What is it again?
The TLDR version is that GA4 is the new (gold) standard for GA; it expands data collection and brings premium features to the masses.
However, this isn’t just tinkering around the edges (I’m looking at you Universal Analytics), it’s a complete rebuild and radical rethink of the platform.
Implementing it is effectively a migration project. And that means not just new tagging, but training for your teams and adapting existing processes and workflows.
Where are my reports?
Let’s start by addressing the elephant in the room – if you were to open GA4 right now, you might find it a little… empty.
For the past 10 years, we’ve become accustomed to Google providing us with a wealth of pre-built reports, allowing us to dive straight in from day one, but in GA4 we only have a handful of, often very simple, reports.
This may lead to a perception that GA4 has less to offer, and I’ve heard it suggested multiple times, but the truth is that the power of GA4 needs a little more work to unlock – and trust me, it’s worth it.
The standard reports have been rethought to focus on providing quick summaries of the most important areas, such as user acquisition and sales, with the detailed reporting tasks moved into Explore, which offers more powerful reporting tools.
This is a recognition that different data, and therefore reports, are important to different businesses, and that app and web data is fundamentally different in some respects – let’s not forget that GA4 is all about being able to track anything across any platform.
GA4 has also unlocked the power of BigQuery for all users, which is huge! It allows for more powerful analysis, provides a route to unsampled data, and all of that can still be linked into your Data Studio dashboards seamlessly. The challenge is you’ll need some SQL skills to use it, so leaning on an analytics teams is necessary now.
If you need any of your custom dimensions or metrics in your reports, you’ll need to use one of these two routes.
You’re also going to need to build some reports from scratch, and not all of those will be best served within Google Analytics 4 itself. It’s therefore worth spending some time mapping out your reporting requirements and identifying the best way to tackle them and giving your analytics team a seat at that table now will save repeating yourself later.
Enhanced Measurement
GA4 brings a powerful new feature: Enhanced Measurement. This allows you to configure some of the more common events within the GA4 interface, without the need for additional tagging (think outbound link clicks and video playback events).
However, this is not a silver bullet and you’ll still need to implement events tags to get the most insightful data about your website or mobile app, because that data is specific to you and some of the enhanced measurement events are just a little too basic. For example, scroll events don’t provide scroll depth measurement, only a single event for users who reach 25% depth.
To make the most of it you’ll need to plan your event tracking requirements in advance and document them. This way you can identify which are covered well in Enhanced Measurement and which will need custom event tracking.
It really is worth investing the time though; you’ll need the documentation elsewhere, and you’ll be able to focus development and analysts on the most important areas, filling in the simple bits with a few clicks yourself.
Custom Dimensions and Parameters
The collection of custom data has been massively expanded in GA4, but it’s not quite as simple as having extra custom dimensions available to you.
Custom data is passed to GA4 as parameters within your event tags, these then need to be defined as custom dimensions and metrics within GA4 for use within Explore and audiences.
But it’s worth noting that the collection quota for parameters is much higher than the number of custom dimension slots available – 25 user-scoped dimensions, 50 event-scoped dimensions, and 50 event-scoped metrics, but up to 25 parameters per event (a theoretical maximum of over 12,000 parameters).
If you haven’t defined a parameter as a custom dimension or metric, you won’t be able to see that data in GA4, but it will still be collected and become available within BigQuery. And this opens up some really interesting options for working with custom data.
To my mind, the best route forward is to reserve custom dimensions and metrics for marketing use, as this is the team that will get the most benefit from Explore and you’ll want them for creating audiences to pass to Google Ads. Analytics teams can then work with their custom data within BigQuery, which is much more capable as an analytics tool.
No more squabbling over who gets to use the dimension slots? Utopia!
But that’s only possible if data collection is carefully planned in advance, so everyone is clear on what data is available and where it can be found, and important marketing data isn’t invisible to your teams.
And you’ll want to identify whether any of the output from BigQuery analysis should be sent back to GA4 – think customer lifetime value segments! That’s not going to be possible if you don’t know what that team is collecting.
AI at your fingertips
Google has been adding AI features to GA for a few years now but has been quite limited in scope – really there’s nothing of note outside the Insights panel on the Home page, unless you’re lucky enough to have a 360 subscription.
With GA4, Google is turning things up a notch. Not quite to 11, but they’ve definitely made 10 louder.
The most exciting feature right now is predictive audiences as it allows you to create segments based on automated predictive insights, such as which users are most likely to convert in the next few days. These can be applied to reports – and also activated within Google Ads.
We’re sure there are many more AI features in the pipeline, and we’re really excited to see where they’ll take us, but there are opportunities to go beyond predictive audiences right now… with a little support from your friends in the analytics team.
BigQuery contains some interesting pre-built machine learning models, which you can easily apply to your GA4 data thanks to the free exports. This means AI-powered insights without a full data science team; you just need someone with some handy SQL skills.
Marketing x Analytics
GA4 is the future – there’s no getting away from it. It’s powerful and very different, but incredibly valuable once you adapt to get the most out of it.
By now, I’m sure you’ve spotted the theme that planning is more important than it’s ever been and so is a close working relationship between marketing and analytics teams.
Over the years businesses have focused a lot on making teams self-sufficient, and this just isn’t practical in a GA4 world. Instead, make use of the skills available within your business or your agency partners to free up your marketeers to do what they do best; optimising returns on marketing spend.
If you need any help with planning your GA4 adventure, getting your teams up to speed on the new capabilities, or just a spot of extra SQL knowledge, find out how we can help.
