- 1. What's changed: analyzing GA4
- 2. New in analytics. We sort it out by category
- 3. Advantages and disadvantages
- 4. Worth implementing Google Analytics 4
- 5. For now it's better to wait with the transition
Google Analytics Update 4 was released last October. According to the developers, there are many differences from Universal Analytics. In fact, it is Firebase Analytics – it is from there that the update interface and data accounting logic are borrowed. However, GA4 contains additional functionality that is worth paying attention to.
Fig. 1 – Updated Google Analytics 4 interface
Let’s see if there are any advantages to switching to new analytics right away.
1. What’s changed: analyzing GA4
- First, Google Analytics 4 provides comprehensive machine learning to simplify analytics, Natural-language processing features, and full reports. This means that with the new version of analytics, it became possible to more successfully analyze demand trends, assess the likelihood of conversion or abandonment, and create campaigns in Google Ads. In Universal Analytics automation and reporting by device and platform was limited.
- The indicator around which the analytics was united has been changed. Now it is built around events. Google intends to abandon the artificial concept – sessions – and focus on user behavior. Now you can track the entire user’s path by collecting consistent, reliable data from all platforms and devices and without duplicating the action done by one person on different gadgets. This system is called cross-platform, and it has become one of the key features of GA4, because now you can count the real number of users, and not the electronic devices they used.
- Google Analytics 4 priority is preserving user privacy. Here, IP anonymization is configured by default, and in addition, the gtag.js library is used, which does not require the installation of cookies for data transfer.
- Integration with all Google products has been implemented, in particular with YouTube. Now there is an opportunity to better evaluate advertising campaigns.
2. New in analytics. We sort it out by category
Directly in the functionality:
- IP anonymity is set by default;
- Google Analytics code is based on Gtag with dimension ID (G-XXXXXXXXXX);
- Google Analytics 4 uses 2 cookies to track user actions: “_ga” cookie “and” _ga_XXXXXX “;
- increased power of segmented analysis. Segment overlay is now available to all users. There are more parameters, metrics and links between them and events;
- regular expressions are available in this version of analytics, but they are not available in standard reports.
In working with data:
- in Google Analytics 4 each action is classified as an event and 25 custom parameters can be added to each;
- if before that we were guided by the data model “sessions/page views”, then in the new version another model is relevant – “event/parameter”.
Fig. 2 – Google Analytics 4 Events
In data collection:
- the new analytics assumes restrictions on data collection: if we are talking about “user data”, then you can see them for the last 14 months;
- the earlier period can be captured using Google BigQuery. However, cart additions, checkout and page views will remain available.
Fig. 3 – Google Analytics 4 Data
In data streams:
- in the new analytics, data is collected from the web and from applications at the same time, and metrics and indicators are consistent with each other. You can configure multiple data streams and create up to 50 such streams for each resource;
- you can connect existing analytics for Android and iOs applications, but for the web you will have to re-create it.
Fig. 4 – Configuring data streams
In the custom path:
- the presence of path analysis in the GA4 functionality;
- the presence of a funnel analysis, which helps to analyze user behavior.
- in Google Analytics 4, you can add up to 50 custom metrics and parameters (once created, you cannot change the data layer). It is forbidden to transfer personal information;
- the list of automatically triggered events has been expanded: you can enable the display of scrolls, outbound links, site search, actions with video and file downloads;
- engagement is tracked differently: by interactions during sessions and by time of engagement. Sessions must be more than 10 seconds long, consist of at least two page/screen views, or include a conversion.
Fig. 5 – Google Analytics 4 Events
- you can analyze user behavior and fully customize certain reports;
- 4 blocks of standard reports are available: users, demographics, behavior and devices;
- there is integration with the BigQuery cloud data storage;
- three types of data filters: exclusion of data by IP, data from test devices, event filter with a special marking of the required events;
- when comparing data, advanced parameters, and applying filters in standard reports, sampling can be applied. But for this you need to either exceed 10 million events if the generated report is not standard, or when the date range is more than 60 days.
Fig. 6 – Google Analytics 4 Reports
- free integration with BigQuery, suitable for projects with an impressive amount of data.
3. Advantages and disadvantages
The benefits of the new analytics are clear.
- The focus is on user actions, moreover, combined across different platforms. Almost all parameters are taken into account. You can analyze his behavior and interaction with the business much deeper than before. Accordingly, it is easier to build communication based on personalized information.
- In addition the speed of the update has been increased.
- It has become easier to create and segment audiences, track user actions on various devices.
However, there are also disadvantages.
- The new GA version is more complicated. It becomes necessary to preliminarily consider and configure the parameters for collecting, storing and transferring data.
- Out of the box, only basic reports are provided, the rest must be configured manually. So, there are no reports comparing different attribution models, multichannel funnels.
- There are similar parameters, which can be confusing for the user.
- Filtering functionality is limited. You cannot download additional data from systems outside the site or application. In addition, the query generation option does not work.
4. Worth implementing Google Analytics 4
In terms of the specifics of their activities, new analytics are needed for companies that:
- are largely dependent in business on interaction through mobile applications: it-firms, job search sites, large corporations (banks, travel, real estate, game development);
- support many communication channels;
- lead projects with a long sales cycle.
- reason: for these businesses, “session” as a measurement parameter is not informative enough. You need to understand how a particular funnel is formed.
In terms of technical capabilities, GA4 is suitable if:
- the adjustments that need to be made to scripts and code syntax are minimal;
- collection of data on the site occurs through Google Tag Manager and Data Layer;
- Firebase is used, or at least the logic for collecting its data is known;
- YouTube Ads and User ID based remarketing are connected.
5. For now, it’s better to wait with the transition
Despite the fact that Google recommends using both analysts at the same time for a while, not all companies will be happy with the idea of actively using GA4. These are the cases:
- a site with many subdomains, and each of them is tracked independently;
- the site does not have a mobile application, a personal account in the system, a competent analyst on the staff;
- site code or GTM is used as the main tracking method, and at the same time there are many tags in the container;
- the web resource has applications, but there is no general event hierarchy or a built-in data collection scheme;
- there is no single approach to this hierarchy and a system of metrics (it is not clear how to prioritize);
- there are no specialists for working with Firebase Analytics and data in BigQuery;
- requires conversion export and integration with other Google products.
And, of course, there is no need to actively implement advanced analytics for business card sites, media sites and small online stores that do not have voluminous data.