Customer Life Time Value Model Framework Using Gradient Boost Trees with RANSAC Response Regularization

For many years, Customer lifetime value has been a mainstay concept in direct response marketing and has been increasingly considered. This paper presents a mathematical model framework for determination of customer lifetime value. The proposed multi-layer framework is based on a systematic theoretical taxonomy and on assumptions grounded in customer behavior. CLV, a recent marketing paradigm, pursues long-term relationship with profitable customers. It can be a starting point to understand relationship management and measure the true value of customers to be deployed toward the targeted customers and profitable customers, to foster customers' full profit potential. Corporate success depends on an organization ' ability to build and maintain loyal and valued customer relationships. In this paper, we propose a framework for analyzing customer value and segmenting customers based on their value. We also conducted an in-depth data analysis to find each member's behavior and important attributes which plays a significant role in calculating CLV in multiple revenue bands. From the experimental validation, we concluded that our proposed framework works much better in predicting customer's life time value in terms of revenue compared to other methods in past literature.