In October 2021, Google announced the arrival of GA4, the newest update to analytics and possibly the most significant change to occur. Now, Google has also announced that as of 1st July 2023, all existing properties will flow into GA4 analytics properties and therefore standard Universal Analytics properties will stop processing data and become obsolete. All of your existing properties that are using Universal Analytics can be upgraded to use GA4, which means that a new property will be created to start collecting the new data and your old GA account won’t be affected; however, no new data will be collected, and any new accounts will use GA4 analytics so there’s no choice but to move.
Since this new update is being called the biggest change Google has ever made to the Analytics set up, it is important to understand some of the key differences between Universal Analytics and GA4.
GA4 vs Universal Analytics
One of the biggest differences between Universal Analytics and GA4 is the reliance on cookies to record events. The introduction of cookie consent meant that data in Universal Analytics could be skewed as a result of users not consenting to cookies; however, with GA4 analytics it will use machine learning to effectively fill gaps of where user consent has not been given. This will hopefully make the data collection more stable and avoid inconsistencies, which is particularly important with the phase out of the third-party cookie.
Events
One of the main focuses for analytics in GA4 is events and parameters and this is what their measurement model is based on, compared to Universal Analytics, which is based on sessions and pageviews.
GA4 will automatically start tracking some events once you have set your property up, such as ‘first_visit’ and ‘session_start’ and even a ‘page view’ can be considered as an event.
There are now 4 different categories of events that you can set up instead of having an event category, action, and label. The categories are:
- Automatically collected events
- Enhancement measurement events
- Recommended events
- Custom events
Automatically Collected Events
As the name suggests, automatically collected events are triggered and logged by GA4 on pre-defined user activities such as downloading a file or reaching the bottom of a web page. In order to track these events, you don’t need to add any code or tag to your website or app.
Enhanced Measurement Events
Similarly, enhanced measurement events are also automatically logged, but only when enhanced measurement has been enabled. Some examples of these kinds of events are clicking on an external link or performing a search on your website.
Recommended Events
Recommended events are the events recommended by Google and they can only be triggered and logged through manually implementing them, using the exact same event name and parameters are given by Google. These events could be things such as ‘add to cart’ or ‘add to wishlist’.
Custom Events
Finally, custom events are events that you specifically create and use, or in other words any activities that are not tracked by default such as button clicks or form submissions.
With each event you can set up 25 customer parameters, which supplies additional information.
It is possible to migrate the events you have set up in Universal Analytics into GA4, but you will have to set up a new configuration, which is basically setting up a new project, with a new data structure and new naming conventions. In GTM, you can use the same container, but create a new set of tags, using the GA4 configuration. You must do this while Universal Analytics is still running, as you might be able to reuse variables which makes the migration faster. Additionally, check that any of the events that you’re currently tracking in Universal Analytics won’t be automatically tracked by GA4, as there is no need to migrate these over.
Reporting
Another key difference of GA4 vs Universal Analytics is the changes in the reporting interface. Firstly since, the measurement model of GA4 is different compared to Universal Analytics, you should not expect to see the same reports available. Additionally, many of the metrics that are familiar to Universal Analytics have been replaced or removed in the new changes.
For example, dashboards have been replaced by the new customisable GA4 homepage. This change comes with less limitations compared to dashboards such as the shape, placement, and size of the scorecards. Additionally, in general the interfaces for GA4 use less tables, in favour of scorecards and widgets, which will hopefully make it easier to spot trends, patterns and anomalies in your data.
Additionally, another difference between GA4 and Universal Analytics in reporting is the ability to create different reporting views per property. In Universal Analytics you were able to create 25 views per property; however, in GA4, you can only use one and there is no option to create any additional reporting views.
Tracking IDs
The tracking ID that was used in Universal Analytics to set up any type of tracking via Google Tag Manager (GTM) has now been replaced by a measurement ID. These two IDs are not the same as each other. Measurement IDs identify data streams, which are the flow of data from customer touchpoints to GA4 analytics. Data streams can come from a website, or a mobile application and you can send data from up to 50 data streams into one GA4 property to consolidate your data.
There are many differences between Universal Analytics and GA4, and there are probably many more to appear thanks to the enhanced capabilities of GA4. It is important to understand and get to grips to the new changes that are being made to analytics with GA4, especially since there is no alternative to use after July 2023. Therefore, if you already have existing properties in Universal Analytics, you should start considering migrating into GA4 sooner rather than later. If you would like advice on the new changes being made to analytics with GA4 or any more advice on technical SEO, our experts are always happy to help!
Get in touch with us today!
Photo Credit: Stephen Phillips
