Kalman Filters to Generate Customer Behavior Alarms

Aggressive marketing campaigns to attract new customers only covers customer churn, resulting in neither growth nor profitability. Retaining current customers, increasing their lifetime value, and reducing customer churn rates, thereby allowing greater efforts and resources to be dedicated to capturing new customers are the goals of a commercial director. But how can that loss be detected in time and avoided—or at least reduced? There is the 3A program to keep customers loyal, based on analyzed information from our customers, to construct an expert alarm agent and one-to-one retention actions. In this paper we show how to apply the Kalman filter and study how to configure it to predict the normal behavior of customers by projecting their consumption patterns into the future. Abnormal behavior detected by the Kalman filter triggers alarms that lead to commercial actions to avoid customer churn.