Customer churn analysis : A case study on the telecommunication industry of Thailand

Abstract

Customer churn creates a huge anxiety in highly competitive service sectors especially the telecommunications sector. The objective of this research was to develop a predictive churn model to predict the customers that will be to churn; this is the first step to construct a retention management plan. The dataset was extracted from the data warehouse of the mobile telecommunication company in Thailand. The system generated the customer list, to implement a retention campaign to manage the customers with tendency to leave the company. WEKA software was used to implement the followings techniques: C4.5 decision trees algorithm, the logistic regression algorithm and the neural network algorithm. The C4.5 algorithm of decision trees proved optimal among the models. The findings are unequivocally beneficial to industry and other partners.