The Research On Dynamic Measurement Of Enterprise Cutomer Value And Strategy Analysis

The computation of customer value, as the theoretical basis of differentiated service, has attracted widespread attention from enterprisers and scholars at the time of "customer-centric" business philosophy. On the basis of data mining technology, Bayes, RFM and Pareto/NBD and using research methods such as literature review, investigation and data mining modeling, this dissertation has done in-depth research on dynamic measurement of customer value and provided a specific strategy for enterprise to improve customer management activities. This dissertation was done from the perspective of elements of the customer value, customer behavior characteristics, customer behavior prediction.This dissertation introduces the definition of customer value, expounds the relationship between customer value and information technology application, scientific decision-making, precision marketing research, and emphasizes the theoretical and practical importance of related research. Then, this dissertation makes a review of the existing studies and management practices on customer value, and points out an urgent need of customer value measurement in the enterprise management activities, which means effective dynamic measure model of customer value is to be set up.Firstly, this dissertation does an explanatory analysis about the components of customer value using factor analysis and finds the factor, customers’"recency ", is the key of dynamic measuring of customer value, whose influence on customer value has amounted to more than 50%. Furthermore, this paper is the first to prove that the survival probability of customers decreases with recency. This finding can help enterprises identify different types of customers’survival probabilities so as to analyze the differences of customer value.Secondly, this paper constructs the customer survival probability model using the posterior estimate method of Bayes statistic in order to solve the problem that customer survival probability fitted out from traditional model has difficulty to meet enterprises’future demands in the changing market. Using historical data to get customers’prior information, this method can combine current information to make dynamic posterior estimate of various parameters. This model can obtain every customer’s curve of survival probability in each period in order to achieve dynamic customer value measurement. What’s more, this model can do survival probability estimates at any time point, expanding the application of the customer value theory in practice. Instead, traditional customer value computation requires the starting time of the customer first purchase behavior to be identical.Thirdly, this dissertation uses the time series model to predict customer’s future purchases each time, which overcomes the shortcoming of the traditional method using the mean forecast. We find the model proposed in this dissertation not only enriches the traditional customer purchase prediction theory but can greatly improve the prediction quality and precision of customer purchases in enterprise’s application, so that the precision of customer value calculation is improved.Finally, after identifying the elements of customer value, measuring customer survival probability and predicting customer purchases, this dissertation adds other important customer value elements which are prices and costs. Thus the dynamic measurement model of enterprise customer value is constructed. Comparing with the traditional customer value calculation method, this model has greatly improved the prediction accuracy of customer survival probability and customer purchase. At the same time, to be different from the traditional method only to calculate the same type of customer value, this model can analyze the specific circumstances of the individual customer value, which makes the innovation in both theory and application. The enterprise customer dynamic value measurement model can help enterprises accurately analyze customer value constitution, providing data support for the fine management of customers, so as to improve the management and core competence of enterprises.