Factors Affecting Customer Churn and Customer Churn Prediction Model using Machine Learning Case Study

  • Yousef Mohammed Rashed Alaskar Alnuaimi

Student thesis: Doctoral ThesisDoctor of Professional Studies

Abstract

This thesis focuses on examining customer churn in Etisalat, a telecommunications operator in the UAE, and explores the relationship between customer satisfaction, brand variables, customer loyalty, and customer churn. The primary objective is to identify the factors influencing customer churn and investigate how customer satisfaction and brand contribute to customer loyalty to reduce churn. Additionally, the research proposes the utilization of a machine learning algorithm to predict customer churn specifically in Etisalat. The study's findings indicate significant and positive associations between customer satisfaction, brand, customer loyalty, and customer churn. It is revealed that pricing perceptions and meeting customer expectations are crucial for customer satisfaction, as they create value for the services provided. Moreover, the study suggests that Etisalat should focus on organizational improvements and offer a dynamic platform that emphasizes creativity and innovation in personalized services to meet customer demands. The main contributions of this research are twofold. Firstly, a customized customer churn framework is proposed for Etisalat, providing a structured approach to address customer churn. Secondly, a novel methodology is introduced to develop a customer churn prediction model specifically tailored for Etisalat. To achieve these contributions, the thesis employs a combination of expert interviews to identify factors influencing customer churn and subsequently develop the customer churn framework. This framework is then validated through questionnaire analysis to confirm the factors influencing customer churn and to gain insights into the data sources available within Etisalat. The research concludes that the random forest algorithm exhibits the highest accuracy metrics for building customer churn predictions using the Etisalat dataset.
Date of Award2023
Original languageEnglish
Awarding Institution
  • Aberystwyth University
SupervisorJulie Jones (Supervisor) & Richard Jensen (Supervisor)

Keywords

  • customer churn
  • customer loyalty
  • customer churn prediction
  • machine learning
  • Artificial Intelligence
  • random forest
  • accuracy metrics
  • service personalization
  • Etisalat

Cite this

'