What templates are supported on the Engine?

Customer Churn Prediction

The customer churn prediction template on the AI & Analytics Engine supports two different types of churn use cases:

• Transactional option: Predicting the likelihood of churn for active customers based on their purchasing/financial activity. These customers have no “contractual” relationship with the business. Churn occurs when their activity level drops below a certain threshold over a period of time.

• Subscription option: Predicting the likelihood of churn for customers who have active subscriptions to a service. To build the prediction model, users are required to supply historical churn dates of customers. This particular option is suitable for business scenarios in which customers explicitly indicate their intention to churn. For example, a business may consider a customer churned if they cancel their subscription, and do not renew within 30 days.