Maximizing Customer Lifetime value using Dynamic Programming: Theoretical and Practical Implications

Dynamic programming models play a significant role in maximizing customer lifetime value (CLV), in different market types including B2B, B2C, C2B, C2C and B2B2C. This paper highlights the main contributions of applying dynamic programming models in CLV as an effective direct marketing measure. It mainly focuses on Markov Decision Process, Approximate Dynamic Programming (i.e. Reinforcement Learning (RL)), Deep RL, Double Deep RL, finally Deep Quality Value (DQV) and Rainbow models. It presents the theoretical and practical implications of each of the market types. DQV and Rainbow models outperform the traditional dynamic programming models and generate reliable results without overestimating the action values or generating unrealistic actions. Meanwhile, neither DQV nor Rainbow has been applied in the area of direct marketing to maximize CLV in any of the market types. Hence, it is a recommended research direction.