Customer Churn Prediction in B2B Non-Contractual Business Settings Using Invoice Data

Customer churn is a problem virtually all companies face, and the ability to predict it reliably can be cornerstone for successful retention campaigns. In this study, we propose an approach customer prediction in non-contractual B2B settings that relies exclusively on invoice-level data feature engineering uses multi-slicing maximally utilize available data. We cast as binary classification assess of three established classifiers when using different definitions. also compare classifier performance amounts historical are used engineering. The results indicate robust models definitions derived by alone more creating some features tends lead better performing classifiers. confirm dataset creation yields compared traditionally single-slicing approach.