Customer Segmentation and Preference Modeling of Indonesian Mobile Telecommunication Industry: A Data Mining Approach

Abstract

The number of mobile internet users in Indonesia is overgrowing over the past years because of the emerging mobile telecommunication. However, lately, the growth is declining year by year. It makes each company struggles to increase their revenue. The continually increasing number of mobile internet users does not necessarily mean positive growth for the five major mobile telecommunication providers in Indonesia. Understanding customer segmentation and preferences are vital to increasing customers’ repurchase intention (i.e., data consumption per user) and customer satisfaction. In response, we present novel insights on mobile internet users’ segmentation and preference. We use online questionnaire data from more than 500 respondents and apply k-modes clustering and multiclass logistic regression analysis to establish the customer segmentation and preference models based on numerous attributes: internet speed, internet package price, internet quota size, call duration bonus, SMS bonus, network coverage, signal strength, customer care quality, service center availability, and other exclusive offers. We propose two dominant customers segments and characterize the main customers’ preference profile for each of the five major mobile telecommunication providers.