چكيده به لاتين
Abstract
Ever-increasing competition and rapid growth of businesses make organizations put identification of customers on the top of their marketing programs. Furthermore, segmentation and identification of customer groups, especially customers who create value, have been the priority of organizations. Today, organizations have used the customer lifetime value as an appropriate pattern to classify customers. In addition, data mining techniques have enabled organizations to analyze customers more deeply.
This research has been carried out in a software company to cluster customers based on factors of customer lifetime value and data mining techniques. Based on customer lifetime value model, LRFM, and its factors including length, recency, frequency and monetary, transactional data of 1865 customers has been gathered and analyzed through CRISP method. Moreover, four factors of customer lifetime value have been emphasized based on feature selection algorithm. They also have been prepared for clustering using quintile method. To determine the optimum number of cluster, silhouette and sum of square error index have been used. Additionally, k-means algorithm has been applied to help us classify the customers into four groups, namely high value loyal customers, uncertain lost customers, uncertain new customers, and high consumption cost customers. In the next step, customer lifetime value has been evaluated according to weights of factors and the average of normalized amounts in each cluster. Accordingly, the first cluster customers with the highest number and the highest customer lifetime value are the most valuable customers and the forth, third, and the second cluster customers are in the second, third, and forth positions respectively. Finally, the attributes of customers in each cluster have been analyzed and the marketing strategies have been proposed for each cluster.
Key Words: Customer lifetime value, LRFM model, Data mining, Clustering
