Modelling customer satisfaction for use in a lifetime value model

Smith, Andrew (2018) Modelling customer satisfaction for use in a lifetime value model. MSc(R) thesis, University of Glasgow.

Due to Embargo and/or Third Party Copyright restrictions, this thesis is not available in this service.


In a retail banking context, understanding the long term value of a customer can inform and steer the direction of banking strategies, and can ultimately add value to a bank. With the belief that customer satisfaction plays a role in the long term value of a customer, this thesis aims to explore the possibility of incorporating a customer satisfaction factor in a customer lifetime value model with the overall aim of improving estimates of long term customer value and ensuring the model includes all relevant variables. With no bankwide customer satisfaction metric currently available, the aim of the thesis is to develop a model which produces a customer satisfaction score for all customers in the bank and incorporate this new metric into the existing customer lifetime value model. Chapter 1 introduces a number of studies that investigate customer lifetime value models. Studies on the value of
customer satisfaction are also explored, motivating the main investigation of this thesis. Chapter 2 introduces the methodology used throughout the thesis. In particular, binary and multinomial logistic regression are explored and used in the modelling of the customer satisfaction metric. Chapter 3 introduces the data considered in the modelling of customer satisfaction. A binary logistic regression model and a proportional odds logistic regression model are developed and tested. Advantages and limitations of both models are explored and arguments for which to be used in the implementation of customer satisfaction in the customer lifetime value model are presented. In chapter 4, methods of incorporating the customer satisfaction model developed in chapter 3 are explored briefly and the new
customer lifetime value model is developed. Comparisons of the new and old model are presented to illustrate the benefits of including a customer satisfaction factor in the model. Finally, chapter 5 provides a summary of the previous and discusses the limitations and issues met during this study.

Item Type: Thesis (MSc(R))
Qualification Level: Masters
Additional Information: Due to confidentiality issues the full text of this thesis cannot be made available online. Access to the printed version is available once any embargo period has expired.
Keywords: Logistic regression, customer lifetime value.
Subjects: H Social Sciences > HA Statistics
Colleges/Schools: College of Science and Engineering > School of Mathematics and Statistics
Supervisor's Name: McColl, Professor John
Date of Award: 2018
Embargo Date: 23 April 2023
Depositing User: Mr Andrew Smith
Unique ID: glathesis:2018-8996
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 24 Apr 2018 13:26
Last Modified: 22 May 2018 06:47

Actions (login required)

View Item View Item