Case study: Churn prevention in the energy environment

For a supplier in the energy industry, COSMO CONSULT has developed a solution to forecast the probability of losing lucrative customer groups. Today, the company can recognize the potential of losing customers ahead of time and counter that with suitable measures.

• Identifying customers that are likely to stop doing business with the company
• Realizing competitive advantages in the highly competitive market
• Paying more attention to customer contacts

The increase of digitization, online comparison portals and the lack of changing restrictions has made the task of changing an electricity provider easier and simpler than ever before. That is why it is particularly important in the energy industry to retain valuable customers who are willing to switch to another provider. However, it is far from easy to identify such groups of customers early and act accordingly and quickly to gain competitive advantage.

COSMO CONSULT was commissioned to develop a solution with which the probability of losing a customer can be precisely determined in order to detect and counteract the potential of such lost at an early stage.

To design the model, the project team first needed to have access to the customers and their contract data. As in the classical regression model, the Cox Proportional Hazard model also has certain predictive variables that lead to a higher mortality rate, which refers to the termination of an existing contract. Among the most important variables are:

• Consumption
• Branch
• Type of customer
• Place
• Contract history or
• Competitive prices.

Active contracts were important in the design of the model, and thus they were also included in the probability calculation. Only in this way could the probability that a customer would terminate a contract be plausibly determined.

In the daily run, the probability of losing a customer and the so-called Customer Lifetime Value (CLTV) are calculated for each contract and for each specified period of time. The current value of all future (expected) payments by a customer can be used to measure the value of a customer relationship. Customers with a high probability of switching to another provider and at the same time with a high CLTV are looked upon like the “stars” among the customers in the energy industry. The competition revolves around such customers when it comes to new customer acquisition.

The result of the model calculation was prepared by COSMO CONSULT in the form of various key performance indicators (KPIs) and customer-specific data in QlikView. Users can identify the “stars” easier and faster and initiate direct customer contacts. The ability to retain profitable, long-term customers who are willing to switch to another provider has a direct impact on the economic success of the company. Other advantages arise because the time required to select customers has shortened. For employees, it is now much easier to focus on customers whose defect would have a real impact on the company.