Model of Customer Lifetime Value with Two Customer and a Former Customer State

A model of Customer Lifetime Value can be designed by a stochastic process of Markov Chain. In the model, there are several states that must be passed by a customer. Every two states have a transition probability as a representation of state moving probability by a customer. Transition from a state to another state is assumed to fill Markov chain. In previous research, the model only has one customer state and some former customer states. Here, the model is built by two customer states and a former customer state. In real conditions, many companies are found to have more than one product variant for one type of goods/services. Typically variant differences are based on differences in taste, quality, volume/mass or services. Each variant usually has its own customer. But do not rule out that a customer can buy the product by alternating variants. The problem is how to calculate the customer lifetime value for each product variant. The expected result is an increase the total customer lifetime value compared with the previous model.