to worry about this if you are not actively operating in or targeting people in those overseas markets. However, if you are, then a range of regulations will apply to you. We are, of course, mainly talking about GDPR in the EU and CCPA/CPRA in California. In the case of the EU. Courts in a number of different EU countries (Austria, Finland, France, Norway, Sweden, Italy and Denmark) have now ruled that using GA in that country is illegal. There are signs that this might change now that the European Commission has decided that safeguards in the US for personal data protection are adequate. However, at the same time this decision is likely to be challenged in what seems to be a never-ending saga. Hopefully things will get to a point where GA is considered “legal” in the EU, but it’s also worth having a contingency plan.
#10 Pat yourself on the back
Congratulations! You got there. It’s always hard having to work to a deadline and deliver before a cut-off date, so whether you did that a year ago or on the eve of the 30th of June with an MVP setup: Well done! You deserve credit for that.