Google analytics 4

Companies in B2B sectors, in particular, can benefit from the upcoming switch to Google Analytics 4 in 2023. The implementation of the new machine learning-oriented analysis solution enables a cross-channel understanding of customer needs. Read about the other benefits here.

In March, Google announced the end of the Universal Analytics tracking solution. By mid-2023 at the latest, all users of the popular web analytics tool must have implemented the successor, Google Analytics 4, in order to be able to continue running the required analyses. In this guide, you can find out how the solutions differ, what companies need to know now, and how best to approach the changeover.

How does Google Analytics 4 work and what is different than before?

In the past, Google’s tracking solutions have recorded and evaluated user behavior based on third-party cookies. But this strategy has been on the decline for years, as users see their privacy restricted and prevent use and evaluation through browser settings or ad blockers. The EU Privacy Directive could also ban the use of such solutions in the medium term or, as is already the case today, make it more difficult with consent management banners.

Google Analytics 4, therefore, does without it completely and instead works more event-based and across devices with a complex machine learning algorithm. This not only means the need to rethink operator guidance and reporting templates, but also changes a lot of possible metrics and KPIs: Popular key figures such as the bounce rate can only be determined indirectly, but an engagement rate and engaging sessions, for example, are used give, with the help of which one can reach one’s goal in a different way. For this purpose, events such as scrolling, interactions with videos, or downloads are automatically recorded in order to better record user behavior on the website and to feed the machine learning algorithm.

Does Google Analytics 4 achieve similarly good results as Universal Analytics?

That remains to be seen and is currently one of the most exciting questions in the online marketing world. It is clear that Google is caught between the two chairs here – because the company has heard the call for more privacy and fewer cookies from users on the one hand, but on the other hand, it also earns part of its money with online advertising and the associated targeting. Because the companies are of course also interested in a similarly good conversion as before.

However, it is also true that the projections and simulations that are used as an alternative to the proven third-party cookies always represent “only” a kind of alternative solution. Nevertheless, cross-device user tracking with its individual touchpoints is a promising approach and a reasonable compromise.

Why and how do B2B companies benefit from using Google Analytics 4?

Event-oriented tracking of user activities allows a much better understanding of the customer journey and user behavior – a decisive advantage for sales-oriented teams with complex customer communication and loyalty. Because now every single touchpoint can be recorded and evaluated, which justifies the additional effort, especially in high-priced B2B business areas. The graphic evaluation of user behavior in particular allows new insights that the old system, which could only look at the last point, did not offer. Companies with a complex digital strategy based on many different channels and end devices, apps and websites will benefit the most.

But all of this produces a significantly larger number of data points – and so it remains to be seen how well Google manages to process this wealth of data with its machine learning algorithm and to enrich it with predictions about customer behavior.

Why should companies start implementing Google Analytics as soon as possible?

Google has announced that it will no longer support new analyzes and data evaluations on existing Universal Analytics instances from July 2023. Only the data from the past remains accessible for a further six months (and should therefore be archived externally if you need it beyond that). It is to be expected that the relevant agencies and technical service providers will run out of resources in the months leading up to this.

In addition, the new GA4 solution requires some training both in terms of operation and the supported KPIs. But not only do the administrators in front of the screen have to learn, but also the artificial intelligence behind it. The machine learning algorithm becomes more accurate and accurate in its results, the more data points are included here. It also makes sense to be able to include comparable data from the previous year (YOY data), which, from a purely mathematical point of view, requires rapid onboarding with Google Analytics 4 in order to be able to have one year of parallel operation before UA is switched off. In addition to the respective data of an individual company, it is also about how the machine learning algorithm works.

How should B2B companies proceed when introducing Google Analytics 4?

Before the onboarding process, there is the development or adjustment of the strategy and the opportunity to think about the fundamentally required online marketing analyzes and to strategically reposition oneself here. Because much of what was important in the past has lost relevance over time – in favor of other topics and metrics.

For the technical implementation, companies will make use of the services of an appropriate agency and should first draw up a list of the systems to be implemented (or have them drawn up). Because the more complex the IT structure of a company, the more factors can lead to problems – when integrating the customer management, content management, or shop system, but also in the consent management process. It has been shown that native GA4 support is still not available for all systems. It is therefore often necessary to work with plug-ins here. Then it’s time to create templates for the Google Tag Manager and reporting templates and to gather experience with the different evaluations. If you can run both systems in parallel for as long as possible, you will gain useful empirical values.


Please enter your comment!
Please enter your name here