Statistical Methods in Customer Relationship Management

V. Kumar, J. Andrew Petersen

• Publisher: Wiley
• Year: 2012
• Print ISBN: 9781119993209
• eBook ISBN: 9781118349199
• Edition: 1
• Topic: Mathematics
• Subtopic: Probability & Statistics
• Index: Mathematics

Statistical Methods in Customer Relationship Management focuses on the quantitative and modeling aspects of customer management strategies that lead to future firm profitability, with emphasis on developing an understanding of Customer Relationship Management (CRM) models as the guiding concept for profitable customer management. To understand and explore the functioning of CRM models, this book traces the management strategies throughout a customer's tenure with a firm. Furthermore, the book explores in detail CRM models for customer acquisition, customer retention, customer acquisition and retention, customer churn, and customer win back.

Statistical Methods in Customer Relationship Management:

• Provides an overview of a CRM system, introducing key concepts and metrics needed to understand and implement these models.
• Focuses on five CRM models: customer acquisition, customer retention, customer churn, and customer win back with supporting case studies.
• Explores each model in detail, from investigating the need for CRM models to looking at the future of the models.
• Presents models and concepts that span across the introductory, advanced, and specialist levels.

Academics and practitioners involved in the area of CRM as well as instructors of applied statistics and quantitative marketing courses will benefit from this book.

Chapter 1

Customer Relationship Management

1.1 Introduction

Henkel, a European multinational corporation that operates in three business areas (home care, personal care, and adhesive technologies), also operates in a highly competitive, fast-moving consumer-goods industry that also has a global outreach. Some of the key operational challenges include low product margins on its products and a lack of direct customer contact, among other region-specific challenges. Henkel has managed to overcome these challenges by implementing customer relationship management(CRM) practices. Management at Henkel realized the importance of identifying and understanding the needs of individual high-value customers, in order to target, establish, develop, and retain long-lasting relationships with customers. Through these practices, Henkel has actively pursued the development of strong relationships with its customers, and at the same time increased its profitability [1].

Just like Henkel, many corporations are increasingly adopting CRM as a means to forge their competitive advantage – the ability to understand individual customer needs, and therefore to manage their marketing efforts more efficiently. Such firms are also under tremendous pressure to adjust quickly to rapid changes in the marketplace with regard to the customer, technology and marketing functions. Customers are becoming not only more value-conscious, but also less loyal and less tolerant of low service levels. Consequently, markets are becoming more fragmented, making differentiation more difficult and competition more intense. These changes are driving companies to be customer-centric, and shifting their marketing functions from product-based to customer-based ones. At the same time, the exponential growth in data storage technology has made it possible for firms to process a much more substantial amount of customer-level information. All of these changes have had a significant influence on the rapid growth, increasing the awareness and adoption of CRM worldwide.

While most firms recognize the benefits of adopting CRM practices, not all firms have been successful in their CRM implementations. We believe that having the right approach to CRM planning is critical to a firm's success. Over the years, while technology has played a key role in the success of CRM implementation, it is but one component of CRM implementation. An important part of CRM is identifying the different types of customers and then developing specific strategies for interacting with each customer. Examples of such strategies are developing better relationships with profitable customers, not loyal customers. That means locating and attracting customers who will be profitable, and finding appropriate strategies for unprofitable customers, which could mean eventually terminating the relationship with customers who are causing the firm to lose money.

In this book, we discuss CRM and its related strategies from a modeling perspective, with a specific focus on methodologies that can be used to obtain insights on customer metrics within a company's own customer database. To help understand the discussion on CRM models and methodologies, we start this first chapter by providing an overview of CRM and its components. First, we present a formal definition of CRM and discuss the relevant concepts in CRM, such as customer value and customer databases. Next, we explain what is required to implement CRM strategies from a manager's perspective. In essence, managers ultimately want to evaluate their firms' marketing performances based on quantifiable indicators called customer metrics. Managers usually have to go through several data-processing steps before being able to generate the customer metrics they want. In this section, we focus on the database sources, the impact of technology on the implementation process of CRM, as well as providing a list of common customer metrics used to evaluate managerial performance. Finally, we introduce the role of analytical methodologies to help managers with the necessary levers/drivers to maximize the performance metrics.

1.2 What is CRM?

As the conceptualizations of CRM have evolved significantly, there are various definitions of CRM depending on the perspectives looked at. An important concept in CRM is customer value. Customer value is essentially the financial value of the customer relationship to the firm. It can be expressed in terms of contribution margin or net profit. Customer value is widely used by firms to evaluate their marketing efforts. However, it is a general term which does not refer to any specific time or duration. A better term that gives managers an idea of how the value of a client has evolved over time is customer lifetime value (CLV). CLV refers to the net economic value of a customer to a firm over his/her entire lifetime (three years in most cases – but it can depend on the length of the average purchase cycle of customers) with the company [2].

By definition, CRM refers to the practice of collecting, storing, and analyzing customer-level information, and incorporating the results into the decision-making process of a firm [3]. This also involves automating, enhancing, and integrating core business processes such as production, operations, sales, marketing, and finance, among others. The power of CRM lies in its adaptability to further the performance of any individual activity of the business, or even the entire business as a whole.

Companies are now beginning to design and implement CRM initiatives from a CLV perspective so that it becomes easier for them to measure the impact of CRM activities. Apart from evaluating the value of customers, such an approach to CRM activities also provides a basis for the competitive advantage of a firm: the customer-centric organization. By improving customer satisfaction, and customer loyalty, CRM helps firms acquire and retain profitable customers, and reactivate dormant customers. The ultimate goal of CRM is to maximize the lifetime value of each individual customer to the firm, thereby increasing firm profitability. In this regard, most of the CRM initiatives can be attributed to one of the following four categories:

1. Customer acquisition. Customer acquisition is the process of acquiring new customers, the foundation step of the whole CRM process.

2. Customer retention. Customer retention is the process of keeping and developing relationships with the customers after the company acquires them.

3. Customer churn. Customer churn, sometimes referred to as customer attrition, is the process of managing the rate of existing customers leaving a firm.

4. Customer win-back. Customer win-back is the process of reacquiring the customers that have left a firm through customer churn.

Now that we have reviewed the reason for adopting CRM practices, in the next section we will discuss the essential inputs needed for any CRM implementation.

1.3 What is Needed to Implement CRM Strategies?

The implementation of a CRM strategy is an ongoing process of developing and executing a series of small projects aimed at satisfying the business needs and enhancing the value proposition to customers. In this section, we focus on three essential ingredients needed to implement CRM strategies from a modeling perspective: database, technology, and metrics.

1.3.1 Database

The database is the core of any CRM planning. Companies gather information to store, analyze, and make marketing decisions based on the results of data analysis. This section provides a basic overview of the categories of databases and the sources from which data can be collected.

1.3.1.1 Categories of Databases

There are several types of databases and various ways to categorize them. This can be done according to firms' main business function, information contents, underlying marketing activities, or database technology. As the focus of this book is on data modeling, we look at the following two types of databases in detail: the transaction-related database and the customer database.

• Transaction-related database. This database refers to all the information associated with the transactions that customers have made. Examples of this type of information are:

– What transactions have the customers conducted?

– What type of product was purchased?

– How frequent is this type of product purchased by the customer?

– How much was spent in the transaction?

• Customer database. This database is essentially a collection of information about a firm's customers. In general, the following information may be included in customer databases:

– Basic information: name, address, ZIP (post) code, and telephone number.

– Demographic information: age, gender, marital status, education, number of people ...

Table of contents

• Cover
• Title Page
• Copyright
• Dedication
• Preface
• Chapter 1: Customer Relationship Management
• Chapter 2: CRM in Action
• Chapter 3: Customer Acquisition
• Chapter 4: Customer Retention
• Chapter 5: Balancing Acquisition and Retention
• Chapter 6: Customer Churn
• Chapter 7: Customer Win-back
• Chapter 8: Implementing CRM Models
• Chapter 9: The Future of CRM
• Appendix A: Maximum Likelihood Estimation
• Appendix B: Log-linear Model—An Introduction
• Appendix C: Vector Autoregression Modeling
• Appendix D: Accelerated Lifetime Model
• Appendix E: Type-1 Tobit Model
• Appendix F: Multinomial Logit Model
• Appendix G: Survival Analysis – An Introduction
• Appendix H: Discrete-Time Hazard
• Appendix I: Proportional Hazards Model
• Appendix J: Random Intercept Model
• Appendix K: Poisson Regression Model
• Appendix L: Negative Binomial Regression
• Appendix M: Estimation of Tobit Model with Selection
• Index