(and how to prevent them — find out more here)
It is 2022 and the summer is finally approaching. So is the day that Universal Analytics is sunsetting. This is why more and more analytics users are migrating from Universal to GA4.
I have been following several groups where GA4 enthusiastics, and less enthusiastics, gather. Turns out that not all of them are positive of their migration experiences and questions remain unanswered. Sounds familiar?
This post summarizes the biggest challenges currently faced by analytics users migrating from universal to GA4. And of course, I will give you some solutions as well.
- Losing data
The sunset will mean for those who did not export their historical data and saved it, the data will be lost. There are ways to prevent this: Export direct CSV files from universal analytics and store your data somewhere where it’s safe. Datastudio can help you to visualize this data, and some professionals have made it easier for you by creating templates. Find one example here.
You can also transfer both your UA and GA4 data into BigQuery and then manipulate it there. You can follow the steps as described here.
*Since the data structure is different, we do not recommend comparing GA4 data to Universal data. Hence is why it is important to start to collect current metrics in GA4, alongside your traditional setup.
2. Data discrepancies
So, when comparing 3 and 4, it’s like comparing apples and pears. This means conversions can be higher in GA4 or the other way around. This causes frustration, because which data can you trust? Acceptance starts with understanding. An important reason for data discrepancies to occur is the different tracking model, but also modeling capabilities that are available in GA4, and not in GA. Google predicts whether a conversion would have happened based on multiple signals when cookies have not been accepted, and counts the conversion if the likelihood is high. Another reason is that data is processed in 24/48 hours in GA4, or sometimes a little slower. The reason for the data delay is not given, but often a solution for the problem if you don’t see your expected conversions straight away. Patience is key, if you have the time.
3. Not seeing any conversions
Event measurement is automatically implemented when you install the GA4 tag on your webpage, and more events are measured after enabling enhanced measurement. However, one thing that does not go automatically, is the measurement of conversions.
You will first need to toggle the ‘mark as conversion’ button in the events table.
A simple step, but oh so crucial. You just need to know about it! Find out how to do it here.
4. No more views
If you were counting on views as a metric to measure the success of your SEO and SEA, you will get disappointed by GA4. This metric is simply not available anymore. Instead, users and engagement metrics of users are the new source of truth.
Until 2023 you can continue to measure views in Universal Analytics, but after that, it is time to evaluate performance differently.
5. Reporting differences
Comparing metrics, viewing real-time performance or a funnel flow report: things you were pretty used to in Universal, suddenly became less straightforward. It is not that these options are not available anymore, they just exist in different locations. View your debug mode reports to see what is happening in real time, build your own funnel flow report in the explore menu, and build segments in order to evaluate subsets of your data side by side. This will help you to understand your data in the way you were used to before.
And let’s also hope that Google’s product teams are working further on the development of these features so that the usage will be much more intuitive.
6. Setting up e-commerce tracking
It is true that currently, e-commerce tracking is still in an early phase in GA4, and the implementation requires a bit more time than you hoped. The naming of e-commerce events and parameters have changed for Google Analytics 4 properties. That’s why Google recommends leaving your UA implementation unchanged and creating duplicate events for your GA4 property. Use the new event names and parameters that are required for Google Analytics properties. Setting up the parameters in correct format are required to start seeing your e-commerce data, otherwise they will just show up as custom events. This is all the info you need in order to set up your e-commerce tracking correctly (reach out if you need help)
7. No more bounce rate
GA4 does not report on how many of your website visitors did not interact with any other items on your webpage, and left the page without engagement. Instead, you have to rely on different metrics such as: Engaged Sessions, Engagement rate and Average engagement time. It provides almost the same information, but framed differently.
In short, some work needs to be done before you can rely on your GA4 reports and get insights that can be actionable. But once you get a hold of it, the product offers some great benefits.
Benefits
- For example, GA4 has excellent automated insights that help you catch trends in your data instantly. A sudden drop in your conversions? Huge spike in sales? Decrease in engagement? GA4 will automatically and instantly show you the information, so that you can act upon the insights straight away.
- GA4 also offers the possibility to find your highest value customers with lifetime value reports and automated audience lists based on predictive metrics (usage probability, purchase probability and churn probability), which you can select in Google Ads to target twitch the highest conversion likelihood.
- On top of this, GA4 comes with a free connection to Big Query, so you can access the raw GA4 data and manipulate it here. This helps in more precise and multilevel data analysis of your users.
This, and the fact that the traditional Google Analytics will not be there for you anymore in the near future to make sense of your customer behavior, are reasons enough to migrate now. So let’s focus on solving the challenges, and benefitting from improvements.
Need help with the migration? Don’t hesitate to reach out. You can also sign up for my 10-week Performance Analytics course. Find out more here:
https://performance-marketing-analytics.getlearnworlds.com/
