Profit-driven optimal segment churn prevention actions for a financial institution. A multi-action and shared-segment budget constraint approach
Loading...
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Customer attrition is a significant source of revenue loss for organizations, and attracting new customers is estimated to be five to six times more expensive than retaining an existing customer. This paper presents a profit-driven optimization approach to prevent customer churn using a multi-action customer retention action portfolio subject to a budget constraint. The proposed approach integrates the problem of predicting customer churn through analytical models with the issue of selecting specific customer retention actions from a set of defined actions with different costs and effectiveness while enforcing budgetary constraints. The main contribution is developing an optimization framework that matches the customer with the actions, maximizing the expected Customer Lifetime Value retention by applying a given action to a set of customers subject to a budget constraint. This proactive approach is compared with industry customer-action matching heuristics such as sorting by Expected Profit of a Customer, churn ratio, and Customer lifetime Value. Results show that the proposed model significantly outperforms the mentioned heuristics, using a shared budget over a specific campaign budget is preferred, an adequate portfolio of retention actions is crucial for maximizing value, and overestimating a campaign’s effectiveness can reduce the value. The proactive model was tested in a set of 150,000 customers and increased the return on investment of the retention actions by over a 185% compared to the reactive approach of the company.
Description
Tesis (Master of Science in Engineering)--Pontificia Universidad Católica de Chile, 2024
Keywords
Customer churn, CLV, Optimization of marketing campaigns