Google Analytics 4 is a different tool from the Universal Analytics most of us grew up on. It is event-based, it leans on machine learning, and it was built for a world with fewer cookies. Here are the ten benefits that make it worth knowing well.
It gives you a level of insight that was not possible in the older versions, once you know where to look.
Google Analytics 4 was officially launched with machine learning at its core in October 2020.
The previous versions of Google's platform for web analytics include:
- GA1: Classic Google Analytics (ga.js JavaScript library)
- GA2: Universal Analytics (analytics.js JavaScript library)
- GA3: gtag.js JavaScript library.
The short version
- GA4 is event-based, machine-learning-led, and built for a cookieless web, so it is a different tool from Universal Analytics, not just a new coat of paint.
- Its standard reports are unsampled and collection is unlimited, so decisions rest on your full data rather than an extrapolated fraction.
- It follows one user across devices and apps in a single schema, and predicts behaviour with purchase, churn and revenue metrics.
- Enhanced Measurement tracks page views, scrolls, outbound clicks, site search, video and downloads out of the box, with no extra tags.
- Privacy is baked in: IP anonymization is on by default, and Google is adding modeling to fill cookieless gaps.
What makes GA4 a real upgrade?
Here are the ten that matter most, from the way GA4 collects data to what it can predict.
#1 Unsampled Data
Earlier versions of Google Analytics capped collectible data at 10 million hits per property, and sampling kicked in once a segment passed 500k sessions.
GA4 drops sampling from its standard reports, and collection is unlimited. That matters because every decision then rests on your full data set, not a fraction of it that the tool extrapolated from.
Even though sampling is practical with segments and secondary dimensions, there is a risk of losing information scope if you rely only on sampled data.
#2 Extended Segmentation Possibilities
GA3 conditioned marketers to split users and their interactions by device and platform type. Time was also a problematic factor to incorporate into reporting.
But, now, Audiences in GA4 allow for a much more actual marketing approach: you can now create segments based on events and add the concept of time. You can analyze the time users take to progress through the sales funnel.
No more unnecessary splitting of users in GA4! And, all the Audiences published in GA4 will automatically be added to Google Ads for the most precise reach yet.
#3 Complete User Journey
With the new section in GA4 called Life Cycle, it's never been easier to focus on customer behavior and get insights about which part of your marketing strategy needs improving or whether you should change its focus.
The reports here split into Acquisition, Engagement, Monetization, and Retention, each mapping to a stage of the customer journey.
Acquisition and Engagement tell you how people find you and what they do once they arrive, which is where you judge your awareness campaigns and paid ads.
Two more things are crucial parts of the user journey: How many revenue goals you're hitting and if you're doing enough to build customer loyalty. Monetization and Retention reports will give insight into how well you're retaining customers and how much money campaigns bring.
GA4 moved from a goal-oriented model to a user-centric, event-driven one. It is built to follow a whole customer journey, so when someone discovers you on their phone and converts later on a laptop, GA4 can still join those visits into one user.
#4 More Intelligent Tracking
With privacy laws tightening, third-party cookies are getting harder to rely on. GA4 leans less on cookies and more on AI and machine learning to fill the gaps, which is the direction web analytics is heading.
Vidhya Srinivasan, the Vice President of Engineering at Google Ads, Google said:
"Because the technology landscape continues to evolve, the new Analytics is designed to adapt to a future with or without cookies or identifiers. It uses a flexible approach to measurement, and in the future, will include modeling to fill in the gaps where the data may be incomplete. This means that you can rely on Google Analytics to help you measure your marketing results and meet customer needs now as you navigate the recovery and as you face uncertainty in the future."
#5 Predictive Metrics & Audiences
One of the strongest uses of that machine learning is predicting what users will do next. GA4 can estimate which users are likely to buy in the coming weeks and how much revenue they will bring. It does this with three predictive metrics for ecommerce sites:
- Purchase probability. The chances that an active user in the last 28 days will purchase within the next 7 days
- Churn probability. The chances that a user active in the last 7 days will be inactive within the next 7 days
- Revenue prediction. The expected revenue for the next 28 days from a user who's been active in the past 28 days
#6 Better ROI
All of this rolls up into better return on ad spend. Cleaner data, real attribution across devices, and predictive audiences mean you can point budget at the people most likely to convert, and pull it back from the segments that do not. The reporting is only useful if it changes where the money goes, and GA4 gives you the signals to make that call.
#7 Enhanced Measurement
One more essential benefit of GA4 is the Enhanced Measurement event that tracks data streams automatically without additional tags. In a new way, it measures things like:
- page views
- scroll
- outbound clicks
- site searches
- YouTube Video engagement
- file downloads
There is also the option to create custom events if they don't already exist.
#8 BigQuery Integration
BigQuery is Google's data warehouse that processes SQL queries at very high speeds. This is important if you need to analyze and process terabytes of raw data and get insights quickly. BigQuery lets you use the advantages of machine-learning capabilities and even stream data directly to BigQuery.
This feature isn't entirely new, but it was only available to GA360 users. There are still limits on storage and data size, but the free monthly allowance goes a long way: 10GB of storage and 1TB of query processing.
#9 Machine Learning
Machine learning runs through GA4 in ways you do not always see. It flags trends and anomalies in your data automatically, surfaces which products are gaining demand, and points out shifts you might have missed in a manual review.
#10 Privacy-Centric Data
Privacy rules keep tightening, and Google has to follow them too. GA4 was built to adapt to a cookieless web rather than fight it.
What is remarkable about GA4 is the option to control how data is collected and retained. IP anonymization is enabled by default, which is a step forward from GA3, where it was disabled by default. There is also granular control for ads which lets you choose when to use data for optimization or measurement.
GA4 is planning to use modeling to fill in some of the gaps in the cookieless tracking.
How is GA4 different from Universal Analytics?
Those are the benefits. The other half of the picture is how GA4 actually differs from Universal Analytics day to day, which takes a little getting used to.
In a separate post, we've broken down the differences between the metrics in these two analytics platforms. But here are the most vital differences between Google Analytics 4 and Universal Analytics to speed up the process. (and other previous versions)
Reporting Interface
The GA4 reporting view looks unfamiliar at first, but it settles quickly once you know the layout.
Yes, reports and metrics are not in the same place as before, and they have been either removed or replaced.
Reports have changed; there aren't as many because they're generated only when you start tracking events.
If you used Firebase, you'd see that the reporting interface in the GA4 looks very similar to it, and it's not by accident: GA4 is built on Firebase Analytics.
It is a real change from the GA3 view, but you get used to it, and that is what this guide is here to help with.
Measurement Model
If you've even scratched the surface of GA4, you may know this one.
The primary difference between GA4 and GA3 (or UA) is the data model, which is event-based. The earlier case was pageviews batched into sessions in UA.
So, the main point is that GA4 gives users much more space and flexibility than previous GA versions because it has a unified tracking system for websites and apps, while it's not bound entirely to URLs.
But, pageviews and sessions still exist, but they're listed as separate events.
Set Up Tracking IDs
Measurement IDs enable any tracking in Google Analytics via GTM.
But in Universal, this was Tracking ID.
If you already have a GA4 property with a web data stream, your Measurement ID begins with G, for example, G-SV5AS78HKN.
But if you have set up a GA3 property, then it has a Tracking ID, and it begins with UA, for example, UA-5789025-9.
Event Tracking Set Up
UA has been an excellent tool for a long time, but everyone could see the limitations. With Universal, every event was trackable if it followed the schema of category-action-label-value, like this:
GA4, on the other hand, doesn't impede this on users but has a much more flexible event tracking setup. There are more parameters beyond the four old ones, which you can use to assign any event type.
Event Tracking Automation
There was no automatic tracking in Universal Analytics / GA3. But now, the Enhanced Measurement feature is built in the GA4 property, and it allows automatic tracking for events like:
- Scroll tracking
- Video tracking
- Exit tracking
- Site search tracking and more.
Automatically tracked events require no additional coding or tagging.
User and Event Data Retention
With previous versions of Google Analytics, there were more User and event data retention options. Basically, this feature lets you set the time for which GA retains user-specific data for inactive website users like cookies, and user and advertising identifiers, before deleting it automatically.
In GA3, you can set time to 14 months, 26 months, 38 months, 50 months, or Don't automatically expire.
In GA4, there are either 2 months or 14 months, and no other options are available.
Cross-device & Cross-platform Tracking
One of the most noteworthy changes from GA3 to GA4 is including web and app data into the same schema. This feature wasn't available with Google Analytics 3, but with GA4 is.
That single schema gives you more reliable data than GA3 could, because web and app behaviour finally sit in one place.
Custom Dimensions
Custom dimensions are different in GA4 than in GA3. You can create a new custom dimension by creating a new custom event parameter in GA4.
In GA3, there is the option to set or change the scope of the custom dimension to Hit, Session, User or Product. But GA4 has only the Hit dimension, while Session doesn't exist anymore.
There are some loopholes for creating custom dimensions with User or Product scope:
- If you want to create a custom dimension with the User scope, create User properties in GA4.
- If you want to create a custom dimension with the Product scope in GA4, make use of item parameters like: item_category, item_category_2, item_category_3 and so on.
Custom Metrics
These are created differently in GA4 than in previous versions, so what you need to do in GA4 is create a new custom event parameter that has specified the unit of measurement.
In GA3, the scope can be changed to Hit or Product, like custom dimensions. And like with custom dimensions, in GA4, there isn't the possibility to change or set the scope of your custom metric, but only the Hit scope.
Debugging with DebugView
Reporting in GA4 provides you a DebugView report used for validating analytics configuration from the reporting interface. It wasn't possible to view this in GA3.
IP Anonymization
Google Analytics records the identity of your website users via IP addresses and stores it for future use but does not display this information in reports.
Because of local privacy laws taking action and GDPR, IP anonymization in GA4 masks website visitors' IPs.
*Under GDPR, your IP address is considered personal data.
The last three digits of your visitor's IP will be automatically deleted when you anonymize IP addresses. In a more technical sense, this feature sets the last octet of IPv4 user IP addresses and the last 80 bits of IPv6 addresses to zeros, and GA4 will mask the IP automatically.
In GA3, on the other hand, IP anonymization is disabled by default, something that new privacy policies around the world don't allow anymore!
If you're still using GA3, it's possible to enable/disable this feature.
In GA4 properties, anonymization is enabled by default without the option of disabling it.
Reporting Views
With the new version of Google Analytics, you can only have one reporting view. In GA3, there were 25 reporting views per property. You can't really add more views in GA4 at this time.
But there is a bypass! You can also create new Audiences or Data streams and use them in place of filtered views.
Advanced Analysis Reports/Custom Reports
GA4 includes advanced analysis tools that let you create impromptu funnels. This is new in Google Analytics, at least in the free version, and GA360 had this feature available before.
Advanced analysis reports let you do a few specific but highly beneficial things.
- A funnel exploration you can customise with far more options than the old funnels.
- The option of Path Analysis helps identify the routes users take most, like visiting a page and then subscribing.
- A free-form exploration where you pivot dimensions and metrics into the table or chart you need.
- With the segment overlap report, you can see correlations between targeted segments.
- A user explorer report that drills into the individual users inside a segment you choose.
Final Thoughts
If for some reason you are still on the old setup, move to Google Analytics 4 now. Universal Analytics stopped processing new data on July 1, 2023, so GA4 is the only version Google still runs.
In this post, we've highlighted some of the features of GA4 that are especially beneficial for mobile app data and website usage metrics.
When you are ready to set it up properly, our GA4 setup checklist walks through it step by step. If you would rather have it done for you, that is what our Google Analytics 4 setup service is for.
Frequently asked questions
What are the main benefits of GA4? The biggest are unsampled data with unlimited collection, a cross-device view of the whole user journey, predictive metrics powered by machine learning, free BigQuery export, automatic Enhanced Measurement, and privacy-by-default. Together they give you cleaner data and sharper targeting than Universal Analytics could.
Is GA4 data sampled? Not in standard reports. GA4 drops the sampling that kicked in on large Universal Analytics segments, and collection is unlimited, so every decision rests on your full data set rather than a fraction the tool extrapolated from.
What are GA4's predictive metrics? Three, for ecommerce properties: purchase probability, the chance an active user buys in the next 7 days; churn probability, the chance an active user goes inactive; and revenue prediction, the expected revenue over the next 28 days. GA4 builds them with machine learning.
Does GA4 track events automatically? Yes, through Enhanced Measurement. It captures page views, scrolls, outbound clicks, site searches, video engagement and file downloads with no extra tags or code, and you can add custom events on top.
How is GA4 different from Universal Analytics? GA4 is event-based where UA was session-based, it fits web and app data into one schema, it adds DebugView and impromptu funnels, and it anonymizes IPs by default. We break the metric-level differences down in our UA vs GA4 comparison.
