‘Head Start’ List for Google Analytics 4 (GA4)

To excel as an online ad specialist (which has been my goal for years), it’s crucial to understand how results get tracked and measured. The world of online tracking is a complicated one that can never be completely mastered. But in 2021, I decided to up my game by studying the two most significant changes/developments in the online tracking world:

• Apple’s App Tracking Transparency (ATT) policies
• Google Analytics 4 (GA4)

The more I learned about GA4, the more I found it interesting and encouraging. Hopefully, this list can give you a head start if you haven’t delved in deeply yet.

1. GA4 is not a new and improved version of Google Analytics 3 (a.k.a. Universal Analytics). It is a complete, ground-up rebuild of the platform.
2. GA4’s technology is derived from Firebase, a company and app development framework that Google acquired in 2014. The Firebase framework includes analytics for tracking interactions within apps.
3. GA4 endeavors to show the overall user journey through an entire acquisition funnel – app and web.
4. In GA4, the same property can gather data from apps and the web. (In previous versions, there was no integrated property type.)
5. GA4 employs something called the “event-driven data model,” which counts every interaction as an “event.” This is better because of the different ways people use apps. (For example, people often use fitness apps to track their physical activity while viewing other apps on their screen. Metrics like bounce rate and pageviews aren’t very relevant in this scenario for measuring engagement.)
6. GA4 focuses on users and events, whereas previous versions focused more on sessions and pageviews.
7. GA4 gathers three types of data: events, event parameters, and user properties.
8. In GA4, properties have Data Streams, which are the data collection methods. The types of data streams are: IOS, Android, and Web.
9. In GA4, every “hit” is an event. No distinction is made between hit types.
10. If you want to define a particular event as a conversion action, you can simply toggle it on to do so.
11. In Universal Analytics (UA), you would create “views” to see data in a customized way. (For example, with bot traffic filtered out, people on Samsung phones, or traffic from Portugal.) In GA4, there are no “views.” If you want to see data in a customized, segmented way, you create an “audience.”
12. In GA4, known bots are filtered out automatically.
13. There is no bounce rate metric in GA4. But there is an “engagement rate” metric. An engagement is counted if a visit lasts more than ten seconds, OR involves more than one page or screen view, OR includes a conversion.
14. Configuring a custom event for a specific pageview (such as a “thank you” page) involves using the page_view event and making it conditional based on the URL or page title. (A page_view event contains the parameters page_location, page_referrer, and page_title.)
15. GA4 employs the Global Site Tag, which allows changes to be made in the user interface rather than in the code (such as turning on Enhanced Measurement tracking).
16. Enhanced Measurement tracking gives the ability to track interactions such as outbound clicks, scrolls, file downloads, video engagement, and site searches.
17. GA4 is built with sophisticated modeling/machine-learning technology, which will work in a world without browser cookies.
18. Google Signals is a built-in collection system that gathers data from users who are signed into their Google accounts and have Ads Personalization enabled. It gathers data such as: location, search history, YouTube history, and data from partner sites. It is used to provide aggregated and anonymized insights into cross-device behaviors. (The Ads Personalization setting can be changed for users by going to https://myaccount.google.com/.)
19. Google Signals data collection can be toggled on or off by country in the GA4 data collection settings. For example, an advertiser might turn it off for European Union counties where the GDPR applies.
20. Google Signals can now be used to create audiences rather than relying on first-party cookies.