Customer Churn Prediction in Telecom Sector

Abstract

Purpose: The purpose of this research is to determine factors that lead to customer churn in the telecom sector and to understand how customer response can make an impact on customer churn in the future.

Methodology: It is an exploratory study involving respondents of different age groups and professions selected across India.

Findings: K-Means clustering is the best algorithm that suits the data. It generates 3 clusters and each cluster has a different range of quality index. Thus, companies can use the quality index as an indicator of the Net Promoter Score and identify detractors

Practical Implications: This will help companies identify the customers which might churn in the future. Thus, companies can come up with targeted marketing strategies to curb customer churning.

Originality/Value: This research will focus on unstructured features whereas most of the existing research is based on structured features

Article Classification: Research Paper