Machine Learning in Google Analytics 4
In 2020, Google announced that it would be launching a new and improved version of Google Analytics, the world’s more widely used web analytics service. This newest version, called Google Analytics 4 builds on previous iterations and specifically builds on the App + Web property launched last year. Google Analytics 4 has machine learning at its core to automatically surface helpful insights and give you a more robust understanding of how users are interacting with your business across devices and platforms. This change should help your business to get an improved ROI from your long term marketing strategy and the new focus on machine learning will help give your business a more complete understanding of your users.
What is Google Analytics 4?
Google Analytics 4 is a complete update of the platform and takes over from the previous generation called Universal Analytics which was launched back in October 2012. Google Analytics 4 is focused on customers – with a shift away from sessions and a focus on interactions and engagement over time with better integrations across Google’s marketing products.
Some of the areas in the platform that will immediately be noticeably different to veteran Universal Analytics is that now on the left hand side, we can see that Google Analytics splits areas by customer life cycle stages:
- Monetisation and
Search Engine Land tells us that the new Google Analytics offers deeper integration with Google Ads including the ability to maintain audiences between Web and Apps. Other noticeable updates include the ability to measure more on-site or in-app actions without the need for Google Tag Manager and more options for Data Controls to help with GDPR compliance.
How does Google Analytics 4 use Machine Learning?
One of the core elements of the new Google Analytics is the inclusion of more insights and predictions based on Google’s Machine Learning algorithms. Vidhya Srinivasan, Google’s Vice President of Measurement and Analytics said in the announcement that: “By applying Google’s advanced machine learning models, the new Analytics can automatically alert you to significant trends in your data – like products seeing rising demand because of new customer needs. It even helps you anticipate future actions your customers may take.”
For example, it calculates churn probability so you can more efficiently invest in retaining customers at a time when marketing budgets are under pressure. She goes on to say that Google are continuing to add new predictive metrics, like the potential revenue you could earn from a particular group of customers. This allows you to create audiences to reach higher value customers and run analyses to better understand why some customers are likely to spend more than others, so you can take action to improve your results.
Machine learning in Google Analytics 4
With machine learning playing such a key and central part of the new Google Analytics 4, businesses can now get a much more comprehensive understanding of their data. With the new functionality in Google Analytics 4, you can now predict the probability of conversions and build audience forecasts around those probabilities, identify and be warned of key trends, and more easily identify anomalies. What’s more, using natural language processing, you can search for things in in Google Analytics 4 using natural language to get instant results. For example, you can search for ‘conversions last week from London”. This means it’s now easier and quicker to find the data that’s important to you.
Google plans to continue to build on and develop Google Analytics 4 adding new forecasts so that more businesses can build more effective marketing strategies to increase their return on investment.
Setting up Google Analytics 4
Google Analytics 4 is available now to all Google Analytics users. All newly created Google Analytics properties will use the new experience by default. If you have an existing GA account and you want to look at the new features you can set up a new property alongside your existing ones to try out Google Analytics 4.