Customer Segmentation Of China Telecom

The market situation of mobile telecommunications and IPTV is constantly changing. And the competition is becoming extremely fierce. As the user’s growth slow down, how to develop the value of existing users is a big issue faced by the enterprise. In the big data times, customer segmentation and precision marketing strategies which is based on the data mining technology seem to be a way out. Yet the China Telecom company still use the old methods to segment its customers and do the marketing. The dimensions are limited and rough. Valuable data resources have not been fully used, which suffered the shackles in the efficiency and benefit of operation.After reviewing the related literatures of market segmentation and data mining, with CRISP-DM data mining process as reference, this paper approached the study in the following steps:business understanding analysis, sample processing, variable selection and data preparation, research methods selection and model establishment, data analysis and discussion. Faced with the problems and shortcomings of the telecom operator’s current segmentation strategy, this paper proposed customer segmentation framework based on the real telecom users’data. This framework studied the customer segmentation from the perspectives of customer value and behavior, customer demographic characteristics and customer interests preference, which provided decision-making basis for the enterprise to carry out the precision marketing strategies of mobile communication services and IPTV service operation.The main research contents are as follows:1. According to the customer value theory and mobile telecommunications business characteristics, the author constructed the customer value and behavioral characteristics of three aspects which included the customer profit contribution, loyalty and usage behavior. K-Means cluster analysis method was used to segment the customers into five sub groups and they were named as potential type group, flow preference type group, traditional type group, business type group and low value type group. When the model was established, priority attention was given to the traffic related variables.2. The chi square test and correspondence analysis method were conducted on the customer’s gender, age and other demographic characteristics, which further characterized the subdivision market to profile customer images and the consumption level. This was beneficial for enterprise to manage customer relationship and acquire new customers.3. Based on the customers access data of the ITV service, binary logistic regression analysis was used to study the different preference of customers of different gender, age, consumption levels in 8 categories of interests, which provided decision reference for the introduction and making of the ITV programs resources. And also knowing the customer created the possibility and condition for the ITV business oriented marketing strategy.This article studied the customer segmentation of China Telecom Company based on the data mining technology, which was a meaningful practice. In the background of the era of big data, data mining technology will have a wide range of applications in the telecommunications industry.