Google Analytics 4 (or GA4) is a completely new version of Google Analytics that uses Firebase Analytics on the backend. In this post I will explain the implications of that, but the overall benefit of GA4 is that it enables marketers to measure users interactions within mobile and single page apps, and it attempts to correct data that has been corrupted by cookie restrictions and user consent choices in a privacy-safe way.

Migrated from ken-williams.com
Migrated to new CMS with minor updates.
Updated this post to explain how conversion and behavioral modeling differentiate GA4 from other Analytics tools.

Where Does the 4 Come From?

If you’ve been following Google Analytics for a while, you’ll remember that there have been two major overhauls of the platform since it was acquired:

2008: Transition from Urchin to classic Google Analytics (ga.js)
2013: Transition from classic Google Analytics to Universal Analytics (analytics.js)

Now, we are in the middle of a third major overhaul from Universal Analytics to Google Analytics 4, which makes this the 4th version.

Google Analytics 4 has been in development since the initial release of "Google Analytics for Firebase" back in 2017, and a beta version was released in July 2019 under the name "App + Web Properties".

What You Need to Know About GA4

Google Analytics 4 has been developed and launched at a time when two storms are hurtling towards us that threaten Google’s business:

The first storm is the move away from static websites, and toward single page and mobile applications. This is the smaller of the two storms, but threatens a marketer's ability to understand how well users are engaging with websites and mobile apps. The prior version of Google Analytics was released in 2013, and was not designed to handle apps well.
The second and larger storm is the increasing demand for user privacy. This threatens a marketer's ability to identify an individual user as they navigate the web, which in-turn threatens Google’s core business of presenting individuals with personalized advertisements.

Google Analytics 4 includes a huge number of changes that are designed specifically to enable marketers and analysts to continue working in the midst of these two storms.

There are two fundamental changes in GA4 that make this possible: 1) the event-driven data model, and 2) modeling.

The Event-Driven Data Model

If you are familiar with the event-driven data model, feel free to skip to the next section. Otherwise, here is a quick overview:

In the early days of web analytics (say mid-2000’s) websites were composed of pages that loaded every time they were viewed. As a result, we all agreed that it was a good idea to group these page views into sessions every time we noticed a break for longer than 30 minutes. Then mobile and single page apps came along, and the concepts of page views and sessions didn’t always apply any more.

As an example, consider how important the page view is to web analytics. It is used to calculate bounce rate, time on site, time on page, pages per session and more (not to mention the importance we place on “landing page”). Apps that simply load and run long processes (like games, video players, etc) cannot meaningfully describe user behavior with the traditional page view approach.

Solution: The Event Driven Data Model

Mobile app Analysts solved this problem by discarding the notion that the fundamental building block of a session is a page view, and replacing it with a flexible system of events, parameters and user properties. These new concepts can be flexibly applied to any new application, but they can also be applied nicely to a traditional website. For more info on the event driven data model, read my post onDifferences Between Traditional Analytics and GA4.

Google did not invent the event driven data model, and it’s not new — it has been the core of Firebase Analytics since it was acquired in 2014. It is a tested and proven approach that has been around for years, but it has rarely been applied to traditional websites before GA4 came along.

The second fundamental change in GA4 is the way Google uses modeling to repair data that has been fragmented by browser restrictions and consent settings.

Behavioral and Conversion Modeling

Analytics data is imperfect. Some users use run ad blockers that prevent analytics tools from collecting data. Other users visit your website or app from an iOS device, which forces cookies to expire in 7 days. And other users opt out of cookies in the consent management tool. These are not new problems for Analytics tools, but they have become increasingly severe since 2018.

To help marketers fill the gaps in your data that are created by cookie restrictions and cross-device behavior, Google Analytics 4 has started differentiating high quality data (consented, first-party data that has not been fragmented by browser restrictions) from all other data. Then, they use modeling to identify trends within high-quality data that can be used to repair the known problems within the low quality data.

As a result, every report in Google Analytics 4 is generated from data that follows into one of these two categories.

This is a major enhancement over prior versions of Google Analytics (as well as over other Analytics tools in the market). To learn more, check out my article How Conversions are Modeled in Google Analytics 4.

What are Some Benefits of Google Analytics 4?

There are a variety of benefits to GA4, and the ones that you find most important will depend on whether you are an Analyst, a marketer, or a developer. Since this blog is for Analysts, let's start there:

Cross-Platform Rollup

If you are an analyst who manages both website and mobile applications, you will finally have the ability to roll up your data across web and mobile. This is far more powerful than the rollup properties you may have used in the past because the data under the hood for each platform will use the same schema.

Explorations

Even if you do not manage a mobile app, you will notice that GA4 is a significantly more powerful analysis tool than legacy versions of Google Analytics. This is because the “Exploration” reports that were previously only available to GA360 users are now available for free, and they’ve been greatly enhanced so that you can easily explore data, analyze individual users, create custom conversion funnels, compare segments, and conduct pathing analysis. Also, the BigQuery export is now available to everyone as well, so even users on the free version of GA can get access to their raw data to join it with other sources or run SQL queries on it.

So those are the two big updates for Analysts, but if you are a marketer you will arguably benefit the most from upgrading to GA4. Here are the three top new features for you:

Propensity & Churn Audiences

GA4 allows you to group users into audiences based on the probability that they will make a purchase or churn within the next 7 days. This is similar to programmatic ad buying, but it leverages your first-party web behavior to generate predictions, rather than relying on data in a third-party’s black box.

New Engagement Metrics

If your goal is to drive users to engage with your website or app (like a healthcare or news site might be for example), it is important for you to know which users are truly engaged with your content by actually reading articles, scrolling, watching videos, etc. GA4 launches an innovative new way to help you track user engagement and completely removes the concept of a bounce rate (read more).

Cross-Platform Campaign Attribution

And of course, if you do manage both websites and mobile apps, the ability to truly roll up data means that your campaign attribution and the audiences you create will finally be intelligent enough to consider all of the user’s touch points with your brand, regardless of platform.

This brings us to developers (or anyone who is responsible for installing and maintaining your Analytics tags).

Enhanced Measurement

If you use GA4 on a website, many of the common user interactions that you previously had to track using custom javascript have been built into the tool. Scroll tracking, video plays, exit link clicks and much more can be enabled automatically with the flip of a switch (learn more).

DebugView

I’m very excited about this last one: GA4 actually has a built-in report called “DebugView” that allows you to isolate the real-time data flowing in from your own machine while you’re testing your code. This means that you no longer have to teach non-technical teams how to use Chrome developer tools or a proxy.

A Warning: Migrating to GA4 Will be Difficult

Despite all of the benefits of GA4 listed above, Analysts and marketers who are familiar with legacy versions of Google Analytics often find it difficult to migrate to GA4.

This is because GA4 is a complete rebuild of the Google Analytics you are familiar with.

Many of the default reports that marketers have come to rely on have been removed or replaced.
Popular dimensions and metrics such as "medium" and "bounces" no longer exist.
The process for designing your implementation and adding tags is very different as well.

To assist with this, Google began automatically migrating properties to GA4 in 2023. This auto-migration allows you to start immediately collecting data in GA4, but it only works well for simple implementations that only use the out-of-box functionality. Custom tracking applied to Universal Analytics will also need to be customized to fit the new GA4 data model.

How DiveTeam Can Help

If you are looking for confidence that Google Analytics 4 has been installed correctly and has been configured to support the needs of your business, submit the form below to speak with our team!

DiveTeam has designed a quick Google Analytics 4 audit to ensure you are following the latest best practices, we assist with correcting or enhancing your tags, and we also offer training to make sure your team is equipped to pull the data they need!

What is Google Analytics 4?

Where Does the 4 Come From?