An important Energy company requested the study of a Machine Learning solution to prevent the risk of customer abandonment (Churn Retention) to be combined with traditional customer segmentation monitoring systems present in the company. More specifically, the problem was focused on the knowledge of the Customer at 360 ° by observing the workflow of payments and Credit Collection actions.
The solution was the provision of Advanced Analytics based on the observation of the payment workflow and Credit Collection actions on the entire Customer Base of the Company. The Credit Collection is typically a flow that involves different Business Units and different types of data, from the Management and Control Business Unit that holds payment data, contractual and personal data, to the Credit Collection Business Unit that operates with actions on the Customer generating data generated by the phases of the Credit Recovery flow, from the Commercial and Marketing Business Unit that uses segmentation criteria to define/renew contracts or promotional actions.
Advanced techniques of Clustering and Pattern Recognition have made it possible to reclassify the segmentation of customers, based both on the risk of abandonment, on the dynamism and frequency of payments and their delays, and on the value of the economic damage.
The benefit of using a Machine Learning system alongside or replacing rules-based criteria consists above all in the value of “Continuous Learning” which allows you to monitor the variation of the “behaviors” of the Customer constantly.
Such a reclassification of Customers based on the risk of both loss and economic damage allows the Business structure to intervene with specific policies to adjust/renew contracts for certain segments or build specific campaigns for a certain “risk” segment.
Implementation of the Machine Learning Pilot service, from the creation of Machine Learning Models to Advanced Analytics. Planning of the Executive Project of Risk Governance with Machine Learning tools.