Customer Equity Measurement And Maximization

Customer equity management theory was brought forward in the end of the 20th century. Unlike other traditional management theories, it treats customer as critical asset and endeavor to optimize its maximal value. The most important part of customer equity management is its evaluation. However in the noncontractual settings,as customer's behaviors are unobservable to company, it shows great difficulties in correctly evaluating customer equity. Recently, academics have shown interest and enthusiasm in the development and implementation of stochastic customer base analysis models, such as the Pareto/NBD model and the BG/NBD model. On the other hand, there is also a growing frustration among academics insofar as these models have not found their way into wide managerial application because of complexity of calculation. According to this situation, this paper put forward several forecasting models which can not only achieve high accuracy as other complex stochastic models do but also meet the efficiency as managers require. The main work of this paper is as follows:(1) Put forward a customer defection model under noncontractual settings. Assuming that past purchase levels and the observable duration since last purchase are both predictable factors, the whole customer base will be divided into four classes according to the future purchases. And then find the cutoff threshold to satisfy the maximum of correct classifications. We show via experiment that our method performs better than other models.(2) Put forward a predicting future purchase level model. We benchmark some published models'predicted number of transactions against a simple management heuristic: Every customer continues to buy at his or her past mean purchase frequency while considering the effects from losing customers. Comparing the performances of the Pareto/NBD model, the BG/NBD model, and another simple heuristic model with our model, the results show that our model outperform on the aggregate levels but, in term of individual level not so perfect as stochastic models.(3) Consider the negative effect brought by unprofitable customers to the whole customer equity, we build a customer equity measurement system exclude those non-value customers. The comparing results shows abandoning unprofitable customers will be more realistic and more effective when calculating customer lifetime value. (4) Regarding of identification of future best customers, our method can also perform very well. This paper use past 20% best customers in a customer base to identify if they are also the future 20% best customers. The results show that our method works better than other models. Meanwhile this paper also point out that when predicting future best customers, purchase amount and corresponding cost should be considered instead of purchase times only.(5) Combining the achievements of other researchers, summarize a whole frame work for assessing how marketing actions affect customers'lifetime value to the firm.