“Adaptive” Learning and “Proactive” Customer Relationship Management

CRM is about introducing the right product to the right customer at the right time through the right channel to satisfy the customer’s evolving demands. Ideally, it should follow the development of each individual customer and develop integrated multi-segment, multi-stage, and multi-channel CRM decisions in order to maximize the total customer lifetime profit. However, most existing CRM practice and academic research focuses on methods to select the most profitable customers for a scheduled CRM intervention. This campaign-centric approach deviates from the goal of customer-centric CRM. In this article, we discuss the two-step procedure (“adaptive” learning and “proactive” CRM decisions) and three-components for customer-centric CRM, adaptive learning (of customer individual preference), forward-looking (into future marketing consequences of current CRM interventions), and optimization (to optimally balance cost and benefit). We then formulate CRM interventions as solutions to a stochastic dynamic programming problem under demand uncertainty in which the company learns about the evolution of customer demand as well as the dynamic effect of its marketing interventions, and make optimal CRM decisions to balance off the cost of interventions and the long-term payoff with the goal of maximizing each customer’s “longterm” profit. The framework allows us to integrate all the interand state-dependent factors that drive the CRM decisions and results in inter-temporally related path of CRM solutions that are consistent with customer-centric CRM. Finally, we choose two examples to demonstrate the input, output, and benefit of “adaptive” learning and “proactive” CRM. The proposed solution meets the recent trends of companies seeking real-time solutions for integrating database and CRM decisions that are empowered by the advancement of technology.