Customer Lifetime
Value Forecast
The key to lasting success
How to identify loyal customers?
The Problem
In the subscription business, it is often unclear which types of customers and customer relationships deliver the most long-term value. We worked with a major international press house to identify loyal customers and develop customized business strategies.
Our Solution
Customer Lifetime Value Prediction
To ensure the success of the project, we first analyzed and structured the existing data together with our stakeholders. We then asked ourselves: Which customer data can we use for a prediction and which are relevant for the lifetime value? To find the answer we analyzed variables such as the type of subscription, the monthly price or the average time spent on the website in detail and then identified the most important ones.
​
We visualized the results for our stakeholders in a dashboard and jointly programmed an AI that uses them to accurately predict lifetime value. The predictions are now used to target and retain customers with a high lifetime value in the long term. A success for everyone!
Decision Trees for CLV Prediction
Our Technology
For the technical implementation of our project, we relied on decision trees. This machine learning algorithm is known for its high interpretability and was therefore ideal for our client. With its help we were able to derive simple decision rules that clearly show which factors have the strongest influence on customer lifetime value.