Customer Accounting

SPRINGER BRIEFS IN ACCOUNTING

Massimiliano Bonacchi Paolo Perego

Customer Accounting Creating Value with Customer Analytics

1 23

SpringerBriefs in Accounting Series editors Peter Schuster, Schmalkalden, Germany Robert Luther, Bristol, UK

More information about this series at http://www.springer.com/series/11900

Massimiliano Bonacchi • Paolo Perego

Customer Accounting Creating Value with Customer Analytics

Massimiliano Bonacchi Faculty of Economics and Management Free University of Bozen-Bolzano Bolzano, Italy

Paolo Perego Faculty of Economics and Management Free University of Bozen-Bolzano Bolzano, Italy

ISSN 2196-7873 ISSN 2196-7881 (electronic) SpringerBriefs in Accounting ISBN 978-3-030-01970-9 ISBN 978-3-030-01971-6 (eBook) https://doi.org/10.1007/978-3-030-01971-6 Library of Congress Control Number: 2018957984 © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Foreword

It is self-evident that customers are essential to business enterprises. This was already the case when the first barter transaction in history was concluded. What is relatively new is the ability of many businesses now, particularly those who charge a subscription fee for their services, to track their customers, identify their preferences, customize products to people’s tastes, and learn about their experiences and satisfaction level. This wealth of information derived from the footprints of customers of Internet service providers, media and entertainment firms, and insurance companies, among other sectors, radically transformed corporate customer management. But this transformation is a work in process with lots of unanswered questions for both corporate managers and their shareholders. That is the reason this book on customer accounting is such a welcome addition to the literature of management, marketing, operations research, and of course accounting. The core of the book is the introduction of the highly useful concept of a company’s lifetime value of customers, which for many enterprises is their largest and most consequential, value-creating asset. The computation of customers’ value (customer equity) and the various uses of this important metric in management and capital market investment decisions are clearly discussed in this book. The many real-life examples provided by the authors, both experts on the subject, demonstrate the power of this new metric and make the book fun to read. Customer value and the related measures introduced and demonstrated by the authors are particularly important to investors, given the sharp decline in the usefulness and relevance of the traditional accounting and financial variables used in investment analysis. In this book, both managers and investors will find new measures and methods to manage customers and enhance corporate value. Who will benefit from this book? Corporate executives responsible for the management of their customers to create corporate value and also CFOs; financial analysts and investors striving to value business enterprises and frustrated with the traditional, failed financial measures based on accounting asset and earnings; and

v

vi

Foreword

last but not least, business students, both at the undergraduate and graduate (MBA) levels, will benefit considerably from this book in finance, marketing, and accounting courses. Philip Bardes Professor of Accounting and Finance NYU Stern School of Business New York, NY, USA

Baruch Lev

Contents

1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Customer-Centricity in a Fast-Evolving Landscape . . . . . . . . . . . . 1.2 Motivation and Objectives of This Book . . . . . . . . . . . . . . . . . . . 1.3 Theoretical Framework: Organizational Architecture . . . . . . . . . . . 1.4 Outline of This Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. 1 . 1 . 4 . 5 . 8 . 10

2

Customer Analytics: Definitions, Measurement and Models . . . . . . . 2.1 Customer Analytics: Definitions of CP, CLV and CE . . . . . . . . . . 2.2 CLV Formulae: Sources and Variations . . . . . . . . . . . . . . . . . . . . 2.3 Applications of CLV in Subscription-Based Business Settings . . . . 2.4 CLV Scorecard and Cohort Analysis: An Application in an SBE . . 2.4.1 The CLV Scorecard as a Performance Measurement System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.2 Benefits of CLV Scorecard . . . . . . . . . . . . . . . . . . . . . . . . 2.4.3 CLV Cohort Analysis: Rationale . . . . . . . . . . . . . . . . . . . . 2.4.4 CLV Cohort Analysis: A Practical Illustration . . . . . . . . . . 2.5 Conclusions and Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . .

13 13 16 17 19

. . . . . .

20 24 25 28 33 33

. . . . . . . . .

37 37 41 42 42 43 45 47 48

3

Customer Analytics for Internal Decision-Making and Control . . . . 3.1 Review of Accounting and Marketing Literature . . . . . . . . . . . . . . 3.2 Evaluation of the Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 A Case Study on the Adoption of Customer Analytics . . . . . . . . . 3.3.1 Case Background and Research Methodology . . . . . . . . . . 3.3.2 Organizational Structure . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.3 The Performance Measurement System . . . . . . . . . . . . . . . 3.3.4 The Reward System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.5 Conclusions and Implications from the Case Study . . . . . . 3.4 An Exploratory Cross-Sectional Survey on the Adoption of Customer Analytics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. 50 vii

viii

4

5

Contents

3.4.1 Sample and Data Collection . . . . . . . . . . . . . . . . . . . . . . . 3.4.2 Descriptive Statistics and Univariate Analysis . . . . . . . . . . 3.4.3 Multivariate Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.4 Conclusions and Implications from the Survey . . . . . . . . . . Appendix Chapter 3: Questionnaire . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

Customer Equity for External Reporting and Valuation . . . . . . . . . . 4.1 Customers as the Most Valuable (Intangible) Asset . . . . . . . . . . . . 4.2 Customer Franchise Is Missing in IFRS/US GAAP Financial Statements: How to Value It? . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Describing SBEs Business Model Using Customer Metrics . . . . . . 4.4 Valuing SBEs Using Publicly Disclosed Customer Metrics: A Parsimonious Model to Estimate Customer Equity . . . . . . . . . . 4.5 Customer Equity and Stock Returns: Empirical Evidence . . . . . . . 4.6 Beyond GAAP: Customer Metrics Reporting . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. 67 . 67

Conclusions and Trends to Look Forward . . . . . . . . . . . . . . . . . . . . 5.1 Looking Back and Looking Ahead . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Linking Online with Offline Commerce . . . . . . . . . . . . . . . . . . . . 5.3 Enhanced Forms of Corporate Non–financial Reporting . . . . . . . . 5.4 The Rising Impact of Artificial Intelligence on Modeling Customer Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

53 54 56 58 59 64

. 68 . 69 . . . .

71 77 79 81

. . . .

83 83 84 85

. 86 . 87

Chapter 1

Introduction

The primary function of a business is to serve the customer and the primary goal of your business is to create customers. —Peter Drucker

1.1

Customer-Centricity in a Fast-Evolving Landscape

During the Nineties, the business environment was affected by technological advances resulting from “combinatorial innovations” triggered by liberalization of the telecommunication industry and the Internet (Varian et al. 2004). Those innovations created the basis for many of the innovative services introduced over the past decade, such as cell phones, satellite radio, cable TV, financial services (e.g. direct banking) and internet services (games, music, entertainment, etc.) (Libai et al. 2009). At the same time, the information technology (IT) revolution introduced extraordinary improvements in methods of collecting, storing, analyzing, and transmitting huge amounts of information (Varian 2006, 2009). Firms realized that this presented great opportunities to invest in IT to manage customer relationships, since data could reveal actual customer preferences rather than merely their intentions, making sampling unnecessary since information on customer behavior became available for the entire population of customers (Gupta et al. 2006). For instance, advertising models evolved from a focus on “brand awareness” to “direct and measurable” customer acquisitions (Economist 2006a, b, 2007; Epstein 2007; Epstein and Yuthas 2007; French 2007). Unlike television advertising, Internet advertisers paid only when a user clicked through to their website, gaining a reliable measurement of customer acquisition costs (Court 2005; Laffey 2007; Mulhern 2009). In recent years, firms have continued witnessing a period of transformative developments that emphasize the central role of customers in all industries. We

© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2019 M. Bonacchi, P. Perego, Customer Accounting, SpringerBriefs in Accounting, https://doi.org/10.1007/978-3-030-01971-6_1

1

2

1 Introduction

provide below a few examples showing customer power and the trends shaping the future of marketing decisions into the next decade: • Half of the firms listed in the DAX 30 and DJIA 30 explicitly mention in their mission statements or company strategies the notion of value creation for customers (Kumar and Reinartz 2016) • According to a 2017 Forrester report, we are now fully within the ‘Age of the Customer’, in which newly empowered customers place elevated expectations on every interaction they have with brands. • The 2017 Salesforce report “State of the Connected Customer”, revealed that 70% of consumers now believe technology has made it easier than ever to switch brands to find experiences that matches their expectations. • The results of a 2016 global survey by Forbes Insights showed that firms who increased their spending on retention in the last 1–3 years had nearly a 200% higher likelihood of increasing their market share in the last year compared to those spending more on acquisition. • An online survey by TECH at Harvard revealed that in 2016, increasing customer experience received the highest priority among 908 IT decision makers at global firms. These latest examples clearly indicate that consumers hold far more power than ever before in today’s ultracompetitive and fast evolving business landscape. The transition from a product-centric, transaction-focused business model to a more relationship-oriented or customer-centric view appears as a necessary condition to sustain long-term business performance (Sheth et al. 2000; Shah et al. 2006; Ramani and Kumar 2008). This transition necessitates a radical shift that aggressively relies on interaction response capacity and customer value management (Kumar et al. 2008; Ramani and Kumar 2008). Interaction response capacity is the degree to which a firm can provide successful products and services by exploiting the feedback of a specific customer. At the same time, through customer value management a firm can define and dynamically measure individual customer data and use this information as a guiding principle for tactical and strategic resource allocation decisions. Customer-centric firms thus understand not only what the customer values but, more importantly, the value the customer adds to their bottom line. Customercentricity implies a carefully defined and quantified customer segmentation strategy in which a firm’s operations aim at delivering the greatest value to the best customers for the least cost (Sheth et al. 2000; Shah et al. 2006; Ramani and Kumar 2008; Libai et al. 2009; Fader 2012). Shah et al. (2006) and Fader (2012) emphasize that customer centricity is a necessary condition for twenty-first-century firms that need to address key strategic issues (Kumar and Rajan 2012; Cokins 2015) such as: • Do we push for volume or for margin with a specific customer? How many products can we sell to a specific customer? • How can we develop profitable relationships over a long time span?

1.1 Customer-Centricity in a Fast-Evolving Landscape

3

Table 1.1 Comparison of the product-centric and customer-centric approach (source: Bonacchi and Perego 2012) Basic philosophy Business orientation Product positioning and selling approach Organizational focus

Selling approach

Product-centric approach Sell products Transaction oriented

Customer-centric approach Serve customers Relationship oriented

Highlight product features and advantages

Highlight product’s benefits in terms of meeting individual customer needs Externally focused. Customer relationship development, profitability through customer loyalty. Employees are customer advocates How many products can we sell to this customer?

Internally focused. New product development, new account development, market share growth, and customer trelations are issues for the marketing department How many customers can we sell this product to?

Source: Authors’ elaboration adapted from Kumar (2008a); Ramani and Kumar (2008); Shah et al. (2006)

• How can we identify profitable customer segments and business processes with higher productivity? • Can we influence our customers to alter their behavior to interact differently (and more profitably) with us? In Table 1.1, we summarize the main differences between product-centric and customer-centric orientations after a review of several sources in marketing and management literature (Sheth et al. 2000; Egol et al. 2004; Shah et al. 2006; Kumar 2008a, b; Ryals 2008). In this context, disruptive developments in digital technology, Internet of Things (IoT), sensor data and the social media have accelerated the shift towards customercentricity on an unprecedented scale and pace. In a short time, firms in several industries have started to collect very large quantities of data from their own operations, supply chains, production processes, and customer interactions. The scale and diversity of customer data provide Internet-based firms such as Facebook, Google, Amazon and Netflix rich new sources of business insights, allowing firms to understand and engage with customers in novel ways to both better serve them and maximize profitability. Beyond a basic transaction history, companies currently track marketing interactions, clicks, web or mobile navigation patterns, and online and offline behaviors, on their own platforms or on social media. They also receive large amounts of data from connected objects owned by customers (e.g. mobile phones, tablets, tracking devices). Traditional databases cannot handle such volumes of information and variety of formats, but this is where ‘Big Data’ solutions step in. We are currently witnessing a shift in the breadth and depth of firms’ customer accounting systems. In this book, we use the label customer analytics to broadly denote the metrics, processes and technologies that provide

4

1 Introduction

firms the insight into customers necessary to deliver offers that are anticipated, relevant and timely. Numerous examples are emerging of the potential impact of customer analytics in traditional companies: Tesco and IBM, among other large firms, make increasing use of Big Data to deliver contextual insights about purchase behaviors and marketing response. Several firms are also spinning up new investigative computing or data science practices rooted in artificial intelligence (AI), deep learning and other highly dynamic and multidimensional forms of advanced analytics. Half a decade ago, none of these disrupting technologies were anywhere close to being used in daily practices. In the closing chapter of this book, we will point at these developments further.

1.2

Motivation and Objectives of This Book

An increasing number of academic papers in marketing have examined how a customer-centric focus can provide competitive advantages and emphasized the benefits of providing differentially tailored responses to marketing initiatives, such that the contribution from each customer to overall profitability is maximized (e.g., Verhoef and Lemon 2013). The marketing literature has also started to highlight the organizational steps and barriers critical to initiate and sustain customer centricity (Shah et al. 2006; Kumar et al. 2008). However, there is a dearth of knowledge about the business processes with which CFOs and management accountants interact and coordinate with other CMOs and marketing managers to monitor the attraction, conversion and retention of customers through marketing campaigns and reliance on customer data. Interested readers should refer to recent reviews of the literature dedicated to the marketing-accounting interface (Gleaves et al. 2008; Roslender and Wilson 2008; Kraus et al. 2015). The apparent disjunction between these two core functions emerges clearly in the developments of the accounting literature on customer accounting, defined as “all accounting techniques that measure individual customer’s and/or customer segments’ contributions to firm profitability” (Holm et al. 2016). On one hand, accounting textbooks seem to cover traditional techniques of customer profitability analysis and only marginally treat contemporary topics in customer value management (Gleaves et al. 2008; Bates and Whittington 2009). On the other hand, the academic literature on customer accounting is still embryonic when compared to marketing, pointing at a relevant gap between current practice and theory-driven research in this rapidly changing business area (Guilding and McManus 2002; McManus and Guilding 2008). We will provide a review of this literature in subsequent chapters. In sum, whilst the volume and complexity of customer data today require sophisticated analytic methods that go beyond traditional measurement and reporting, accounting research and accounting textbook knowledge on these topics lag behind. In this book, we contribute to filling this void by examining fundamental issues, challenges and opportunities that typically a CFO or a manager in the accounting & finance function would face when dealing with customer-centricity and the role of

1.3 Theoretical Framework: Organizational Architecture

5

customer analytics in extracting valuable business insights at all stages of the customer lifecycle. To logically map and structure the various implications, we draw upon a theoretical framework that allows an analysis of the main levers involved in the implementation of a customer-centric strategy. Such a conceptualization, labeled ‘organizational architecture’, relies on research conducted in organizational economics and management accounting (Wruck and Jensen 1994; Brickley et al. 1995; Ittner and Larcker 2001; Brickley et al. 2004; Brickley et al. 2009) and has the advantage of being broadly generalizable to several business contexts and industries. In the next section, we provide a definition and a few examples of the three components of organizational architecture relevant for customer-centricity.

1.3

Theoretical Framework: Organizational Architecture

The organizational architecture framework provides the infrastructure with which business processes are deployed and ensures that the organization’s core capabilities are realized across business processes. A key issue is ensuring that decision makers not only have the relevant (i.e. accurate and useful) information required to make decisions, but that they must also be provided with the appropriate incentives to use that information to achieve organizational objectives. Thus, the fundamental tenet behind organizational architecture is that value creation depends on coherence among three primary organizational components, namely, the assignment of decision rights, the choice of performance measures, and the design of compensation and incentive systems, as depicted in Fig. 1.1. The extent to which top management chooses how to design an organizational architecture differs greatly among firms. Such differences are not random but vary in systematic ways with underlying characteristics of the firms themselves. Drawing on the contingency theory of organizations in management (Brickley et al. 1995, 2004; Brickley et al. 2009) and management accounting research (e.g. Gong and Ferreira 2014), consistent relationships and alignment among the three components should ensure the most effective fit with a firm’s business environment and inherent strategy (Ittner and Larcker 1997; Langfield-Smith 1997; Chenhall 2003; Widener et al. 2008; Lee and Yang 2011; Grabner and Moers 2013). Kaplan and Norton (2004) state that “unless an organization links its strategy to its governance and operational processes, it won’t be able to sustain its success”. Put simply, failure to properly design and incorporate the three levers (hence the ‘three-legged stool’ label of the model) in internal decision-making and control systems, is likely reflected in lower organizational performance. Previous management accounting studies recognize that these three key organizational elements are jointly determined and complementary (Nagar 2002; Abernethy et al. 2004; Widener et al. 2008). The role of strategy is indeed a crucial part of a contingency framework, although, as noted by Chenhall (2003: 150), “it is not an element of the context, it is a means whereby managers can influence the nature of the external environment, the technologies of the organization, the structural arrangement, the control culture and the

6

1 Introduction

Contingent Variables Technology − Fast pace of change − Big Data, AI, Social media

Markets

Regulation

− Rising power of the Customer − Diffusion of services

− Deregulation − Globalization

Business Strategy Customer-Centric Strategy

Organizational Architecture Allocation of decision rights

Performance measures

Incentives

Financial performance Fig. 1.1 Conceptual framework: organizational architecture (source: Adapted from Brickley et al. 1995)

management control system.” The marketing literature similarly sees in customercentric strategies a solution to adapt to the new competitive environment characterized by rapid changes in technology, market forces and regulation. In particular, Shah et al. (2006), Fader (2012) and Cokins (2015) emphasize that customer centricity is a necessary condition for twenty-first-century firms that need to address key strategic issues (Kumar and Rajan 2012), such as: • How many products can we sell to the customer? • How can we develop profitable relationships? • How can we identify profitable customer segments? Following this rationale, customer-centric firms should deliberately design and develop features in their organizational architecture that differentiate them from those typical of traditional product-centric firms. The performance measurement system (how a firm’s performance is conceptualized, tracked and evaluated) involves the choice of performance measures to coordinate the efforts of decision makers, to provide feedback to top management for evaluating progress toward strategic objectives and to employees for learning purposes. A critical component of the performance measurement system for

1.3 Theoretical Framework: Organizational Architecture

7

customer-centric organizations is determining how to collect customer-related data to provide a unified, comprehensive, and organization-wide view of a firm’s customer base, irrespective of the products purchased or channels employed by the customer. This entails a substantial IT-related investment commitment to set up an information infrastructure for collecting, tracking, and integrating data at the individual-customer and transaction level. Jayachandran et al. (2005) specified how several information system-related activities can be integrated and allow customercentric firms to successfully build a viable relationship with their customers. Such an integrated database is then made accessible to those responsible for managing the customer relationship to analyze past performance with the goal of understanding the “why” behind customer behavior (Shah et al. 2006). One of the reasons many organizations struggle to deliver value from customer data is the excessive number of possible integration points among the number of different data management and analysis technologies. In recent years, the advent of disrupting digital technologies and Big Data has accelerated and opened up a variety of technical solutions to measure customer-related performance data. Several firms today have multiple data warehouses, data marts, data caches, and operational data stores aimed at a timely collection of customer information. The allocation of decision-making authority (that is, who in the organization is given the authority to make decisions) reflects the contention that delegation and empowering people with specific knowledge is a critical determinant of organizational success. A typical product-centric company that is organized around functional silos defined by product types is not conducive to customer centricity, as each product/sales manager may end up pushing different product offerings to the same customer without first determining what the customer’s true needs are. On the contrary, it can be posited that a customer-centric organization has its functional activities integrated and aligned to successfully serve its customers. The first stage of this organizational realignment is the emergence of lateral coordinating activities that aim to overcome the traditional deficiencies of products or functional silos. This may be achieved by setting up a horizontal organization structure, in which information flows are readily shared among team members (Shah et al. 2006). In this context, ensuring an interface between the Marketing and the Accounting and Finance (A&F) functions becomes crucial. For example, more than a dozen Fortune 1000 firms, such as Coca Cola, Hershey, Intel, HP, and JD Edwards, have created a specialized function, labeled as Chief Customer Officer, to acknowledge the importance of customer-centricity-related issues in the boardroom (Shah et al. 2006; Rust et al. 2009). Wells Fargo has successfully realigned its organization by creating a two-tiered sales structure whereby a relationship manager ensures an interaction orientation (external focus) and a product specialist provides the technical input for product development (internal focus). In this context, the interface between Marketing and A&F is crucial to provide decision makers with relevant information on customer profitability. The third element of an organizational architecture refers to the formal incentive and compensation systems (how a firm rewards its management for success). Incentive systems seek to motivate managers and employees to be more productive,

8

1 Introduction

to focus on organizational objectives and to learn. A broad consensus from a variety of disciplines concludes that the presence of incentives influences behavior. With regard to customer-centric organizations, firms should include selected customer metrics among the key performance indicators that are regularly reported to the top management and the board. Moreover, it is essential to synchronize incentive and reward systems by linking the formal evaluation of employees with customer-centric metrics and targets. For instance, sales/account managers could be rewarded for increasing customer equity, while relationship managers could be incentivized to extend the profitable lifetime duration of the customers. For example, Texas Instruments is reported to have successfully introduced a reward system that includes three marketing metrics tracking the following dimensions: marketing gains for three consecutive years, efficient and timely services and better understanding of customers (Kumar 2008b). In sum, a customer-centric strategy should shape firms in ways that radically deviate from transaction- and product-centric business models. The specific architecture choices in the three dimensions of organizational design likely have an impact on the profitability of the firm. Incorporating several customer data sources into customer analytics, properly allocating decision-rights to move quickly from data to decision, and aligning incentives to avoid dysfunctional triggers remain fundamentally difficult tasks contingent upon the business environment, the industry and the technological developments in which a firm operates.

1.4

Outline of This Book

We acknowledge that the organizational architecture (similarly to other organizational design frameworks) is an abstraction of the complex interdependencies, simultaneous choices, and feedback loops found in practice. However, it provides a useful framework for categorizing the main organizational dimensions and business processes involved in customer-centric firms and the main effects thereof. In this book, we will therefore rely upon the organizational architecture to structure our analysis along two lines: • The current state-of-the-art academic literature: our focus will predominantly be on accounting studies, although we will also highlight main trends and findings in the marketing literature; • Practical applications or field studies that serve the purpose of illustrating with concrete examples and research insights how customer accounting can influence organizations interested to shift towards customer-centricity. We will initially point to the recent developments in the dimension of performance measurement as a foundational element of the organizational architecture required to pursue a specific business strategy—in our setting a customer-centric strategy. The label and contents we adopt will be customer analytics to more appropriately convey the combination of the wide range of data sources and customer

1.4 Outline of This Book

9

metrics with the analytic capabilities used to engage with customers. To determine the relative analytics proficiency of an organization, MIT Sloan Management Review developed the Analytics Core Index based on the organization’s core analytics capabilities in: • ingesting data (capturing, aggregating, and integrating data); • analyzing (descriptive analytics, predictive analytics, and prescriptive analytics); • applying insights (disseminating data insights and incorporating them into automated processes). Our aim is not to dissect every analytic capability; we will focus instead on essential features that are more relevant for the typical challenges faced by a CFOs and CMOs in developing a suitable set of customer analytics. Chapter 2 provides definitions of the most widely diffused customer metrics, namely Customer Profitability (CP), Customer Lifetime Value (CLV), Customer Equity (CE). We refer to the marketing literature that extensively covers these metrics and illustrate their interrelationships. We point at applications in business settings that have a contractual, subscription-based model and mention potential challenges to compute CLV in non-contractual settings. To illustrate the implementation and impact of customer metrics in a real-world context, we provide a case study focused on the computation of CLV in an Internet-based, subscription-based company. The case presents a simulation that applies cohort analysis in an attempt to fill the void between theoretical CLV models and its implementation in practice. The main rationale is to provide CFOs and CMOs a better understanding of new and latent customer preferences in a typical subscription-based business model by directly observing the customer’s purchase behavior and subsequently linking this data to estimate CLV and firm performance. In Chap. 3 we offer a critical evaluation of the literature in accounting that examined the role of customer metrics in internal decision-making and control purposes. We draw on the relationships theorized in the organizational architecture outlined in Chap. 1 to structure our selective review and emphasize key critical gaps in our knowledge, especially vis-à-vis extant developments in the marketing literature. The chapter then presents two empirical studies aimed at generating insights on the adoption of customer metrics for internal decision-making and control purposes. The first study is a qualitative case study conducted within a subscription-based enterprise (SBE). The second study reports a survey about the diffusion of customer metrics in a sample of SBEs. In combination, the empirical evidence highlights relevant take-away points and current challenges about the actual use of customer metrics in performance measurement and management control systems. Chapter 4 reiterates a common critique about current financial accounting models (e.g. IFRS/US GAAP), namely that they cannot capture Customer Franchise as key value creator intangible asset. We approach this issue by characterizing the business model of subscription-based enterprises (SBEs) that offer a for-fee-per-period access to products or services. Specifically, we show how to aggregate publicly available data into a measure of a firm’s Customer Equity value, which incorporates the major value drivers of SBEs, and empirically examine its properties. We build on the idea

10

1 Introduction

that the acquisition and retention of profitable customers is crucial for SBEs to identify the fundamental elements of their business model (e.g., customer base, revenues and service cost per user, and customer turnover). We further argue that companies should disclose in the Management Discussion and Analysis (MD&A) section of their annual report a set of customer metrics useful to investors, such as new subscriber acquisitions, revenue per subscriber, customer dropouts, and cost of customer acquisition. Chapter 5 concludes the book and provides a glimpse on managerial, technological and institutional trends that will likely affect the way customer metrics will be deployed to create business value in the next decade.

References Abernethy, M. A., Bouwens, J., & van Lent, L. (2004). Determinants of control system design in divisionalized firms. The Accounting Review, 79(3), 545–570. Bates, K., & Whittington, M. (2009). The customer is king. Enthroned or in exile? An analysis of the level of customer focus in leading management accounting textbooks. Accounting Education, 18(3), 291–317. Bonacchi, M., & Perego, P. (2012). Measuring and managing customer lifetime value: A CLV scorecard and cohort analysis in a subscription-based enterprise. Management Accounting Quarterly, 14(1), 27–39. Brickley, J., Smith, C., & Zimmerman, J. (1995). The economics of organizational architecture. Journal of Applied Corporate Finance, 8(2), 19–31. Brickley, J. A., Smith, C. W., & Zimmerman, J. L. (2004). Managerial economics and organizational architecture. Boston, MA: McGraw-Hill/Irwin. Brickley, J., Smith, C., Zimmerman, J., & Willett, J. (2009). Using organizational architecture to lead change. Journal of Applied Corporate Finance, 21(2), 58–66. Chenhall, R. H. (2003). Management control systems design within its organizational context: Findings from contingency-based research and directions for the future. Accounting Organizations and Society, 28(2–3), 127–168. Cokins, G. (2015, February). Measuring and managing customer profitability. Strategic Finance, 23–29. Court, D. C. (2005). Boosting returns on marketing investment. McKinsey Quarterly, (2), 36–47. Economist. (2006a). E-commerce. The Economist. Economist. (2006b, July 6). The ultimate marketing machine. The Economist. Economist. (2007). Word of mouse. The Economist, 77–78. Egol, M., Hyde, P., Ribeiro, F., & Tipping, A. (2004). The customer-centric organization: From pushing products to winning customers. McLean, VA: Booz Allen Hamilton. Epstein, M. J. (2007). Evaluating the effectiveness of internet marketing initiatives. In Management accounting guideline. Hamilton: The Society of Management Accountants of Canada, The American Institute of Certified Public Accountants and The Chartered Institute of Management Accountants. Epstein, M. J., & Yuthas, K. (2007). Managing customer value. In Management accounting guideline. Hamilton: The Society of Management Accountants of Canada, The American Institute of Certified Public Accountants and The Chartered Institute of Management Accountants. Fader, P. (2012). Customer centricity focus on the right customers for strategic advantage. Philadelphia, PA: Wharton Digital Press.

References

11

French, T. D. (2007). Confronting proliferation. . . in online media: An interview with Yahoo!’s senior marketer. McKinsey Quarterly, (3), 18–27. Gleaves, R., Burton, J., Kitshoff, J., Bates, K., & Whittington, M. (2008). Accounting is from Mars, marketing is from Venus: Establishing common ground for the concept of customer profitability. Journal of Marketing Management, 24(7–8), 825–845. Gong, M. Z., & Ferreira, A. (2014). Does consistency in management control systems design choices influence firm performance? An empirical analysis. Accounting and Business Research, 44(5), 497–522. Grabner, I., & Moers, F. (2013). Management control as a system or a package? Conceptual and empirical issues. Accounting, Organizations and Society, 38(6–7), 407–419. Guilding, C., & McManus, L. (2002). The incidence, perceived merit and antecedents of customer accounting: An exploratory note. Accounting, Organizations and Society, 27(1–2), 45–59. Gupta, S., Hanssens, D., Hardie, B., & Kahn, W. (2006). Modeling customer lifetime value. Journal of Service Research, 9(2), 139. Holm, M., Kumar, V., & Plenborg, T. (2016). An investigation of customer accounting systems as a source of sustainable competitive advantage. Advances in Accounting, 32, 18–30. Ittner, C. D., & Larcker, D. F. (1997). Quality strategy, strategic control systems, and organizational performance. Accounting, Organizations and Society, 22(3–4), 293–314. Ittner, C. D., & Larcker, D. F. (2001). Assessing empirical research in managerial accounting: A value-based management perspective. Journal of Accounting and Economics, 32, 349–410. Jayachandran, S., Sharma, S., Kaufman, P., & Raman, P. (2005). The role of relational information processes and technology use in customer relationship management. Journal of Marketing, 69(4), 177–192. Kaplan, R. S., & Norton, D. P. (2004). Strategy maps. Boston: Harvard Business School Press. Kraus, K., Håkansson, H., & Lind, J. (2015). The marketing-accounting interface – Problems and opportunities. Industrial Marketing Management, 46, 3–10. Kumar, V. (2008a). Customer lifetime value – The path to profitability. Foundations and Trends in Marketing, 2(1), 1–96. Kumar, V. (2008b). Managing customers for profit: Strategies to increase profits and build loyalty. Upper Saddle River, N.J.: Wharton School. Kumar, V., & Rajan, B. (2012). Handbook of marketing strategy. Cheltenham: Edward Elgar. Kumar, V., & Reinartz, W. (2016). Creating enduring customer value. Journal of Marketing, 80(6), 36–68. Kumar, V., Venkatesan, R., Bohling, T., & Beckmann, D. (2008). The power of Clv: Managing customer lifetime value at Ibm. Marketing Science, 27(4), 585–599. Laffey, D. (2007). Paid search: The innovation that changed the web. Business Horizons, 50(3), 211–218. Langfield-Smith, K. (1997). Management control systems and strategy: A critical review. Accounting, Organizations and Society, 22(2), 207–232. Lee, C. L., & Yang, H. J. (2011). Organization structure, competition and performance measurement systems and their joint effects on performance. Management Accounting Research, 22(2), 84–104. Libai, B., Muller, E., & Peres, R. (2009). The diffusion of services. Journal of Marketing Research, 46, 163–175. McManus, L., & Guilding, C. (2008). Exploring the potential of customer accounting: A synthesis of the accounting and marketing literatures. Journal of Marketing Management, 24, 771–795. Mulhern, F. (2009). Integrated marketing communications: From media channels to digital connectivity. Journal of Marketing Communications, 15(2), 85–101. Nagar, V. (2002). Delegation and incentive compensation. The Accounting Review, 77(2), 379–395. Ramani, G., & Kumar, V. (2008). Interaction orientation and firm performance. Journal of Marketing, 72(1), 27.

12

1 Introduction

Roslender, R., & Wilson, R. M. S. (2008). The marketing/accounting synergy: A final word but certainly not the last word. Journal of Marketing Management, 24(7–8), 865–876. Rust, R. T., Moorman, C., & Bhalla, G. (2009). Rethinking marketing. Harvard Business Review, 88(1), 94–101. Ryals, L. (2008). Managing customers profitably. Chichester: John Wiley & Sons. Shah, D., Rust, R. T., Parasuraman, A., Staelin, R., & Day, G. S. (2006). The path to customer centricity. Journal of Service Research, 9(2), 113–124. Sheth, J. N., Sisodia, R. S., & Sharma, A. (2000). The antecedents and consequences of customercentric marketing. Journal of the Academy of Marketing Science, 28(1), 55–66. Varian, H. R. (2006). The economics of internet search. Rivista di Politica Economica, (IX–X), 177–191. Varian, H. R. (2009). Hal Varian on how the web challenges managers. McKinsey Quarterly. Varian, H. R., Farrell, J., & Shapiro, C. (2004). The economics of information technology: An introduction. Cambridge: Cambridge University Press. Verhoef, P. C., & Lemon, K. N. (2013). Successful customer value management: Key lessons and emerging trends. European Management Journal, 31(1), 1–15. Widener, S. K., Shackell, M. B., & Demers, E. A. (2008). The juxtaposition of social surveillance controls with traditional organizational design components. Contemporary Accounting Research, 25(2), 605–638. Wruck, K. H., & Jensen, M. C. (1994). Science, specific knowledge, and total quality management. Journal of Accounting and Economics, 18(3), 247–287.

Chapter 2

Customer Analytics: Definitions, Measurement and Models

The world’s most valuable resource is no longer oil, but data. —The Economist

2.1

Customer Analytics: Definitions of CP, CLV and CE

In a recent discussion on the future of management accounting, Cokins (2013, 2014) pointed out that cost accounting techniques like Activity-Based Costing were conceived as causal cost tracing approaches to manage the complexity caused by increasingly diverse types of products, services, channels and customers. He labelled the period from 1980 to date as the ‘consumer era’ and suggested moving forward into the predictive analytics era, with a shift in emphasis from a backward-looking to forward-looking perspective of strategy and operations. Cokins (2013: 25) identified the expansion from product to channel and customer profitability analysis and called for management accounting to support the sales and marketing function to find “the best types of customer to retain, grow, win back and acquire” in order to maximize shareholder value. Consistently, with such a call to expand the toolkit of traditional cost accounting techniques, a survey by Deloitte in 2016 found that more than half of responding North American CFOs, broadly speaking, were investing substantially in (or were planning to invest in) customer analytics, with finance/accounting analytics running a close second in terms of priority. The central tenet behind any performance measurement system is the type and sophistication of tailored business performance metrics or indicators that allow managers to gauge a firm’s performance against targets. Literature reviews by Kumar and George (2007), Villanueva and Hanssens (2007), Kumar (2008a) and Petersen et al. (2009) provided exhaustive coverage of a new generation of customer-metrics in the marketing literature. Three core marketing-related indicators have been crucial in ensuring the shift towards a customer-centric strategy:

© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2019 M. Bonacchi, P. Perego, Customer Accounting, SpringerBriefs in Accounting, https://doi.org/10.1007/978-3-030-01971-6_2

13

14

2 Customer Analytics: Definitions, Measurement and Models

• Customer Profitability; • Customer Lifetime Value; • Customer Equity. We briefly define these fundamental customer analytics consistently with the marketing and accounting literature and will rely on these definitions accordingly for the remainder of this book (Pfeifer et al. 2005; Villanueva and Hanssens 2007; Gleaves et al. 2008; Kumar 2008a; Kumar and Shah 2009). Customer Profitability (CP) is defined as the difference between the revenue earned from, and the cost associated with, a customer relationship during a specified period (Smith 1993; Smith and Dikolli 1995; Foster et al. 1996). This metric is usually gauged in one accounting period (e.g. monthly, quarterly, half yearly and/or yearly) in which all revenues and costs have to be traced or allocated to customers. CP belongs to the necessary toolkit that helps to make decisions about: (a) which customers to select for targeting; (b) determining the level of resources to be allocated to the selected customers; and (c) selecting the customers to be nurtured to increase future profitability (Kumar 2008a). Customer Lifetime Value (CLV) in its classical definition is the value of future cash flow attributed to a single customer or a group of customers, discounted using the average cost of capital of the firm (Kumar 2008a). It is a leading informative indicator that drives customer profitability (Kumar and Rajan 2009). CLV can also be defined in terms of profit instead of cash flow (see, Gupta and Lehmann 2005). If we assume that cash flow equals profit, CP becomes a special case of CLV with the lifetime period set at one accounting period (Gleaves et al. 2008). CLV is measured using three main components, namely customer retention rate, margin per customer, and cost of customer acquisition. CLV is a pivotal metric that is useful both for customer profitability analysis and in valuing companies (cf. Chaps. 3 and 4 of this book). Customer profitability is positively associated with the forward-looking perspective offered by CLV (Kumar and Bharath 2009), in particular, when a firm has to decide which customers to acquire/retain because CLV is the upper limit of what one should be willing to spend to acquire/retain a customer unless one wants to lose money. CLV allows assessing which customers to nurture, with the underlying tenet that management should focus on customers with high CLV. Finally, the incorporation of CLV in decision-making should improve resource allocation, with marketing resources that should strive to maximize CLV. Similarly, equity valuation will benefit because CLV offers the algorithm that helps to estimate one of the most important assets of a company: the value of its customer base. In fact, CLV provides a valuation model that allows understanding of the mechanisms by which individual customer metrics (i.e. ARPU, churn, cost of customer acquisition) affect a firm’s sales/earnings, and ultimately its stock return (Bonacchi et al. 2015). We elaborate further on this topic in Chap. 4 of this book. Finally, Customer Equity (CE) is a combination of a firm’s current customer assets and the value of the firm’s potential customer assets (Villanueva and Hanssens 2007). CE is defined as the sum of the CLV of all a firms’ existing and potential customers. In other words, CLV is a disaggregate measure of customer profitability,

2.1 Customer Analytics: Definitions of CP, CLV and CE

15

while CE is an aggregate measure. CE is an intangible asset of the firm influenced by the ability to acquire, retain, and increase the customer base (Gupta et al. 2004; Kumar and Shah 2009; Bonacchi et al. 2015). In sum, the key distinctions between these three concepts, which all measure customer value, relate to the timescale under consideration (1 year, multiple years), and to whether the analysis refers to one or all of a firm’s customers. For a visual representation of the inter-relation among Customer Profitability, Customer Lifetime Value, and Customer Equity, refer to Fig. 2.1 and Gleaves et al. (2008). For the sake of completeness, the figure also shows the operating profit that, under the assumption that all costs have been traced to customers, is the sum of the customer profitability from all customers the firm has served within a single accounting period. According to past reviews (Kumar and George 2007; Villanueva and Hanssens 2007; Kumar 2008a; Petersen et al. 2009), the literature on CLV and other customer metrics in mainstream marketing research presently provides a rather consolidated stream of research concerned with the development and refinement of modelling approaches in various business settings. The first modelling stream attempts to use deterministic equations in which some inputs are entered into the equation in order to calculate CLV (for a review of these models see Berger and Nasr (1998)). More recently, in order to control for some endogenous parameters, researchers have proposed stochastic models to estimate CE. Inherent in all these models which try to value the long-run financial contribution of a customer, is the expected length of the relationship. The most interesting are statistical models used to predict the probability of churn (or retention) (see Villanueva and Hanssens (2007) for a review of these models). Some researchers have also developed a parsimonious model in which the parameters can be easily obtained, even in Microsoft Excel (Fader and Hardie 2007b). With regards to practitioner’s literature that contains applications of CLV, initial evidence is currently available in recent books such as Gupta and Lehmann (2005), Kumar (2008b) and Ryals (2008). Case studies written with pedagogical purposes Fig. 2.1 Classification of customer metrics All customers

A single customer

Operating profit

Customer Equity

Customer Profitability

Customer Lifetime Value

Current accounng Period

All Future accounng periods

16

2 Customer Analytics: Definitions, Measurement and Models

about best-practice CLV techniques are also emerging (Ofek 2002; Asis and Narayanan 2007). Bendle and Bagga (2017) provide an exhaustive list of relevant cases, notes and teaching materials on CLV. The next paragraph outlines the formulae applicable to compute CLV.

2.2

CLV Formulae: Sources and Variations

For exhaustive reviews of CLV models and their underlying logic, we invite the reader to refer to Jain and Singh (2010), Ascarza et al. (2017) and Kumar (2018) as excellent literature reviews of the marketing literature. The definitions available on how to compute CLV vary depending on underlying assumptions and different notations (Fader and Hardie 2012; Bendle and Bagga 2017). A quite commonly used definition of CLV is the one provided by Rust et al. (2009): “The customer lifetime value metric evaluates the future profits generated from a customer, properly discounted to reflect the time value of money”. Despite the variation and at times inconsistency across definitions, the rationale behind CLV computation therefore resembles the Net Present Value in finance, where a constant series of cash flows over time is discounted to take into account the time value of money (d ). Most common CLV definitions therefore assume the following equation, with m the (average) contribution margin generated from a customer (or customer segment/ channel) in a year or other period (cf. Steenburgh and Avery 2017). mr mr2 þ þ  ð1 þ d Þ ð1 þ d Þ2   t 1 X r ¼m 1þd t¼0

CLV ¼ m þ

A fundamental element of any CLV computation refers to the churn rate (r), defined as the percentage of customers who end their relationship with the company in a given period. The churn rate is typically defined at the segment level, and it is implicitly assumed that all individuals in that segment have the same probability of ending the relationship with the firm. In each subsequent period, the probability that a customer leaves is modelled as a survival probability function that decreases over time along the entire lifetime of the customer. The series of survival probabilities thus determine the expected cash flows (proxied by the periodic contribution margin) in a given period. If we sum the discounted expected contribution margins over a customer’s lifetime, for the properties of infinite geometric series we obtain a simplified version of the CLV formula that nevertheless may differ depending on two factors:

2.3 Applications of CLV in Subscription-Based Business Settings

17

1. whether we include the customer’s first payment in the calculation; 2. whether the net cash flow associated with each period is “booked” at the beginning or at the end of the period. Different assumptions of these two factors explain the slight variation across available CLV models as conveniently summarized in Fader and Hardie (2012). Moreover, Jain and Singh (2010) emphasize that “the context of CLV measurement plays a key role in the methods proposed for measuring CLV and the issues that become important both from a modeling point of view and the managerial point of view. By context, we mean the context of the customer-firm relationship that generated the data to be used for estimating CLV. From a modeling perspective, the context defines the data available to estimate a CLV model, and from a managerial perspective, the context defines the issues that become important in managing customer profitability.” Jain and Singh (2010) outlined several factors that might have an effect on CLV modeling and forecast. Among them, the costs associated with the acquisition and retention of customer segments are of particular importance for CFOs and management accountants. The next paragraph illustrates how the marketing literature carefully distinguishes different business contexts and develops alternative statistical models to measure CLV in an accurate way.

2.3

Applications of CLV in Subscription-Based Business Settings

Fader and Hardie (2009) and Jain and Singh (2010) referred to different contexts that fundamentally impact how to define and model CLV. A classification of customer bases usually distinguishes two dimensions: a) opportunities for transactions (continuous versus discrete) and b) type of relationship with customers (non-contractual versus contractual). The contractual or subscription-based business model refers to the case in which a customer pays a subscription fee to have access to the firm’s products or services. The model was pioneered by magazines and newspaper publishers but is now used by a growing number of businesses. Rather than selling products individually, a subscription-based firm sells periodic (e.g. monthly, yearly, or seasonal) use or access to its products or services. The number and variety of subscription-based enterprises (SBE) is fast-growing (Economist 2009). Industries based on the subscription model include telephone companies, cable television providers, cell phone companies, internet providers, pay-tv channels, software and business solution providers, financial service firms, as well as the online versions of traditional newspapers and magazine publishers. Additions include online storage, photo sharing, social networking, and online games. The defining characteristic of an SBE is that the acquisition and departure of a customer is observed (unlike the case of brick-and-mortar retailers, for instance). In fact, because a subscription typically involves a contractual agreement, the vendor

18

2 Customer Analytics: Definitions, Measurement and Models

knows at any point in time the number of active customers. We therefore contend that the notions of retention and the churn rate of customers are at the heart of any contractual or subscription-oriented business model.1 The retention rate for period t (rt) is defined as the proportion of customers active at the end of period t–1 and still active at the end of period t. The churn rate for a given period is defined as the proportion of customers active at the end of period t–1 but who dropped out in period t. It is easy to perform this type of customer data tracking and analysis for SBEs because these companies always know who are still active customers and who recently churned. Recent years have seen a dramatic increase in SBEs to offer services, such as mobile telephone, cable television, and e-banking. In these types of companies, a customer pays a subscription fee to have access to the firm’s products/services. The growth of the Internet has broadened the provision of innovative services in a contractual setting (such as music, games, movies and e-books) and a wave of successful web start-ups (such as FriendFinder, HomeAway, LinkedIn, Pandora, Skype, Zynga). In addition, Internet SBEs are currently able to gather a massive amount of aggregate and granular data about customer characteristics and preferences through the analysis of repeated purchasing behavior transactions (Varian 2006). Managers need to gain a more nuanced understanding of the strategic, financial, and operational implications of a subscription-based model in order to answer the critical question: What does it take for SBEs to succeed? As highlighted in Kumar and Rajan (2009), managers need performance measurement reports that are able to convey information useful to diagnose the health of their business and to assist them in making strategic and tactical decisions such as: • Which type of actual customer or future prospect should a firm retain, grow or acquire? • How much should a firm invest to retain, grow and acquire customers? • Which advertising channels are more effective and efficient? • What is the value of the customer base (i.e. the most important asset for these kinds of companies)? Stemming from several marketing studies, Kumar and Reinartz (2016) explained a widely applicable approach useful to address these key dilemmas. They outlined a series of steps starting from the definition of a CLV model with available data, realizing the need for a forward-looking metric as allegedly the most critical decision. Once this step has been achieved, the following activities only strengthen a company’s position in enhancing a customer-centric strategy. Despite the fact that the beneficial effects of a customer-centric approach are widely emphasized in the management accounting academic and practitioner literature—see for instance a recent IMA Statement on Management Accounting on customer profitability management (2010) anecdotal evidence suggests that not

1

Customer churn is the standard term used among SBEs.

2.4 CLV Scorecard and Cohort Analysis: An Application in an SBE

19

many firms practice such a strategy in a systematic and effective manner. We contend that one of the reasons may be a lack of empirical evidence in academic research in management accounting that could generate practical guidance on how to measure CLV. In the next paragraph, we present a real-world application of CLV measurement that could be useful to adopt in an SBE setting.

2.4

CLV Scorecard and Cohort Analysis: An Application in an SBE2

In this paragraph, we focus our attention on two tools—the CLV scorecard and cohort analysis—that CFOs, management accountants and marketing managers in subscription-based enterprises can potentially use to have a better understanding of the way they measure and manage customer profitability. We first provide some background information about the case company in which the tools are adopted. We then illustrate the firm’s performance measurement system and discuss how the CLV paradigm can be successfully implemented. Our research site, Company.net (the firm’s real name has been disguised for confidentiality purposes), is a typical example of the fast-growing subscriptionbased business model in which a customer pays a subscription fee to receive a specified number of downloads of content, namely ringtones and music MP3s. Company.net has contracts with all major record labels that supply content through downloads, it has agreements with the most important carriers that deliver content online and it bills customers on their mobile accounts. Customers are contacted and acquired predominantly through paid search advertisements, using keywords such as ‘free-ringtones’ or ‘free music’.3 The industry value chain unfolds through a series of activities illustrated in Fig. 2.2. They are labeled as follows: content origination; service management; marketing and display; network delivery; customer relationship management; and billing. Company.net follows a typical SBE strategy which can be synthetized as follows: (1) Acquire new customers through aggressive marketing campaigns and marketing tools aimed at customer acquisition to expand its customer base;

2

This section is based on and include excerpts from pp. 27–39 of Bonacchi and Perego (2012). Paid search advertising entails advertisers competing for top listing positions through bidding in ongoing auctions and then paying when users click on their advertisements, making paid search a flexible and accountable form of advertising (Laffey 2007). Pay per click ads are one of the most cost effective advertisement tools. These are the ‘sponsored ads’ that are displayed at the top and right of the search results on Google, Facebook, Yahoo, Bing and similar search websites. Payment of these ads occurs whenever someone actually clicks on an ad, not when it gets displayed. The cost per click can range anywhere from a few cents to several dollars, depending on the type of industry and keywords. 3

20

2 Customer Analytics: Definitions, Measurement and Models

Company.net

Content Owners

Content origination

v R&D activity Design and project of the product/service serv rvice

Mobile Content & Service providers Service management

Aggregation & publishing of content and services in different formats (www, Text, MMS, WAP..)

Media companies

Mktg & Display

Marketing and promotion of VAS/music through MEDIA: TV, print or online networks

Telcos

Network delivery

Networkk used b to distribute s to VAS/music r end users (Mobile operators)

User

Revenues Cost of delivery Contribution margin Cost of acquisition Customer Margin Fixed cost EBIT

Fig. 2.2 Entertainment industry value chain of Company.net. Source: Bonacchi and Perego (2012)

(2) Retain existing customers by measuring the lifetime of users and estimate their value with techniques that tend to stimulate user retention and minimize churn rate; (3) Grow: after an initial period in which the user base is built and the churn rate is stabilized, the focus shifts towards planning for an organic growth of the customer base, and defining the new target of customer acquisition for each period (e.g. month or quarter) in order to balance the churn rate and achieve the target growth rate.

2.4.1

The CLV Scorecard as a Performance Measurement System

As prior literature highlights, a firm’s performance measurement system (PMS) shapes, and is shaped by, a firm’s strategy (Chenhall 2003; Chapman 2005; Kober et al. 2007; Henri 2010). At Company.net, the PMS has been built around two metrics: customer lifetime value (CLV) and customer equity (CE), defined previously in Sect. 2.1. CLV is the value of future cash flow/profit attributed to a single customer or a group of customers discounted using the average cost of capital of the firm. Customer equity (CE) is the sum of CLV across all firms’ customers, thus

2.4 CLV Scorecard and Cohort Analysis: An Application in an SBE

21

including existing and future customers. It is the most important asset for SBE, influenced by the ability to acquire, retain, and increase the customer base. Whereas the marketing literature provides several methods of computing CE, we refer to two metrics that can be leveraged when evaluating the expected profitability of a firm’s customer base: a) Current Customer Equity (CEcur): the sum of the future profit margins generated from the customers that have already been acquired by the end of the period (Villanueva and Hanssens 2007). b) Total Customer Equity (CEtot): the sum of the future profit margins generated from current (CEcur) and future (CEfut) customers of the firm (Hogan et al. 2002a; Kumar and Shah 2009).4 From a management accounting perspective, the challenge is not only to measure CLV (and thus CE as a summation of each customer’s CLV), but especially to manage its drivers. The linkages among such drivers is crucial since the variables of the CLV formula are interdependent rather than independent. Whenever managers’ attention focus towards one direction, it becomes more difficult to ensure that the other metric moves in the same direction and pace. The CLV scorecard is a managerial tool that includes the interaction among CLV and its drivers as its key attribute (Bonacchi et al. 2008). Figure 2.3 exhibits the CLV Scorecard that identifies Measures ARPU CONTRIBUTION MARGIN

YIELD Cost of service

LIFETIME VALUE (LTV) Instant churn LIFETIME Customer Lifetime Value (CLV)

Historic churn

Attraction COST OF ACQUISITION (CoA)

Conversion Cost of contact (CPC, CPA)

feedback

Fig. 2.3 Customer lifetime value (CLV) scorecard at Company.net. Source: Bonacchi and Perego (2012)

4 CEcur can be considered a special case of CEtot, where the acquisition of future customers is a project with a Net Present Value of zero (Bonacchi et al. 2015).

22

2 Customer Analytics: Definitions, Measurement and Models

Lifetime Value (LTV) and Cost of customer Acquisition (CoA), two essential CLV drivers for Company.net. CoA is a pretty straightforward metric, which in this setting can be estimated in a very precise way for every single customer. Company.net uses search engine advertising that measures Cost per Click (CPC) and Cost per Acquisition (CPA) metrics rather than less traceable banner advertising (measured by Cost per impression model, aka CPM), or traditional channels advertising such as magazine and TV ads. This evolution in advertising also implies innovative marketing strategies that have shown new challenges in measuring their effectiveness from the early stages of the Internet.5 Since CoA is influenced by the type of marketing policies, the first necessary step for managerial actions implies an accurate analysis of the impact of such policies on attraction and conversion rate. Examples of possible indicators consist of metrics that capture better targeting, better advertising, better landing pages, optimization of the flow through to checkout, more and better payment options, and so on. The examination of LTV requires a sophisticated analysis of its main drivers (i.e. Lifetime, Margin and YIELD). The firm needs to carefully examine the relationships between the lifetime value, synthetized by the following equation6: Lifetime Value ¼ Lifetime  Margin  YIELD where: 1. Lifetime is the period during which a customer remains with a firm. This metric is a function of the rate of attrition (cancellations/average users per period) over a period of time that subscriber-based customers ‘churn out’ (unsubscribe) from the customer base. Churn is a proxy of the customer satisfaction associated to the goods and service provided. The underlying rationale captures a fairly simple relationship between churn per month and the number of months customers stay with the company (i.e. ratio 1/churn). Thus, a 2% churn means 1/.02, or 50 months average customer duration. To gain a good grasp of the churn metric, Company. net applies two approaches: a. Historic churn: the number of subscribers cancelling during the period (day, week, month) having initiated their subscription before that period, i.e. join

5

Advertisers can now set a bid they are willing to pay to reach a certain typology of Facebook users. These are audiences Facebook can reliably deliver thanks to demographic data collected among its users, while other ad exchanges might have to guess or infer about who fits an advertiser’s desired demographic. For more details on this topic, please refer to: http://techcrunch.com/2012/09/18/ facebook-mobile-ad-network/ 6 In theory, the issue of the time value of money must also be considered, usually discounted in the CLV formula, in order to get an estimate of future profits. Discounting may be very appropriate in some businesses, particularly for firms operating in markets with a very long-cycle, high ticket retail and B2B. However, it is believed that the discounted practice may confuse the analysis in a B2C where the environment is very dynamic and it is better to adopt a simplified approach.

2.4 CLV Scorecard and Cohort Analysis: An Application in an SBE

23

date < quit date. This ratio is computed as follows: subscribers that quit in period N/subscribers alive at the end of month N + 1. Historic churn is the figure used to transform churn in lifetime; b. Instant churn: number of subscribers cancelling the service during the period (day, week, month) in which they initiated their subscription, i.e. join date ¼ quit date. Gross Addition is the number of new subscribers during the specified period. In other words, the following ratio is calculated: subscribers that quit in period N/gross addition period N. This metric is devised to judge the quality and effectiveness of specific marketing and advertising decisions. 2. Margin is the contribution margin per customer and equals Average Revenue per Users (ARPU) less cost of service per customer. For instance, a subscription to a mobile service (i.e. ringtones) of $10 and a service cost per customer for content delivery of $4 would generate a margin of $6. 3. YIELD is defined as the ratio between subscribers successfully billed and customer candidates to be billed. For instance, given a total number of 1000 candidates of which 550 actually pay (for example because the sim-card of a subscriber might be empty), would generate a YIELD of 55%. Monitoring the abovementioned metrics is beneficial to evaluate strategic and tactical marketing choices—such as different advertising campaigns, the launch of a new product or a traditional product in a new country—cross-sell and upsell campaigns to increase ARPU, improve margin, and so on. Company.net reports internally on LTV driver with a typical report that focuses on a product/service (music, ring tones, and other value-added services) providing the following information: country, subscription flight (i.e. the scheduling of advertising for a period of time), telecom operator used to deliver the service (refer to Table 2.1). Comparisons at country-level allow evaluating the success of a product already launched in other countries. Linking the type of marketing campaign to the acquired customers’ churn is a distinctive analysis conducted to gauge the effectiveness of a subscription campaign on a service/product. For example, a customer acquired using Google AdWords could have a different churn rate than a customer acquired through Yahoo! My Display Ads, or Facebook. Furthermore, customers acquired from a specific mobile operator (e.g. AT&T, Vodafone, or T-Mobile) could have different attitudes to churn and a diverse YIELD profile. Table 2.1 Analysis of churn data: example of Company.net’s internal report Dimension Country Subscription flight Telco

Metric Instant churn

Source: Bonacchi and Perego (2012)

Historic churn

Yield

Margin

24

2 Customer Analytics: Definitions, Measurement and Models

These drivers can be considered leading, forward-looking indicators and must be monitored on a daily basis to verify that the company’s results fit with the strategic objectives of the business. Whereas company presentations often display an improvement of all customer metrics overtime, however, this is unlikely to play out in reality. For instance, if a firm attempts to raise ARPU (price), it will predictably increase churn. Similarly, whether a firm intended to grow faster by spending more on marketing, CoA would likely rise as inevitable consequence. Churn may increase as well, because a more aggressive marketing campaign will likely capture customers of a lower quality. In sum, the availability of accurate customer data and the peculiar nature of the business examined (in which all transactions are made online and the log files of each transaction are constantly and accurately recorded) represents a prerequisite for a timely assessment and effective management control of these metrics in order to assess a firm’s marketing strategy.

2.4.2

Benefits of CLV Scorecard

The main benefit associated with the CLV scorecard is the opportunity to analyze past customer behavior and understand the drivers of CLV. For example, it may be valuable to study the differences between CLV or COA of an average customer acquired through Facebook or Google ad campaigns. In other words, the analysis of the CLV drivers allows managers to test cause-and-effect relationships between managerial actions and outcomes of such actions in terms of both customer metrics (Churn, CoA, Margin, Yield) and financial results (CLV and CE). In the context of this case setting, we observed that Company.net regularly monitors the following objects or units of analysis: • specific marketing campaigns in place. In fact, different channels contribute with different results in terms of CLV. For example, customer acquired through Google or Facebook could have a tendency to churn differently than those coming from Yahoo or Bing. The analysis should exploit the retention behavior (instant and historic churn) of customers acquired through these two channels in order to understand which lifetime those users project. • different countries in case the same service is offered internationally. This information is quite useful when a service is introduced in a new country, since data gathered in a similar country provides an appropriate benchmark. For example, the pay-back of marketing investments in Brazil could be a good proxy to infer what would happen if the firm invested by offering the same services in Chile. • different time frames to monitor different consumer behavior. In this way, a daily or hourly fine-tuning of a marketing campaign should be performed, for example,

2.4 CLV Scorecard and Cohort Analysis: An Application in an SBE

25

by changing the bid price on the keyword advertising.7 In the paid search model (Laffey 2007), monitoring the behavior of click-throughs via paid search is essential, because it provides a precise measurement of the success of the advertising method in terms of achieving the objectives set forth. Data collected from such tracking should then be fed back into the process to make performance reviews more effective. It cannot be excluded, for example, that poor quality prospects are attracted (perhaps through the use of the wrong keywords), or that clicks from some sources work better than others. Furthermore, monitoring can be used to ensure that a firm is not excessively paying for clicks; for instance, being in second or third position may generate as much business as being in first position, where the listing position is determined by how much an advertiser is prepared to pay for a keyword or phrase. • different customer characteristics. Although top Company.net managers are interested in computing the lifetime value of their customers, they are similarly keen on identifying the drivers of a profitable duration in their customer-firm relationships (Reinartz and Kumar 2003). More specifically, the telecommunication carrier can be considered a proxy of income. Moreover, a longer subscription length is a proxy of customer satisfaction, since it is well known in the literature that the longer a subscriber remains with the company, the lower the probability they are going to churn (Gupta and Lehmann 2005; Fader and Hardie 2007a, 2010). Conceptually, the following can be thus formally posited: Profitable lifetime duration ¼ f ðtelephone operator; length of subscription; advertising campaignÞ

2.4.3

CLV Cohort Analysis: Rationale

Churn rate is the main driver of customer lifetime and churn data are crucial to judge the ex-post success of a customer acquisition campaign. The marketing literature however raises serious concerns about the typical approach of projecting a historic churn rate into the future to infer the lifetime of the acquired customer and then their profit stream (i.e. LTV). Gupta et al. (2004) demonstrate that the widespread method for the conversion of retention rate to expected lifetime (1/churn rate) and then the calculation of present value over that finite time period overestimates lifetime value. Fader and Hardie (2007a, 2010) showed that the retention rate (defined as the opposite of the churn rate) is an increasing function of time. Therefore, the longer subscribers are with the company, the lower the probability that they are going to

7

When a user searches for a specific keyword, the order of the results the user obtains is determined by current bids in the auction. Payment is made by advertisers each time a user searches for a term and then clicks on their link.

26

2 Customer Analytics: Definitions, Measurement and Models

churn. Researchers account for two main reasons that explain this phenomena (Blattberg et al. 2008). As the customer uses the product or services, he/she increases preference or satisfaction or is locked-in and that causes the decreasing churn rate. Another explanation of an increasing retention rate may be due to cross-sectional heterogeneity in individual retention probabilities across customers in their preference for a product/service (e.g., Gupta and Lehmann 2005, 29–31). The heterogeneity problem can be particularly severe if the company shows substantial growth and thus many first-time customers whose churn can be different from older customers, thus distorting the calculation of how much repeat purchase behavior will occur in the future. Company.net’s Business Intelligence unit has recognized and addressed these crucial issues. The BI unit applies a heuristic methodology to estimate LTV from future acquired customers by projecting into the future the data of each acquired customers’ cohort with a data-mining approach. Cohort analysis has been used by statisticians for decades. However, recent advancements in data collection and processing power have made cohort analysis a viable technique for online businesses to study customer loyalty trends, predict future revenues, and monitor churn. The most popular cohort analysis (and the one we present in this case study) involves segmenting customer groups based on a “join date.” The month, week, or day then becomes the user’s “cohort,” meaning each cohort is the cluster of users who joined in that same time period. The pivotal metrics used in this analysis is called Margin per Thousand Customers (MpK) and focuses on projected margins.8 Company.net estimates across cohorts an average MpK for the last N months of actual data and then, after a normalization based on previous observations, it projects it into the future months. The metric is calculated with different windows, namely daily, weekly (preceding 7 days) and monthly (preceding 30 days). The CLV is computed with the following assumptions: 1) LTV is based on future profit (see distinction between cash flow and profit Gleaves et al. 2008); 2) CLV does not directly consider the CoA (see Pfeifer et al. 2005).9 Formally CLV is obtained by subtracting CoA from LTV, i.e. acquisition costs are not included as part of lifetime value. However, customer acquisition cost is often displayed alongside a customer’s LTV. In this way, Company.net gains insights about whether an unprofitable customer whose CLV (LTV minus CoA cost) is negative is due to high costs of acquisition rather than a low LTV.

8

Using the same rationale at Company.net also computes Revenue per Thousand Customers (RpK) focusing on projected revenues instead of margin. In the following example, we will concentrate our attention on the MpK evolution, however the same report can be adopted for RpK, since in fact gross margin is a percentage of the revenue (in this example 65%) and the difference between Rpk and MpK is merely a question of scale. Our choice to present MpK is because with this metric we are allowed to calculate CE. 9 When it comes to make informed prospecting decisions, there are at least two ways of considering acquisition spending (Pfeifer et al. 2005). Either not include acquisition spending and compare the Lifetime Value (LTV) to CoA; or alternatively, include acquisition spending in the specification of customer value, correctly labeled as CLV, and compare the value of CLV to zero.

2.4 CLV Scorecard and Cohort Analysis: An Application in an SBE

27

The following steps are taken in order to compute the MpK metric: 1) Collect gross additions (acquired subscribers) for each cohort and the corresponding number of billings (subscribers who effectively pay) in each period. It is worthwhile noticing that gross addition is always higher than the number of subscribers billed for two reasons. First, some of the subscribers quit (this phenomenon is measured by churn rate). Second, it is not possible to charge all the candidates subscribers. This phenomenon is measured by the YIELD. 2) Multiply the number of billed subscribers for the Margin per customer in order to determine MpK. 3) Estimate for each considered period of time the average MpK among the different cohorts. By averaging across cohorts, one can determine an average MpK at the end of 1 month, 2 months, and so forth (Eq. 2.1). MpKt ¼ margin

Xn 1000 1 # Billingt t¼1 gross additiont n  ðt  1Þ

ð2:1Þ

where: t ¼ period of time ¼ cohort n ¼ total periods where data is available margin ¼ ARPU – COGS #_Billing ¼ paid subscribers for each cohort The estimation also requires a normalization of the data (i.e. outliers are not considered in the average). In most cases, outliers are due to problems in the firm’s information system that produces the log files and also based on previous experience they can be easily identified. 4) Project average MpK on the lifetime of the customer in order to estimate how much margin can be obtained from the acquisition of 1000 customers today. In other words, with this step it is possible to estimate the tail of the MpK for the future (Fig. 2.4). As the cohorts mature, there are fewer datapoints to average across, and hence the potential for error increases. However, it is still a useful exercise to assess the future margin associated with the acquisition of a thousand customers. How far into the future the estimation can be extended depends on the type of business. At Company.net, the following metrics are computed across different time ranges: MpK1month ¼ MpK1 : expected margin of the next 1000 customer in the first month X6 MpK6months ¼ MpKt: expected margin of the next 1000 customer after 6 months Xt¼1 36 MpK36months ¼ MpKt: expected margin of the next 1000 customer after 36 months t¼1 In theory, CLV models should estimate the value of a customer over their lifetime. However, in many firms, including Company.net, 3 years is considered a reasonable estimate of the horizon over which the current business environment

28

2 Customer Analytics: Definitions, Measurement and Models

$4,500 $4,000 $3,500 $3,000 $2,500 $2,000 $1,500 $1,000 $500 $-

1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 GMpK

Fig. 2.4 Distribution of MpK for the average cohort. Source: Bonacchi and Perego (2012). Note: The graph plots the distribution of the monthly margin of the “next” 1000 customers being acquired. The vertical line at the fifth month separates the estimation based on actual data (first 5 months) from the projection of the tail applying the previous monthly change and a certain rate of decrease (e.g., 95%)

(with regard to technology, competition, etc.) would not substantially change (Kumar 2008a; Kumar and Rajan 2009). Next, we apply the previously outlined equation and develop an example about how a typical cohort analysis is conducted from the firm’s log file to compute the MpK’s drawing on the rationale explained in this section.

2.4.4

CLV Cohort Analysis: A Practical Illustration

Our example assumes that Company.net launched an online service in a new country. The firm’s customer acquisition and collection are described in Table 2.2. The firm acquired a first cohort of 25,016 customers in the first month and these customers produced the following stream of payments: 8614 billing events in the first month, 13,437 in the second month, and so on. Another cohort of 38,862 customers was acquired in the second month, which generated a separate stream of billing events (15,086 the first period, 27,082 the second, and so on). For the calculation of MpK, it is necessary to complete the scenario with ARPU as an input variable. For the example covered here, we hypothesize an ARPU of $10 which is considered to be ‘booked’ at the beginning of the contract period, with a cost of service of $4.50 (i.e., resulting Contribution Margin of $6.50). We also hypothesize a customer acquisition cost of $11 per customer (this figure is an estimate derived from the ratio of marketing cost/gross addition). Company.net raise the following questions raise at the end of May:

2.4 CLV Scorecard and Cohort Analysis: An Application in an SBE

29

Table 2.2 Customer acquisition (Gross addition) and collection (billing events)

Cohort Cohort 1 Cohort 2 Cohort 3 Cohort 4 Cohort 5

Gross addition 25,016 38,862 54,985 68,099 51,851

Period of time January February March Number of customers billed 8614 13,437 15,516 15,086 27,082 32,380

April

May

12,658 22,085 35,906 36,220

9316 15,996 26,036 35,276 19,798

Note: Figures were disguised by a constant multiplier for confidentiality reasons Source: Bonacchi and Perego (2012) Table 2.3 Evolution of gross margin for each cohort

Cohort Cohort 1 Cohort 2 Cohort 3 Cohort 4 Cohort 5

Gross addition 25,016 38,862 54,985 68,099 51,851

Period of time January February March April May Gross margin X # customers billed scaled by thousand customers $2238 $3491 $4032 $3289 $2421 $2523 $4530 $3694 $2675 $3828 $4245 $3078 $3457 $3367 $2482

Note: Figures were disguised by a constant multiplier for confidentiality reasons Source: Bonacchi and Perego (2012)

1) What is the expected contribution, all things being equal, of the next 1000 customers we will acquire? 2) What is the marketing investment’s pay-back period? 3) If we stop investing in this market/service, what is the expected residual value of the customers currently on the firm’s books (i.e. customer equity of current customer base)? To properly answer these questions, they develop a report which follows the MpK rationale. The first step in preparing the report is to transform billing events in monetary terms by multiplying billing events by margin and scaling the result by one thousand customers (Table 2.3). For example, the January figure ($2238) was obtained by multiplying 8614 (January’s billing) with $6.50 (margin) and finally multiplying by 1000/25,016 in order to scale by one thousand customers. The next crucial step requires the computation of an average contribution margin estimated for each period. The question is how much margin should be expected from one thousand customers acquired today. Given the above-mentioned Eq. (2.1), it follows that one thousand costumers acquired today are expected to provide a gross margin of 2905.65 in the next month (average of 2238; 2523; 3828; 3457; 2482); 3908 in the second month (average of 3491; 4530; 4245; 3367), and so forth (refer to Table 2.4). As the cohorts mature, there are fewer data points to average

30

2 Customer Analytics: Definitions, Measurement and Models

Table 2.4 Projection of MpK in the future Time period 1 2 3 4 5 6 7 t 35 36

Average of past data

Forecast decrease (rate  95%)

MpK $2906 $3908 $3601 $2982 $2421 $1988 $1650 ... $123 $118

Standard deviation $694 $567 $484 $434 –

Rate (%) 35 8 17 19 17.9 17.0 16.1 ... 3.8 3.6

Source: Bonacchi and Perego (2012)

over. Hence, for the subscribers that started in month one, we have 5 months of data, for the subscribers that started in month two, 4 months of retention data are available, and so forth. The number of actual data (i.e., number of cohorts) depends on data availability, with the usual range from a minimum of two to a maximum of 12 months of past data. Regarding the normalization of the data, the analysis of the standard deviation allows detecting in which cases the average is biased from outliers. In those cases, the outliers are removed from the average. A typical pattern found in Company.net is that after an initial period, month-bymonth MpK tends to level off. With such a pattern, one can extrapolate forward using the same month-on-month decrease across several months. The following algorithm is applied to forecast the evolution of MpK for the cohort whose data is not available (in our example from the cohort number six onwards): drop from previous month multiplied by an x% of decrease to account for a decrease in churn rate (in this example, the percentage of decrease is supposed to be 95% for each month). In our example, we have 5 months of data and then we extrapolate forward using the same month-on-month decrease on the basis of previous experience for the subsequent 31 months. Stated differently, MpK of period six (1988) is obtained as follows: 2420.61  (19%  95%). This method allows projecting cohort’s MpK into the future, estimating the distribution’s tail (as exhibited on the right side of Fig. 2.4). As already mentioned, at Company.net this estimation procedure usually does not exceed 36 months when forecasting MpK. Further, Table 2.5 provides an answer to the first two questions previously stated, namely: one thousand customers acquired today are supposed to produce a margin of $31,121 (2906 + 3908 + . . . + 123 + 118) during the next 36 months. This result is obtained with the following algorithm. MpK36months ¼

36 X t¼1

MpKt

2.4 CLV Scorecard and Cohort Analysis: An Application in an SBE

31

Table 2.5 Typical report to support decision-making INPUT ARPU $10 OUTPUT

COGS incidence 35%

MpK6months MpK12months MpK36months Cost of Acquisition (CoA)

Margin $6.5

CoA estimate $11

Payback period 4.0 $17,805 $24,621 $31,121 $11,000

Note: This example of report highlights the relation among cost of acquisition and Lifetime Value (estimated using MpK). This type of reports are crucial in order to assess the contribution of new subscribers to value creation Source: Bonacchi and Perego (2012)

This figure must then be compared with CoA in order to obtain a CLV of $20,221 (31,121–11,000), as illustrated in Table 2.5. We obtain the pay-back of the marketing campaign from Table 2.4 summing MpK up to $11,000 (i.e., 2906 + 3908 + 3601 + 2982). This is quite relevant information when firms have to optimize the resources spent on customer acquisition and evaluate the risk of a marketing investment. However, the value of future customers and the payback period alone are not sufficient to obtain a full picture. The other piece of information required to make rational decision in this business setting is Customer Equity (CE). In order to answer the third question, a further step is needed to estimate the CE of current customers extending the actual data of each acquired cohort (from 1 to 5) with the estimation for the future. This means that data in Table 2.6 Column A must be projected into the future using data from Table 2.6 Column B. Hence, for each cohort we sum up the future periods (i.e. for cohort 1 the period 6 to 36; for cohort 2 period 5 to period 36; for cohort 3 period 4 to 36; for cohort 4 period 3 to 36; and for cohort 5 period 2 to 36). The last step is to un-scale the data in order to consider the real gross addition for each cohort, since in fact all the calculations made so far refer to 1000 customers. The result is a CE of the current customer base of $5,328,426 (Table 2.6), representing the value embedded in the acquired customerbase. The PMS of Company.net, due to the normalization of the data, is also able to simulate the effect of other factors that are not under the control of the company, such as competitors’ actions, regulation changes, or technological discontinuities, any of which may affect the consumer behavior in both conversions and retention. This is particular important when it is necessary to neutralize the effect of the abovementioned factors from the metric utilized as the base for managers’ bonuses. As an example, consider the case of a problem in the log files transmitted from the telecommunications carriers to Company.net for a specific cohort.

January February March April May Margin X # customers billed scaled by 1000 customers $2238 $3491 $4032 $3289 $2421 $2523 $4530 $3694 $2675 $3828 $4245 $3078 $3457 $3367 $2482

B Customer equity scaled per 1000 customers $15,303 $17,724 $20,706 $24,307 $28,215

C ¼ (A/1000)  B Customer equity of current customer base $382,826 $688,785 $1,138,524 $1,655,295 $1,462,995 $5,328,426

Note: Customer equity scaled for thousand customers is the value at the end of May of 1000 customers acquired in each of the five cohorts. Customer equity current customer base is obtained from the sum of the un-scaled customer equity of each cohort Source: Bonacchi and Perego (2012)

Cohort 1 Cohort 2 Cohort 3 Cohort 4 Cohort 5 Customer base

A Gross addition 25,016 38,862 54,985 68,099 51,851 238,813

Table 2.6 Estimation of customer equity for the current customer base

32 2 Customer Analytics: Definitions, Measurement and Models

References

2.5

33

Conclusions and Implications

In this chapter, we presented a simulation that applies cohort analysis in an attempt to fill the void between theoretical CLV models and its implementation in practice. The main rationale is to provide CFOs, management accountants and marketing managers a better understanding of new and latent customer preferences in a typical subscription-based business model by directly observing the customer’s purchase behavior and subsequently linking this data to estimate customer lifetime value and firm performance. The application of computing CLV is quite straightforward. It also highlights a few take-away points to bear in mind about CLV modelling: 1) Calculating CLV (and by extension, Customer Equity) is a necessary yet not sufficient condition to strive for customer-centricity, since managers need to examine CLV drivers by developing a framework that makes visible causeand-effect relationships between managerial actions and key customer metrics; 2) There is no such a thing as an average customer lifetime. Since the typical survival curve drops quickly and then levels off, more informed decisions can be taken if customer groups are segmented based on a so-called ‘join date’. Failing to do so will likely undervalue a firm’s Customer Equity; 3) For every decision (launch of a service in a different country, new advertising campaign, and so forth) managers need to rely on reports showing how that decision will impact future CLV (CE). This will allow decision-makers to better answer our initial key questions: Which type of actual customer or future prospect should be retained, grown or acquired? How much should be invested to retain, grow and acquire customers? Which advertising channels are more effective and efficient? What is the value of the customer-base?

References Ascarza, E., Fader, P. S., & Hardie, B. G. S. (2017). Marketing models for the customer-centric firm. In B. Wierenga & R. van der Lans (Eds.), Handbook of marketing decision models (pp. 297–329). Cham: Springer. Asis, F., & Narayanan, V. G. (2007). Internet customer acquisition strategy at bankinter. Harvard Business School Note #9-103–021. Bendle, N. T., & Bagga, C. K. (2017). The confusion about CLV in case-based teaching materials. Marketing Education Review, 27(1), 27–38. Berger, P. D., & Nasr, N. I. (1998). Customer lifetime value: Marketing models and applications. Journal of Interactive Marketing, 12(1), 17–30. Blattberg, R. C., Kim, P.-D., & Neslin, S. A. (2008). Database marketing: Analyzing and managing customers. New York: Springer. Bonacchi, M., & Perego, P. (2012). Measuring and managing customer lifetime value: A CLV scorecard and cohort analysis in a subscription-based enterprise. Management Accounting Quarterly, 14(1), 27–39. Bonacchi, M., Ferrari, M., & Pellegrini, M. (2008). The lifetime value scorecard: From e-metrics to internet customer value. In M. J. Epstein & J. F. Manzoni (Eds.), Performance measurement and

34

2 Customer Analytics: Definitions, Measurement and Models

management control: Measuring and rewarding performance (pp. 193–226). Bingley: Emerald. Bonacchi, M., Kolev, K., & Lev, B. (2015). Customer franchise—A hidden, yet crucial, asset. Contemporary Accounting Research, 32(3), 1024–1049. Chapman, C. S. (2005). Controlling strategy: Management, accounting and performance measurement. New York: Oxford University Press. Chenhall, R. H. (2003). Management control systems design within its organizational context: Findings from contingency-based research and directions for the future. Accounting, Organizations and Society, 28(2–3), 127–168. Cokins, G. (2013, December). Top 7 trends in management accounting. Strategic Finance, 21–29. Cokins, G. (2014, January). Top 7 trends in management accounting – Part 2. Strategic Finance, 41–47. Economist. (2009). Business: The triumph of the monthly bill; Media conglomerates in the downturn. The Economist, 393(8652), 65. Fader, P., & Hardie, B. (2007a). How to project customer retention. Journal of Interactive Marketing, 21(1), 76–90. Fader, P., & Hardie, B. G. S. (2007b). How to project customer retention. Journal of Interactive Marketing, 21(1), 76–90. Fader, P., & Hardie, B. (2009). Probability models for customer-base analysis. Journal of Interactive Marketing, 23(1), 61–69. Fader, P., & Hardie, B. (2010). Customer-base valuation in a contractual setting: The perils of ignoring heterogeneity. Marketing Science, 29(1), 85–93. Fader, P., & Hardie, B. (2012). Reconciling and clarifying CLV formulas. Available http:// brucehardie.com/notes/024/ Foster, G., Gupta, S., & Sjoblon, L. (1996). Customer profitability analysis: Challenges and new directions. Cost Management (Spring):5–17. Gleaves, R., Burton, J., Kitshoff, J., Bates, K., & Whittington, M. (2008). Accounting is from Mars, marketing is from Venus: Establishing common ground for the concept of customer profitability. Journal of Marketing Management, 24(7–8), 825–845. Gupta, S., & Lehmann, D. R. (2005). Managing customers as investment: The strategic value of customer in the long run. Upper Saddle River, NJ: Wharton School Publishing. Gupta, S., Lehmann, D. R., & Stuart, J. A. (2004). Valuing customers. Journal of Marketing Research, 41(1), 7–18. Henri, J.-F. (2010). The periodic review of performance indicators: An empirical investigation of the dynamism of performance measurement systems. European Accounting Review, 19(1), 73–96. Hogan, J. E., Lehmann, D. R., Merino, M., Srivastava, R. K., Thomas, J. S., & Verhoef, P. C. (2002). Linking customer assets to financial performance. Journal of Service Research, 5(1), 26–38. IMA. (2010). Customer profitability management: Statement on management accounting. Montvale, NJ: IMA. Jain, D. C., & Singh, S. S. (2010). Measuring customer lifetime value. In Review of marketing research (pp. 37–62). Bingley: Emerald Group Publishing. Kober, R., Ng, J., & Paul, B. J. (2007). The interrelationship between management control mechanisms and strategy. Management Accounting Research, 18(4), 425–452. Kumar, V. (2008a). Customer lifetime value – The path to profitability. Foundations and Trends in Marketing, 2(1), 1–96. Kumar, V. (2008b). Managing customers for profit: Strategies to increase profits and build loyalty. Upper Saddle River, NJ: Wharton School Pub. Kumar, V. (2018). A theory of customer valuation: Concepts, metrics, strategy, and implementation. Journal of Marketing, 82(1), 1–19. Kumar, V., & Bharath, R. (2009). Nurturing the right customers: By measuring and improving customer lifetime value, you’ll be able to grow your most profitable customers. Strategic Finance, 27–33.

References

35

Kumar, V., & George, M. (2007). Measuring and maximizing customer equity: A critical analysis. Journal of the Academy of Marketing Science, 35(2), 157–171. Kumar, V., & Rajan, B. (2009). Profitable customer management: Measuring and maximizing customer lifetime value. Management Accounting Quarterly, 10(3), 1–18. Kumar, V., & Reinartz, W. (2016). Creating enduring customer value. Journal of Marketing, 80(6), 36–68. Kumar, V., & Shah, D. (2009). Expanding the role of marketing: From customer equity to market capitalization. Journal of Marketing, 73(6), 119. Kumar, V., Venkatesan, R., Bohling, T., & Beckmann, D. (2008). The power of CLV: Managing customer lifetime value at IBM. Marketing Science, 27(4), 585–599. Laffey, D. (2007). Paid search: The innovation that changed the web. Business Horizons, 50(3), 211–218. Ofek, E. (2002). Customer profitability and lifetime value. Harvard Business School Note #9503–019. Petersen, J. A., McAlister, L., Reibstein, D. J., Winer, R. S., Kumar, V., & Atkinson, G. (2009). Choosing the right metrics to maximize profitability and shareholder value. Journal of Retailing, 85(1), 95–111. Pfeifer, P. E., Haskins, M. E., & Conroy, R. M. (2005). Customer lifetime value, customer profitability, and the treatment of acquisition spending. Journal of Managerial Issues, 17(1), 11–25. Reinartz, W. J., & Kumar, V. (2003). The impact of customer relationship characteristics on profitable lifetime duration. Journal of Marketing, 67(1), 77–99. Rust, R. T., Moorman, C., & Bhalla, G. (2009). Rethinking marketing. Harvard Business Review, 88(1), 94–101. Ryals, L. (2008). Managing customers profitably. Chichester, England. Hoboken, NJ: Wiley. Smith, M. (1993). Customer profitability analysis revisited. Management Accounting, 71(9), 26. Smith, M., & Dikolli, S. (1995). Customer profitability analysis: An activity-based costing approach. Managerial Auditing Journal, 10(7), 3–7. Steenburgh, T., & Avery, J. (2017). Marketing analysis toolkit: Customer lifetime value analysis. Harvard Business School Background Note 511-029. Varian, H. R. (2006). The economics of internet search. Rivista di Politica Economica, 8(IX–X), 177–191. Villanueva, J., & Hanssens, D. (2007). Customer equity: Measurement management and research opportunities. Foundations and Trends in Marketing, 1(1), 1–95.

Chapter 3

Customer Analytics for Internal Decision-Making and Control

Without data you’re just another person with an opinion. —W. Edwards Deming

3.1

Review of Accounting and Marketing Literature

Empirical research-based literature that focuses on how organizations embed customer-centricity in their internal management accounting systems is limited. Therefore, existing management accounting papers concerned with customer accounting provide critical commentaries and directions for further research on customer profitability analysis and other cost accounting tools (e.g. Foster and Gupta 1994). In contrast, marketing literature extensively covers customer accounting concepts such as customer satisfaction, customer retention and customer profitability in terms of their use as marketing and implementation strategies. An established stream of research specifically investigated determinants and effects of customer satisfaction and their relationship with customer profitability and performance (Gleaves et al. 2008). In this section, we highlight and briefly review literature examining the interplay between elements of organizational architecture tailored to customer centricity and internal decision-making and control activities. We organize our literature overview by pointing at main relationships between elements of the organizational architecture (see Fig. 3.1). Although our review is not exhaustive, we highlight seminal papers and research streams that appear to be more consolidated as opposed to areas that remain unexplored, and place a particular emphasis on management accounting literature. Chapter 4 covers research examining the use of customer metrics for external purposes, namely in external financial reporting and corporate valuation. (a1) Customer-Centric Strategy ! Performance Measurement In the marketing literature, customer metrics include a variety of constructs categorized into observable/behavioral metrics (e.g. customer retention and customer © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2019 M. Bonacchi, P. Perego, Customer Accounting, SpringerBriefs in Accounting, https://doi.org/10.1007/978-3-030-01971-6_3

37

38

3 Customer Analytics for Internal Decision-Making and Control

Contingent Variables Technology − Fast pace of change − Big Data, AI, Social media

Markets

Regulation

− Rising power of the Customer − Diffusion of services

− Deregulation − Globalization

Business Strategy Customer-Centric Strategy

b) Organizational Architecture Allocation of decision rights

a1)

c)

e)

Performance measures

d) Incentives

a2)

Financial performance Fig. 3.1 Literature review related to the organizational architecture (source: Adapted from Brickley et al. 1995)

lifetime value) and unobservable/perceptual measures (e.g. customer satisfaction) (Gupta and Zeithaml 2006). The implicit assumption in using unobservable customer metrics is that they anticipate or predict observable behavior such as retention or increased consumption. However, as database management and customer relationship management have evolved, researchers and firms find they can bypass unobservable metrics and directly connect a firm’s actions to their financial performance and to customers’ observable behavior. Reviews by Kumar and George (2007), Villanueva and Hanssens (2007), Kumar (2008) and Petersen et al. (2009) provide exhaustive coverage of a new generation of observable customer metrics in the marketing literature, namely Customer Profitability (CP), Customer Lifetime Value (CLV) and Customer Equity (CE), outlined in Chap. 2 of this book (Pfeifer et al. 2005; Villanueva and Hanssens 2007; Gleaves et al. 2008; Kumar 2008; Kumar and Shah 2009). In the accounting literature, few studies investigated the relationships between customer-centric strategy and customer metrics in management accounting systems. A notable exception, Guilding and McManus (2002) examined the incidence of customer accounting metrics usage and the antecedents of customer accounting adoption using a survey of Australian firms across a number of industries. Among their exploratory findings, they documented that customer profitability analysis

3.1 Review of Accounting and Marketing Literature

39

applied to customer segments is the most widely adopted accounting tool and found a positive association between a customer-centric orientation and customer accounting. The need for adopting customer accounting remains highly industry specific; it is more likely in firms facing highly competitive business environments, thereby providing support for the intuition behind the organizational architecture. Insights into the relationship between contingent environmental and strategic factors and features of customer accounting metrics remain, to the best of our knowledge, largely unexplored and warrant further investigation, with both field studies and via survey-based research. (b) Customer-Centric Strategy ! Allocation of Decision Rights Researchers argue that realigning structures away from functional or product-based approaches and toward customer groups creates more customer centric firms, possibly assuming concurrent improvements in responsiveness, customer satisfaction, and business performance (Shah et al. 2006). Alternatively, other researchers argue that changes to a firm’s structural alignment often add coordination costs and introduce greater complexity into decision making, thereby increasing a firm’s costs. In their analysis of longitudinal data (1998–2010) linking Fortune 500 firms to firm performance, Lee et al. (2015) showed that a corporate-level customer centric structure positively affects customer satisfaction, but simultaneously adds coordinating costs. Additionally, the alignment of corporate structure on customer-centric strategic objectives seems to pay off only in specific competitive markets. Thus, overall the impact of structural alignment on performance depends on whether the customer-centric benefits outweigh the additional costs associated with a customercentric structure. Such arguments remain largely empirically untested. (c–e) Allocation of Decision Rights/Performance Measurement ! Incentives Very little research has been conducted on the use of CLV and other advanced customer metrics for compensation and incentive schemes. The theoretical rationale is that encouraging the use of CLV through accountable allocation of decision-rights will likely change decision-making processes and ultimately embed customercentricity into firm culture (Kumar and Rajan 2009a, b). Ryals (2005) and CasasArce et al. (2011, 2017) documented qualitatively how two banks started providing CLV as a forward-looking non-financial metric to employees to improve their decision making. In both cases, customer data allowed the ability to gauge the effects of enriching the information set of the bank’s employees. However, while the information set of customer-related metrics was enlarged, both of the examined case companies maintained the managerial incentive scheme linked to short-term accounting profits. There is clearly a need to further understand how the (mis)alignment of the decision-rights and the reward system towards a customer-centric strategy impacts managerial decision-making and ultimately firm performance. Casas-Arce et al. (2017) found that in the banking sector, investigated branch managers with shorter tenure displayed a stronger response after the performance evaluation system started to emphasize CLV as a key performance indicator, thereby suggesting a potential substitute effect between information and work experience. It would be particularly insightful to cumulate additional insights on the effects induced by a

40

3 Customer Analytics for Internal Decision-Making and Control

forward-looking metric like CLV in the internal accountability mechanisms and decision-making processes. A handful of practical case studies demonstrated the beneficial consequences of focusing on the new generation of forward-looking customer metrics. A good example refers to Capital One, a firm that carefully segments and values its customers to understand their lifetime value and to identify the best utilization of their resources. The firm normally uses CLV to evaluate marketing campaigns and investments (Anand et al. 2001; Rierson and Lattin 2007). In the entertainment industry, for instance, evidence suggests that Harrah increasingly drives its business operations by relying on observable customer metrics (Loveman 2003). (a2) Performance Measurement ! Financial Performance A rather consolidated stream of literature in both marketing and accounting examined customer metrics as leading, forward-looking indicators of future financial performance (Gupta and Zeithaml 2006). This stream mainly drew on statistical models developed in marketing to understand the key drivers of customer acquisition, customer retention and customer expansion (Rust et al. 2000; Vogel et al. 2008; Verhoef and Lemon 2013). The impact of unobservable metrics on firm financial performance has been empirically demonstrated by several marketing papers. Some studies attempted to link the usage and sophistication of customer metrics in relationship with their impact on financial performance. For instance, Germann et al. (2013) collected data from 212 senior executives of Fortune 1000 firms and showed how firms attain favorable and apparently long-term financial effects from deploying customer analytics. More intense industry competition and rapid changes in customer preferences seem to act as moderators in strengthening the positive impact of marketing analytics on firm performance. Their paper additionally confirmed that appropriate data and information technology support represent necessary internal factors for the effective deployment of customer analytics, thereby corroborating the intuition behind alignment of organizational architecture to ensure a transition towards customer-centricity. Among the accounting papers, Banker et al. (2000) and Behn and Riley (1999) found positive associations between customer satisfaction measures and future accounting performance in the hotel and airline industries, respectively. Ittner and Larcker (1998) investigation of customer, business unit, and firm-level data also supported claims that customer satisfaction measures are leading indicators of customer purchase behavior, accounting performance, and current market value. To the best of our knowledge, only Ittner and Larcker (1998) examined the relevance of customer satisfaction measures on financial performance. In the literature on management accounting examining the customer component of market orientation, a growing number of studies focused on nonfinancial performance measures such as customer satisfaction and customer loyalty (Ittner and Larcker 1998; Banker et al. 2000; Smith and Wright 2004; Davila and Foster 2005; Dikolli and Sedatole 2007). In general, this literature attempted to demonstrate that nonfinancial (forward-looking) metrics have a higher predictive ability to anticipate future financial performance compared to traditional (backward-looking)

3.2 Evaluation of the Literature

41

accounting metrics. Holm et al. (2016) analyzed a longitudinal dataset analyzing whether using customer accounting systems for resource allocation objectives could be a source of long-term competitive advantage. Their findings suggested that introducing customer metrics as management accounting innovations provided temporary, rather than sustainable, long-term competitive advantage. Holm et al. (2016: 26) concluded that “adopting customer accounting systems constitutes one of a number of preconditions for successful customer asset management, and they serve as important facilitators of the reconfiguration and ongoing allocation of firm resources across customer relationships”.

3.2

Evaluation of the Literature

From the literature review, we argue that much of the prior studies have so far focused on models to better estimate CLV (Haenlein et al. 2007; Villanueva and Hanssens 2007; Kumar 2008) as well as the relationship between nonfinancial performance and the sustainability of a customer relationship (Ittner and Larcker 1998). Scant evidence is available on how customer centricity affects the way in which the elements of organizational architecture are combined and how in turn they impact managerial decision-making processes and business performance. Similarly, evidence of the use and effectiveness of innovative customer metrics for internal (e.g. budgeting and executive compensation) and external (e.g. financial reporting and valuation) accounting purposes is very limited (Guilding and McManus 2002; Ambler and Roberts 2008; McManus and Guilding 2008; Wiesel et al. 2008; McManus 2013). Overall, there is a lack of knowledge regarding the way the Accounting & Finance function support CMOs and marketing managers to verify the attraction, conversion and retention of customers through reliance on CLV and other advanced customer metrics. Such a gap in the extant literature is widely confirmed by a special issue published in the Journal of Marketing Management in 2008 on the interface between marketing and accounting (Ambler and Roberts 2008; El-Tawy and Tollington 2008; Gleaves et al. 2008; Inglis 2008; McManus and Guilding 2008; Phillips and Halliday 2008; Roslender and Wilson 2008). Those papers clearly noted inconclusive evidence and recommended further research to overcome the apparent divide between the two disciplines (‘Accounting is from Mars, Marketing from Venus’). While marketers have traditionally had a good grasp on customer metrics and modeling customer behavior, they frequently ignore the complexities and subtleties of adapting the use of these metrics in organizational processes and structures. On the contrary, it can be expected that accountants need to establish stable interfaces and dialogue with marketing colleagues to exploit the ability of customer-centric metrics to predict future financial performance as forward-looking indicators (Kumar et al. 2008; McManus and Guilding 2008). The next two sections provide insights into the use of customer metrics for internal decision-making and control. Through a qualitative case study and a survey among SBEs, we provide exploratory evidence to address this gap by examining specific

42

3 Customer Analytics for Internal Decision-Making and Control

relationships among the internal elements of organizational architecture tailored to customer centricity for decision-making and control. As a complement to this focus, Chap. 5 is dedicated to the literature that investigates the use of customer metrics for external purposes, namely corporate reporting and firm valuation.

3.3 3.3.1

A Case Study on the Adoption of Customer Analytics1 Case Background and Research Methodology

The objective of this case study was to generate insights into how a firm engages with the adoption and use of customer metrics. A qualitative research methodology was deemed particularly suitable for exploratory purposes (Yin 2003). We performed a single, in-depth longitudinal case study on an Italian-based mobile content provider. Our research site, Company.net (the firm’s real name has been disguised for confidentiality purposes), provides a range of digital products and services in the business-to-consumers mobile entertainment and community. Its corporate mission is to make customer users the central component of the Internet, allowing them to share content and create their own virtual space.2 The unique business proposition of Company.net is the convergence of a vast number of applications and a rich library of content (from entertainment/infotainment to social networking and music store) in a single, all-inclusive subscription model, accessible via personal computers as well as via mobile phones. Company.net is a typical example of the fast-growing subscription-based business model in which a customer pays a subscription fee to have access to a firm’s product/service. The model was pioneered years ago by magazines and newspaper publishers, but it is currently being used by a growing number of businesses and websites. Rather than selling products individually, a subscription-based firm sells periodic (monthly, yearly or seasonal) use or access to its products or services (Fader and Hardie 2009). The competitive environment of Company.net is characterized by an open technology which implies a low access barrier to enter the market. Even if the market is not strictly defined, competitors range from global brands like Apple to small startup companies. A key variable in this market is regulation, since it can heavily affect the business environment and differs across products (digital skill games, music, and ringtones) and countries. Company.net works with all major record labels supplying content and has agreements with all major carriers who deliver content to the network and bill customers on their mobile accounts. Customers are reached predominantly through paid search advertisements, using keywords such as ‘free-

1

This section is based on and include excerpts from pp. 253–267 of Bonacchi and Perego (2012). Company.net is the same firm presented in Chap. 2 to illustrate the cohort analysis in subscriptionbased enterprises. For further details about the case setting, readers are invited to complement the information provided in this section with Sect. 2.4. 2

3.3 A Case Study on the Adoption of Customer Analytics

43

ringtones’ or ‘free music’. Subscribers pay a fixed monthly subscription plan fee that provides the right to receive a specified number of content downloads (e.g. ringtones or MP3s). The industry value chain is characterized by the following activities: content origination; service management; marketing and display; network delivery; customer relationship management; and billing (exhibited in Chap. 2, Fig. 2.2). We believe Company.net is suitably representative for a single case study (Yin 2003) illustrative of the adoption of a customer-centric strategy for three reasons (Ramani and Kumar 2008). First, the mobile/internet platform represents an advance in technology that provides the opportunity to interact with an individual customer or with numerous customers, and takes advantage of the information obtained from them. Second, the firm is an example of a pure B2C firm where the customer is the starting point of every strategy and the unit of analysis. Third, and more importantly, the research company experienced a realignment of the three components of the firm’s organizational architecture to suit a subscription-based business (i.e. a business model whose orientation is by definition highly customer-centric). We conducted face-to-face semi-structured interviews with four company representatives—the President, the Chief Operating Officer (COO), the Business Intelligence Officer, and the Human Resource manager (HR)—identified as key participants in the firms’ strategy definition and design of organizational architecture at different levels. Because of the exploratory nature of our research approach, we developed a few introductory questions and then allowed interviewees to elaborate on them. This semi-structured protocol changed over time since we used each subsequent interview to triangulate the responses from previous interviews to gain insights on various elements of the organizational architecture. We were allowed to take notes and had access to all the participants immediately after an interview session to clarify any information that had taken place. We complemented the faceto-face interviews with reviews of archival company data, such as financial reports, corporate communications, worksheets and presentation slides made available during company visits.

3.3.2

Organizational Structure

Company.net is a typical new economy enterprise with a business environment characterized by radical developments both in the services/products offered and in the deployment of business processes. The interviews revealed that, operating in a highly complex and hypercompetitive market, progressively increased the interdependence among functional units and the need to coordinate the business activities around the customer axis. As Company.net embarked on improving its coordination mechanisms, experimentation and fine-tuning became critical to ensure appropriate alignment among the elements of the organizational architecture as a continuous and adaptive work-in-progress. The firm’s structure is customer-centric both in its primary activities, as well as in its support activities. Primary activities deliver superior customer value through the identification of three decentralized

44

3 Customer Analytics for Internal Decision-Making and Control

customer-segments or business units (Value Added Services, Music and Games). These activities are centered around the segmentation of the customer base with a platform that allows customization of the offer (i.e. each customer has the possibility to choose from a bouquet of products in a single point of contact), interaction with each single customer registered on the platform, and interaction among customers through blogs, forums, and so forth. With such an ‘outside-in’ orientation of the value chain (Gulati 2009), specialization on customer-segments provides the advantage of clarifying decision rights, customizing offerings of services in a timely fashion and identifying new or market opportunities previously unexploited. As a potential drawback of such business model, interviewees pointed out the risk of creating ‘legacy silos’ that tend to preserve function autonomy but create unnecessary overlays that makes decision-making processes more cumbersome. The top management realized the importance of removing roadblocks to coordination of silos through a number of mechanisms in which the support activities play a central role. In particular, Company.net created a Business Intelligence (BI) unit in line of the COO. Such formal appointment made customer centricity one of the most strategic issue discussed in the boardroom (Fig. 3.2). The BI unit therefore shifted role in a radical way. Originally formed as a highly specialized function devoted to delivering reliable customer data, the BI unit plays now a fundamental role in Company.net’s architecture on two fronts. First, it developed a centralized customer information warehouse to promote unified access to customer data and encourage strategically collaborative sharing and use of customer data across business units. The BI unit ensures integration of multiple data sources on customers’ transactions and applies data management solutions and predictive analytics to understand subscribers’ behavior, usage characteristics and application performance. Second, the BI unit provides crucial service assurance solutions that enhance the firm’s internal reporting quality. Furthermore, BI proactively manages and strive to prevent the risk of data manipulation, a threat particularly evident in the mobile/Internet industry where data fragmentation and isolated data gathering makes data consolidation a major challenge. In sum, Company.net positioned the BI unit as a support unit of ‘trusted advisers’ who deliver data solutions to the customer segments and customer profitability analysis, helpful for strategic decisions on customer investment, resource allocation

Fig. 3.2 Organizational structure of Company.net (source: Bonacchi and Perego 2012)

3.3 A Case Study on the Adoption of Customer Analytics

45

and cross-segments opportunities (cf. Bonacchi and Perego 2012). The BI unit operates at the intersection between legacy silos to ensure coordination instead of friction, and deploys a culture of cross-functional collaboration. The presence of a centralized information-sharing infrastructure retains the benefits of decentralized decision-making such as deep product knowledge or economies of scale (Gulati 2009). At the same time, it preserves the ability to deploy customer-centricity across organizational functions.

3.3.3

The Performance Measurement System

Performance measurement system (PMS) shapes and is shaped by a strategy (Chapman 1997; Chenhall 2003; Kober et al. 2007). The three pillars of Company. net business strategy consist of (1) acquire new customers: characterized by aggressive marketing techniques of customer acquisition aimed at building up a new user base; (2) retain existing customers: focused on measuring the lifetime of users, stimulate user retention and minimize churn rate; (3) an organic growth of the user base. The firm defines the new target acquisition number for each period (e.g. monthly or quarterly) in order to balance churn rate and reach the target growth rate. From the evidence collected from face-to-face interviews and archival data, the PMS in Company.net is designed around two main objectives: 1. To support managers in their tactical and strategic decisions on managerial decisions like: (a) How much should be spent on customer acquisition? (b) How to allocate the acquisition budget among different acquisition channels? (c) What are the characteristics of the best prospects? 2. To diagnose the health of the business evaluating the expected residual value of the customer base currently on the firm’s books. In order to reach the main objectives, the PMS in Company.net is built around the notion of CLV, and its drivers (regarding the first objective), and CE (regarding the second objective). The main issue is the measurement of CLV since CE is a function of CLV and number of customers. Two estimation activities are implemented for CLV (refer to Fig. 3.3 that reproduces the CLV Scorecard already exhibited in Chap. 2). On the one hand, future cash flows from the customers (aka Lifetime Value LTV) are forecasted; on the other hand, Cost of Acquisition (CoA) is calculated by dividing the expenses related to advertising to acquire new subscribers by the number of customers acquired in the considered unit of time. In synthesis, the comparison between CoA and LTV is crucial for players in this market.

46

3 Customer Analytics for Internal Decision-Making and Control Measures ARPU CONTRIBUTION MARGIN

YIELD Cost of service

LIFETIME VALUE (LTV) Instant churn LIFETIME Customer Lifetime Value (CLV)

Historic churn

Attraction COST OF ACQUISITION (CoA)

Conversion Cost of contact (CPC, CPA)

feedback

Fig. 3.3 Customer lifetime value (CLV) scorecard of Company.net (source: Bonacchi and Perego 2012)

CLV ¼ LTV  CoA In order to operationalize the measurement of LTV, its main drivers (i.e. lifetime and gross margin) must be estimated. The mathematical relation among lifetime value and its drivers is the following: Lifetime Value ¼ lifetime  margin per customer where • Margin per customer ¼ Average Revenue per Users (ARPU) less cost of service per customer. • Lifetime ¼ 1-churn • Churn ¼ rate of attrition (cancellations/average users per period) over a period of time. The abovementioned drivers are leading, forward-looking indicators and must be monitored on a daily basis to verify that a CE creation process is in place. The peculiar nature of the business, in which all transactions are made online and the log files of each transaction are constantly registered, forces Company.net to apply a timely monitoring of these metrics in order to judge the effectiveness of a marketing investment.

3.3 A Case Study on the Adoption of Customer Analytics

3.3.4

47

The Reward System

The reward system at Company.net relies on the rationale to maximize the profitability of the company by recognizing and rewarding accordingly the key elements that drive customer profitability. Consistently with this logic, the annual managerial incentive plan is formally linked to business and individual performance. The size of the bonus pool depends on two families of metrics: 1. Financial performance, with an emphasis on metrics like EBITDA. 2. Nonfinancial performance: focused on metrics that track net addition, number of downloads and churn. We derived from field evidence that the current performance evaluation system is a top-down process that occurs on an annual basis although the operational budget follows a bottom-up approach. The bonus is supposed to translate in nonfinancial terms the targets quantified periodically in the financial budget. The weight of customer metrics is on average about 30–40% relative to other performance indicators outlined in the managerial scorecard. From the current debate occurring within top management, it seems that Company.net recognized the value inherent in the use of non-financial measurement for incentive systems, however the formalization of the calculation of bonuses still requires a better definition of the underlying rationale. When it was first introduced, the bonus system raised concerns about the way some metrics were measured. For instance, the metric ‘net addition’ refers to both free subscribers and paid subscribers. Another challenge was the need to calibrate the metric for each product offered and align the key performance indicators with the business strategy. For example, in the online music business, key performance indicators are different from other segments like games. In the first case, tracking subscriptions is rather straightforward and allows a relatively easy computation to judge CLV. On the contrary, the cost of customer acquisition for the games segment requires a more complex estimation of the possibility that the player is retained in the future, while controlling for the inherent cost of acquiring an occasional player. Regarding the alignment between a choice of performance measures and a business strategy, by the time a new product is introduced the company is eager to expand the diffusion of the product (in technical jargon, Company.net wants the product to become ‘viral’). Once the customer base is sufficiently large and established, maintaining profitable customers then becomes a priority. For instance, in the first stage, focusing on a metric like gross addition (both free and paid customers) is crucial, while in the subsequent phase the focus shifts to additional indicators like the number of customers actively paying, ARPU and churn. As emphasized in Bonacchi and Perego (2012), “given the business environment in which Company.net operates, timeliness of the metrics is another crucial factor. In theory, the incentive system should be designed with frequent performance appraisal occurring at quarterly/monthly intervals. In practice, this choice is still too administratively cumbersome for the organization”.

48

3 Customer Analytics for Internal Decision-Making and Control

The case evidence suggests that in such dynamic and complex business environments as mobile/Internet services, “there is always a trade-off between accuracy of the information and practical implementation of the incentive system. Threats to accuracy in the information may still occur despite adaptation of the organizational architecture” (cf. Bonacchi and Perego 2012). For example, companies may face silo data in each division with different storage locations. Such a ‘data mart’ approach based on multiple data centers hampers suitable transfers and coordination. Further, different granularity of the data in different domains or obsolete data reporting represent a potential impediment to apply a consistent approach in the selection of appropriate performance indicators.

3.3.5

Conclusions and Implications from the Case Study

In the case study presented in this section, we examine whether and how a successful customer-centric organization affected and realigned a firm’s organizational architecture. In Table 3.1, we summarize the main recommendations emerging from our case study to successfully build a customer-centric strategy (cf. Bonacchi and Perego 2012). The first logical stage of this organizational realignment is the emergence of coordinating activities that aim to overcome the traditional deficiencies of productcentric structures. In the case company, the Business Intelligence unit not only ensures the information flow of individual customer data. It collects, certifies and communicates customer-centric data vertically to top management and the board of directors (upward) and to the business units (downward) with high levels of assurance, although with a different degree of granularity and timeliness. The Business Intelligence unit is illustrative of the implementation of an organizational structure in which a centralized information warehouse infrastructure deploys reliable information to highly decentralized business units. Such choices in the organizational design enable an especially synergistic relationship between marketing and accounting functions, by preventing the formation of functional silos, and enhancing coordination to support customer-centric decisions. The allocation of decision-rights is therefore a crucial starting point to build a customer-centric organization; however, specific organizational arrangements must be tailored to individual firms and business contingencies (Gulati 2009). Our field evidence further reveals the complementary role of a performance measurement system centered on forward-looking and leading indicators, like CLV and CE. Knowing the CLV of individual customers enables managers to improve the customer segmentation and marketing resource allocation efforts; this in turn leads to higher retention rates and profits for the firm. The analytical literature in marketing is constantly evolving in an attempt to refine CLV models to undertake customer base analysis activities that are more forward-looking (or predictive) in nature (please refer to Chap. 2 of this book; Kumar et al. 2006). In dynamic business environments, cause-and-effect relationships are

3.3 A Case Study on the Adoption of Customer Analytics

49

Table 3.1 Comparison of the product-centric and customer-centric approach: implications for a firm’s organizational architecture Organizational architecture— elements Structure and allocation of decision rights

Product-centric approach – The need of coordination is seen as being limited to the functional departments within a firm (functional silos) – Functions are typically structured in: product profit centers, product managers, product sales team

Performance measurement

– Customer data are available – Reliance on backward-looking, historical metrics such as: number of new products, profitability per product, market share by product/sub brands – Information focused on products – No link between customer metric and shareholder value

Rewards and incentive system

– Lack of formal appraisal systems containing customer metrics – No link between customer metrics, employee target setting and compensation systems

Customer-centric approach – The need of coordination is extended to all functions – Creation of functions specializing in relationship management, e.g., Chief Customer Officer – Functions are typically structured in: customer segment centers, customer relationship managers, customer segment sales team – Customer knowledge is a valuable asset – Reliance on forward-looking metrics such as: customer shareof-wallet, customer satisfaction, customer lifetime value, customer equity – Need of information sharing regarding customers – Drivers of customer equity are known – Reliance on formal appraisal systems containing customer metrics – Explicit link between customer metrics, employee target setting and compensation systems

Source: Bonacchi and Perego (2012)

constantly changing and the ability to generate inference from past experience is severely challenged. Organizations are therefore required to engage in active experimentation on these predictive models, particularly through investment in customer intelligence and business analytics. From the company investigated, “what seems important for all firms, notwithstanding their size or development stage, is building a centralized customer information warehouse. This serves as the means to provide a unified, comprehensive, and organization-wide view of individual customers, irrespective of the products purchased or channels employed by the customer” (cf. Bonacchi and Perego 2012). This entails a substantial IT investment commitment to set up an infrastructure for collecting, tracking, and integrating data at the individual customer and transaction level. Jayachandran et al. (2005) specified several systems-related activities that allow customer-centric firms to successfully build a long-term relationship with

50

3 Customer Analytics for Internal Decision-Making and Control

their customers. These activities include: creating a database that contains all exchanges with the customer; making the integrated database accessible to those responsible for managing the customer relationship; and using the database to analyze past performance with the goal of understanding the drivers of customer behavior (Shah et al. 2006). Finally, we illustrate how formal reward systems (both monetary and non-monetary) are a powerful mechanism to steer employees’ decisions toward customer-centricity and align their decisions/performance to maximize customer equity (Banker et al. 2000; Widener et al. 2008). In practice, however, the case company reveals several challenges inherent in the design of appropriate managerial incentive and compensation systems. Bonacchi and Perego (2012) conclude that “firms should first focus their attention on aligning their organizational structure and performance measurement system before modifying their incentive systems. Such a positive reaction to customer-metrics embedded in reward systems depends heavily on the ability of the organizational structures and performance measurement systems to deliver a reliable and timely information flow of these metrics”. In conclusion, the organizational architecture introduced in Chap. 1 provides a useful conceptual framework and practical insights to venture capitalists, entrepreneurs, start-ups and established companies aiming at building a successful customercentric business in the new economy. The three organizational elements examined in this study and in this book provide the building-blocks for full-scale transformation of a company’s operating model and point out the priorities of its executives’ agenda. This study illustrates an exploratory roadmap of how customer-centric firms can succeed in a highly competitive market, where services are quickly becoming a commodity and margins tend to shrink. Since our evidence is collected by a single case company in a specific industry (B2C subscription-business model), the usual limitations apply. Further empirical data is nevertheless necessary to test the generalizability of our findings and corroborate our exploratory case analysis. As an extension of this case study, we conducted an exploratory survey on a sample of SBEs to examine the relationships among the three elements of the organizational architecture and their effects on organizational performance. We argue that firms with a higher alignment of organizational architecture for a customer-centric strategy should likely outperform industry peers. We present the research design and the findings of the survey in the next section.

3.4

An Exploratory Cross-Sectional Survey on the Adoption of Customer Analytics

The study relies on data collected using a survey-based methodology following Dillman (2011). The web-based questionnaire was piloted with faculty colleagues and field-tested with a selection of potential informed respondents from practice. Minor alterations were made in each step. Table 3.2 exhibits the list of variables

3.4 An Exploratory Cross-Sectional Survey on the Adoption of Customer Analytics

51

Table 3.2 Variable definitions Variable acronym AVAa

Variable description Customer accounting metrics— availabilitya Customer accounting metrics—usage Use of customer metrics for planning and control Use of customer metrics for compensation Use of customer metrics for external reporting Use of customer metrics for valuation

Question no. (see Appendix) Q1.

Q2. Q3.–Q4. Q5. Q6.

Customer accounting–organizational structure Function in charge of customer Q7.–Q10. accounting data Ownership of customer accounting data Q11.–Q12. Accuracy of customer accounting data

Q13. Q14.

STRAT

Integration characteristics between Marketing and Accounting & Finance (A&F) functions Information sharing between Marketing and Accounting & Finance (A&F) functions Strategic customer-centricity orientation

Q17.

UNCER

Customer exchange characteristics and heterogeneity customer base Perceived environmental uncertainty

PERF

Organizational performance

Q19.

INTEG

SHAR

Source Literature review, various sources Literature review, various sources Literature review, various sources Literature review, various sources Literature review, various sources Literature review, various sources Barker (2008); Song and Thieme (2006) Literature review, various sources Song and Thieme (2006)

Q15.

Song and Thieme (2006)

Q16.

Ramani and Kumar (2008) Reinartz and Kumar (2003) Ramani and Kumar (2008); Song and Thieme (2006) Deshpandé and Farley (1998); Ramani and Kumar (2008)

Q18.

Note: The table exhibits all of the variables measured by the survey items reported in the Appendix of this chapter. The descriptive statistics and bivariate correlations of the variables in bold are summarized in Table 3.3 a This is a dummy variable capturing the availability of the following metrics (refer to the Appendix): Number of customers (AVA_NUM), Usage or traffic (AVA_USA), Gross customer additions (AVA_GCA), Net acquisitions (AVA_NET), Average Revenue Per Unit (AVA_ARPU), Churn/Retention (AVA_CHURN), Cost of service (AVA_COS), Cost of customer acquisition (AVA_COA), Customer Equity (AVA_CE). For each metric, the dummy takes the value of 1 when a metric is available and 0 otherwise

3.46

3.51

3.52

2.80

2.65

2.59

10. INTEG

11. SHARE

12. STRAT

13. UNCER

14. PERF_BENCH

15. PERF_FIRM

1.10

1.11

0.93

0.99

1.10

0.94

0.48

0.48

0.35

0.28

0.28

0.41

0.841

0.930

0.920

0.858

0.849

ICR

0.668

0.911

0.870

0.756

0.774

Alpha

0.267 0.298a

0.353b 0.489a 0.177 0.094

0.081 0.160 0.369a

0.020

0.300b

0.177

0.391a

0.346b

0.240

0.172

0.108

0.252

0.269

0.272

0.304b

4.

0.252

0.258

0.175 0.030

0.269 0.409a

0.137 0.156

0.272

0.050

0.456a 0.304b

0.034

0.034

3.

0.062

2.

b

0.115

0.137

0.187

0.218

0.353b

0.282

0.236

0.560a

0.462a

0.293b

5.

0.374a

0.307b

0.344b

0.176

0.271

0.490a

0.281

0.088

0.039

6.

0.099

0.137

0.360b

0.459a

0.316b

0.262

0.339b

0.681a

7.

0.081

0.082

0.363b

0.339b

0.463a

0.264

0.458a

8.

0.731

0.174

0.139

0.334b

0.292b

0.345b

0.372a

9.

0.404a

0.227

0.369a

0.244

0.606a

0.817

10.

0.591a

0.189

0.428a

0.418a

0.891

11.

0.413a

0.300b

0.691a

0.810

12.

0.352b

0.343a

13.

0.429a

0.853

14.

Note: The correlations among AVA_NUM and the other variables is not reported because AVA_NUM is constant across the whole sample (n ¼ 49). Bold-faced elements on the diagonal represent the square root of the Average Variance Extracted (AVE) of the five variables analyzed in the PLS model. ICR, Cronbach’s Alpha and AVE of the variables AVA_CE and PERF_BENCH refer to the variables METRICS and PERF in the PLS model. Off-diagonal elements are the Pearson correlations between variables a Correlation is significant at the 0.01 level (two-tailed) b Correlation is significant at the 0.05 level (two-tailed)

0.43

9. AVA_CE

0.47

0.84

6. AVA_CHURN

0.67

0.86

5. AVA_ARPU

0.65

0.92

4. AVA_NET

7. AVA_COS

0.92

3. AVA_GCA

8. AVA_COA

0.37

0.80

2. AVA_USA

S.D.

1. AVA_NUM

0.00

Mean

1.00

Variables

Table 3.3 Descriptive statistics and correlations

52 3 Customer Analytics for Internal Decision-Making and Control

3.4 An Exploratory Cross-Sectional Survey on the Adoption of Customer Analytics

53

investigated and the literature sources. The full list of survey items is reported in the Appendix of this chapter. The questionnaire elicited information regarding the availability of customer metrics by proposing a list of the ten most common indicators from the marketing and accounting literature (variables labeled with the acronym AVA; Q1.). Further, we inquired into the usage of CLV and other customer accounting metrics for internal (i.e. decision-making and performance evaluation) purposes. Respondents were asked to indicate whether customer metrics were used at different firm levels (top management; business unit managers; sales managers; middle managers; and customer care employees) for setting periodic targets and presenting monetary rewards in the case of target accomplishment (Q2.–Q4.). We additionally included a series of questions about the usage of customer metrics for external purposes (i.e. financial reporting and valuation) with the goal of assessing the adoption of these indicators for uses other than managerial accounting (Q5.–Q6.). The questionnaire then focused on the organizational structure and other characteristics that shape the interface between the Marketing and Accounting & Finance functions (Q7.–Q15.). We specified a series of items to identify the integration of customer accounting information (INTEG; Q14.) and the dynamics that shape the interaction (SHARE; Q15.) between these two functions. While largely drawing upon the marketing literature, we developed or replicated previous scales to shed light on the relationships currently occurring at the interface between Marketing and Accounting & Finance. Finally, we measured variables capturing the strategic orientation of the organization to customer centricity (STRAT; Q16.). Items regarding environmental uncertainty (UNCER; Q18.) and organizational performance (PERF; Q19.) came at the end of the survey, with the intention of exploring the links between external factors and the degree of adoption/level of customer accounting (function, experience) and the characteristics of the sampled organizations.

3.4.1

Sample and Data Collection

In this survey, we focused on companies in a contractual or subscription-based business setting for a Business-to-Consumers environment. We were interested in gauging the usage of customer metrics for companies in which a customer pays a subscription fee to have access to a firm’s products or services. By restricting our attention to SBEs, we expect to enhance the likelihood of detecting leading customer accounting practices and associated customer metrics. We believe this research design to be appropriate for two reasons (Dubosson-Torbay et al. 2004; Lyons et al. 2009). First, in a contractual setting, it was easier to overcome one of the major obstacles to managing customers using CLV techniques, namely, the availability of data tracking and data mining of customer metrics. Second, focusing on a single business model provided depth to the study and allowed us to address specific aspects of the organizational architecture in customer-centric organizations. Although the scope of our research was limited to SBEs, we believe the findings

54

3 Customer Analytics for Internal Decision-Making and Control

could also be generalizable in a noncontractual setting and would contribute to the development of a management accounting system for customer-centric companies operating in other settings. Our sampling strategy relied upon the advanced search function on LinkedIn. We initially looked for professionals mentioning the keywords “ARPU” and “Churn” in their LinkedIn profiles. We then proceeded to match a profile with the firm in which the professional was employed to ascertain the contractual setting of the firm. In case of a match, we prepared a letter of invitation with a brief explanation of the aim of the survey study. The letter contained a personalized link that the respondent had to click to access a web-based questionnaire. This method allowed us to track the respondent and match them and the firm s/he worked for. The use of a web-based questionnaire facilitated the administration of interactive and open-ended questions and was considered the most suitable research instrument to collect data among the invited contacts. We sent out 512 invitations to professionals with no geographical restrictions. We obtained 51 responses (10% response rate). After cleaning the database of unusable or partial responses, the results outlined in this paper are based on 49 respondents. One-third of the sample represented the wireless industry, with half of the respondents employed in companies headquartered in the Netherlands. The largest portion of survey participants worked at the headquarters level (43%), while 30% were employed in the foreign subsidiaries of multinationals.

3.4.2

Descriptive Statistics and Univariate Analysis

As reported in Table 3.3, while metrics such as Number of Customers (AVA_NUM) seem widely diffused, our descriptive findings revealed that 43% of the companies surveyed utilize CE (AVA_CE). Interestingly, 30% of the respondents reported adaptations of customer-related metrics, for the most part industry-specific, thereby suggesting cross-sectional variation among current practices in this area. Our findings thus seemed to suggest the wide diffusion of CLV and other metrics, once available in a firm’s accounting or information system, across units and hierarchical functions. Additionally, for respondents claiming the availability of customer metrics, our data (not tabulated here) suggested that on average their internal reporting systems were most commonly updated on a monthly basis and widespread at all company levels. Our results further revealed that 71% communicated the Number of Customers more commonly in corporate financial reports than in press releases.3 Reporting of the Churn rate and CE was less frequently adopted (55% and 27%, respectively).

This evidence is consistent with the findings in Bonacchi et al. (2015) in which conference calls and analyst reports do not reflect customer-related data beyond those available in the company SEC filings. 3

3.4 An Exploratory Cross-Sectional Survey on the Adoption of Customer Analytics

55

A deeper analysis of the use of metrics for valuation purposes indicated that 73% relied on the Number of Customers for corporate valuation, which decreased to 47% for M&A’s and impairment purposes (e.g. customer list impairment). A substantially lower use was reported for CE, with 37% of respondents indicating it was used for corporate valuation, while only 27% and 20% for M&A’s and impairment, respectively. Among the companies surveyed that measured and relied upon CE in their decision-making systems, responses indicated that 31% of the firms investigated relied upon average customer data, while 18% elaborated data from single customers. Among the possible methods that can be used to compute CLV, our findings further suggested the most frequent was associated with so-called deterministic models (50%), i.e. simplified calculations that ignored heterogeneity in customer retention and/or churn rates within a cohort, followed by statistical approaches based on survival functions (15%). For 12% of respondents, CLV was calculated using a 1/churn approach, thereby highlighting significant cross-sectional variation in terms of adherence to potential estimation models from the Marketing literature. Finally, the large majority of companies seemed to use customer tenure to develop their segmentation strategies, with information regarding customer gender and residence type as the least diffused among the respondents. For the organizational arrangements of customer accounting, a large percentage (80%) of companies formally appointed an ad hoc unit or organizational function. ‘Business Intelligence’ and ‘Customer Lifetime Management’ were the titles most frequently cited by the respondents as labels for these functions. Among these units, 16% of staff reported corporate functions, 34% were in line with the Accounting & Finance function, and 8% were in line with the Marketing function. Regarding the type of software package or enterprise information systems utilized to measure and report customer metrics, customized solutions were present in 53% of the firms surveyed. It is noteworthy that the ownership of customer metrics data was highest for the ad hoc function (39%), followed by Marketing (25%) and P&C (20%), thereby suggesting a marginal role of the Accounting & Finance function in managing the process of generating CLV data and other related customer metrics. The company function cited most often as normally in charge of providing assurance to these metrics was nevertheless the CFO, thus signaling an auditing role for the Accounting & Finance function rather than the main user/owner of these types of data. Survey participants were asked to indicate whether customer metrics were attached to a firm’s target-setting policy. Two-thirds of respondents reported that the Number of Customers was linked to a target (more specifically, 50% with an average monthly frequency). Only one-third of the companies adopted the approach of formally linking CE results with a pre-specified company target. Moreover, it appeared CE reliance was equally distributed among monthly, quarterly and yearly performance evaluation cycles. It is worth noting that only one-quarter of our sample linked an explicit monetary bonus to the achievement of a targeted metric. Such a relationship was diluted even more (by one-fifth) in the case of an explicit CE target. Overall, our findings signaled a weak alignment of customer accounting in firm’s incentive and compensation systems.

56

3 Customer Analytics for Internal Decision-Making and Control

Furthermore, our exploratory findings suggested the integration between Marketing and Accounting & Finance (INTEG) had a positive and statistically significant link to the usage of specific customer accounting metrics, namely, AVA_ARPU (r ¼ 0.282; p < 0.05) and AVA_CE (r ¼ 0.372; p < 0.01), while the bivariate correlation among all of the other metrics were not significant. Further, the data clearly indicated a significant relationship among the degree of sharing of customer information between Marketing and Accounting & Finance (SHARE) with all metrics having a correlation between r ¼ 0.316 and r ¼ 0.463 (see Table 3.3). Not surprisingly, the correlation between degree of sharing (SHARE) and the degree of integration (INTEG) was positive and statistically significant (r ¼ 0.606; p < 0.01), thus lending support to the expectation that both these aspects moved in the same direction. We then used our survey results to test the arguments in contingency-based research positing an effect between environmental factors and strategic choices in shaping the organizational architecture of the firm. More specifically, it was expected that firms adapted their corporate strategies and internal structures to respond more effectively to shifts in market, technology or regulatory changes. We found evidence that both the antecedents of the integration and sharing of customer accounting was significantly associated with UNCER (r ¼ 0.369; p < 0.01 for integration; r ¼ 0.428; p < 0.01 for sharing). In other words, customer-centric firms more exposed to changes from the business environment tended to adapt more quickly in aligning their organizational structure. Nevertheless, the degree of customer orientation (STRAT) was only significantly associated with the degree of sharing between Marketing and Accounting & Finance (r ¼ 0.418; p < 0.01), but was not significantly associated to integration. These exploratory findings illustrated the cross-sectional variation still present in the degree of integration between Marketing and Accounting & Finance. Finally, we explored the consequences of modified organizational architecture on performance by testing the correlation among organizational structure (degree of integration and sharing) and two perceptual measures of firm performance (measured relative to a firm’s main competitors and last fiscal year; Q19). The results indicated a positive effect in a firm’s performance compared to the previous year (PERF_FIRM r ¼ 0.591; p < 0.01 for the degree of sharing and r ¼ 0.404; p < 0.01 for the degree of integration).

3.4.3

Multivariate Analysis

In an attempt to explore the interplay among variables, we relied on Partial Least Squares (PLS) because of its unrestrictive assumptions over alternative structural modeling techniques (Chin and Newsted 1999; Haenlin and Kaplan 2004). We followed the procedure advocated by Chin (1998) and Hulland (1999) by first analyzing the assessment and reliability of the measurement models and then testing the significance of the path coefficients in the structural model. The factor loadings of items associated with each construct were high, equaling or exceeding the threshold of 0.707 (Carmines and Zeller 1979). Table 3.3 reports the reliability

3.4 An Exploratory Cross-Sectional Survey on the Adoption of Customer Analytics

57

and the correlation matrix among variables obtained from SmartPLS 2.0. Convergent validity was assessed using the internal composite reliability (ICR) (Fornell and Larcker 1981). All ICRs equaled or exceeded 0.7, indicating satisfactory reliability of all constructs in the model (Hulland 1999). Additional support for convergent validity was provided by Cronbach’s Alphas, with the minimum value at 0.76 for INTEG. Discriminant validity was assessed using Average Variance Extracted (AVE), with all AVEs equaling or exceeding 0.50. As demonstrated in the rows and columns of Table 3.3, the indicators exhibited a higher loading with their construct than with any other construct, thus lending evidence to the discriminant validity in that the variance shared between any two constructs was less than the average variance extracted by the constructs. Figure 3.4 reports the results of a PLS model in which we aggregated five items in the variable labeled METRICS that captured the availability of ARPU, churn, cost of service, cost of acquisition and Customer Equity. For the sake of model parsimony, we aggregated the two performance measures (PERF_BENCH and PERF_FIRM) in a single construct labeled PERF. PLS generated standardized beta coefficients that could be interpreted as ordinary multiple regression analysis. To ascertain the stability and significance of the parameter estimates, we computed the t-values on the basis of 1000 bootstrapping runs. Our results indicated that the customer-centric strategy (STRAT) positively affected a firm’s performance through an increase in the use of customer metrics (METRICS) and sharing of customer-related information between Marketing and Accounting & Finance functions, respectively. The positive effect was fully mediated by the degree of integration (INTEG) between those two functions. Our exploratory findings are in line with the rationale of an aligned organizational

SHARE R2 0.302

0.235

0.502***

0.625*** 0.400***

STRAT

-0.018

INTEG

0.234*

R2 0.419

0.057

0.461***

PERF R2 0.369

-0.175

METRICS R2 0.212

*** significant at 1%-level; ** significant at 5%-level; * significant at 10%-level. Dashed lines refer to relationships not statistically significant.

Fig. 3.4 Results of the multivariate PLS model. The descriptive statistics of the variables STRAT, SHARE, INTEG and PERF are reported in Table 3.3. METRICS is an aggregate variable that captures the average availability of five metrics, namely, Average Revenue Per Unit (AVA_ARPU), Churn/Retention (AVA_CHURN), Cost of service (AVA_COS), Cost of customer acquisition (AVA_COA), and Customer Equity (AVA_CE)

58

3 Customer Analytics for Internal Decision-Making and Control

architecture, in which the deployment of customer strategy requires not only a shift in the performance measurement system (focused on the availability and sharing of advanced customer metrics) but also on enhancement of coordination/integration among organizational functions. Although limited to a small sample of subscriptionbased firms, the PLS analysis revealed that the availability of metrics such as CLV or CE were a necessary but insufficient condition to exploit higher levels of performance, signaled by the lack of a significant direct effect between METRICS and PERF.

3.4.4

Conclusions and Implications from the Survey

We administered an Internet-based survey to investigate the level of adoption of customer accounting and its implementation in an organizational architecture among a sample of SBEs. We were particularly interested in documenting the degree of integration and sharing between the Marketing and Accounting & Finance functions because this link has not been explored in past research. We posit that the interface between Marketing and Accounting & Finance is crucial in ensuring the adequate application of recent managerial instruments and tools underlying customer accounting. In the absence of high degrees of integration and sharing, it can be expected that customer metrics remain unused or are not effectively applied in daily decisionmaking processes, thereby undermining the long-term profitability of a customercentric firm. From our findings, it appeared that firms with a subscription-based model were increasingly embracing the use of CLV and other advanced customer metrics. While the adoption of metrics such as Number of Customers appeared widespread, a substantial number of respondents reported Customer Equity as the metric with the lowest degree of diffusion. The average weak link between CE targets and performance evaluation/bonus made the penetration of this key metric more difficult to achieve compared to other metrics. As a consequence, although the availability of customer data was reported to be high, responses indicated that there was no fit in terms of appropriately exploiting this crucial information for strategic analysis. In particular, we found evidence of a weak integration of these metrics in compensation and incentives systems. Moreover, the majority of companies seemed to struggle in their approach to estimating the financial impact of specific customer-related decisions, particularly in the medium and long-run. A possible cause of this lack of fit was identified in the relatively low ownership of customer metrics by the Accounting & Finance function. While a majority of firms had an ad hoc approach to the measurement and reporting of key customer information, so-called Business Intelligence units, the use of predictive customer-based analytics was still in the early stages for management accountants and financial managers. When examining the degree of integration and sharing of customer information between Marketing and Accounting & Finance functions, our survey data analysis revealed the cross-sectional variance of practice associated with integration. Some

Appendix Chapter 3: Questionnaire

59

metrics (ARPU and CE) showed privilege compared to others as soon as the integration level between the two functions increased. On the contrary, the degree of sharing was positively associated with all of the metrics investigated, highlighting that a necessary requirement to adapt a customer-centric performance measurement system should rely on a closer and more frequent exchange of information between Marketing and Accounting & Finance functions. Finally, a multivariate analysis supported the rationale of an aligned organizational architecture, in which the deployment of customer strategy required not only an adaptation in the performance measurement system (focused on availability and sharing of advanced customer metrics) but also an enhancement of coordination/ integration among organizational functions. Our exploratory data revealed that the availability of metrics such as CLV or CE is a necessary but insufficient condition for a customer-centric strategy to deliver higher levels of performance. Therefore, it can be inferred that firms need to pull the right levers of organizational architecture to fully exploit customer centricity. Our findings shed additional light on the adoption of customer accounting information; however, they should be interpreted cautiously due to the limited number of observations from the survey. Another caveat of this study is the selection of potential respondents based on a subscription-based model. Future research should extend the data collection to firms in other settings to validate the generalizability of the model proposed. Similarly, our study suffers from potential commonmethod bias due to the presence of only one respondent represented by a management accountant/financial professional. It might be interesting to administer the questionnaire to marketing managers with the aim of better understanding the interface between Accounting & Finance in this area. Finally, the effects of customer accounting on organizational performance explored in this paper rely on the perceptions provided by the respondents. It would be advisable to rely upon objective data by matching sampled firms with the financial performance data publicly disclosed in their respective annual financial statements.

Appendix Chapter 3: Questionnaire The survey items and scales used in the online questionnaire are reported next (refer to Table 3.2). Customer Accounting Metrics: Availability Q1. Indicate to what extent the following customer metrics are available in your information system. (Check all that apply). (a) (b) (c) (d)

Number of customers (AVA_NUM) Usage or traffic (AVA_USA) Gross customer additions (AVA_GCA) Net acquisitions (AVA_NET)

60

3 Customer Analytics for Internal Decision-Making and Control

(e) (f) (g) (h) (i) (j)

Average Revenue Per User (AVA_ARPU) Churn/Retention (AVA_CHURN) Cost of service (AVA_CoS) Cost of customer acquisition (AVA_CoA) Customer Equity (AVA_CE) Others (open answer)

Customer Accounting Metrics: Usage This section relates to the use of customer metrics for various purposes. Q2. With respect to the current performance measurement system in place, check all of the metrics listed in question 1 that are used to gauge your firm’s performance. Please also indicate the frequency with which the following metrics are reported and analyzed for the purposes of the internal performance measurement of your firm. (a) Extent and frequency of use per function (Top management, BU manager, sales manager, middle managers, customer care employees; monthly, quarterly, or yearly). Q3. We would like to know whether your firm uses the customer metrics listed in question for performance evaluation purposes. Please indicate whether a formal target is attached to the following metrics and the frequency of the evaluation. (a) Target attached and evaluation frequency per function (Top management, BU manager, sales manager, middle managers, customer care employees; monthly, quarterly, or yearly). Q4. Some firms reward employees with a monetary bonus in case targets are achieved. Please indicate the absolute weight per metric used to assign a bonus to the following functional levels. For instance, a weight of 50% for the BU management means that a metric determines half of the bonus attached to that functional level. (a) Relative weight of bonuses per function (Top management, BU manager, sales manager, middle managers, customer care employees; in percentage of total bonus). Q5. Indicate the way in which information is provided about the metrics listed in question 1 for external reporting purposes. (a) (b) (c) (d) (e)

Conference call Industrial plan IPO prospectus Press releases Financial report

Appendix Chapter 3: Questionnaire

61

Q6. Indicate whether your firm uses the customer metrics listed in question 1 for the following valuation purposes. (a) Internal valuation of the company (b) Valuation in the context of an acquisition (c) Valuation required for the impairment test of the customer list Customer Accounting—Organizational Structure This section addresses the organizational arrangements that relate to the ownership and sharing of customer metrics. Q7. Please indicate whether your company appoints an Organizational Unit/ Function (OU) in charge of customer accounting metrics. (a) No specialized unit/function (b) If yes: Indicate title of the OU (i.e., Chief Customer Officer, Business Intelligence, Customer Profiling, etc.) Q8. Please indicate how this OU is positioned in your firm. (a) Is this OU in the corporate staff? (b) Is this OU in line with the Accounting & Finance (A&F) function? (c) Is this OU in line with the Marketing/Sales function? Q9. How many people are employed in this unit? Q10. What was the budget for this OU in the last fiscal year (US $)? Q11. Which of the following functions is the owner of the customer metrics data? (a) Corporate (b) Marketing (c) Accounting & Finance (A&F) Q12. To what extent can [the 2 non-owner functions] have direct access to the customer metrics data? (1 ¼ no extent and 5 ¼ great extent) Q13. To what extent do the following functions provide assurance as to the accuracy of the customer metrics data? (1 ¼ no extent and 5 ¼ great extent)

(a) In my firm, the COO provides assurance as to the accuracy of customer metrics data. 1 2 3 (b) In my firm, the CFO provides assurance as to the accuracy of customer metrics data. 1 2 3 (c) In my firm, the internal auditors provide assurance as to the accuracy of customer metrics data. 1 2 3 (d) In my firm, the external auditors provide assurance as to the accuracy of customer metrics data. 1 2 3

4

5

4

5

4

5

4

5

62

3 Customer Analytics for Internal Decision-Making and Control

Q14. We would like to know your opinion about the following aspects of your firm. Please indicate the extent to which you agree or disagree. (1 ¼ strongly disagree; 5 ¼ strongly agree). (a) There is a give-and-take relationship between the P&C and marketing functions. Each challenges the other in their meeting discussions and tries to understand the other’s point of view. 1 2 3 4 5 (b) The Marketing and P&C functions are always involved from the very early phases of discussion. 1 2 3 4 5 (c) Conflicts between the P&C and marketing function are resolved in a 1 2 3 4 timely manner

5

Q15. Please indicate the extent to which information between marketing and P&C is shared with regard to the following issues. (1 ¼ no information sharing; 5 ¼ complete information sharing)

(a) The current level of marketing’s degree of information sharing 1 2 3 with P&C regarding customer profitability. (b) The current level of marketing’s degree of information sharing with P&C regarding feedback from customers about 1 2 3 product/services quality. (c) The current level of marketing’s degree of information sharing with P&C regarding competitors’ moves. 1 2 3

4

5

4

5

4

5

Customer Orientation and Profitability Q16. We would like to know about some aspects that characterize your firm and your customers. To what extent do you agree with the following statements? (1 ¼ strongly disagree; 5 ¼ strongly agree) (a) This firm has systems in place that record each customer’s transactions. 1 2 (b) This firm can identify all transactions pertaining to each individual customer. 1 2 (c) This firm analyzes previous consumer transactions at the individual customer level to predict future transactions from that customer. 1 2 (d) In this firm, all customer interfaces possess transaction information on individual customers at all times. 1 2 (e) This firm has an excellent idea of what each individual customer has been contributing to its profits, i.e., this firm selectively acquires customers whose discounted future value exceeds acquisition costs 1 2 (f) This firm predicts what each individual customer will contribute to its profits in the future, i.e., this firm determines the revenue and the payback period to recover the cost of acquisition during the lifetime of the customer 1 2 (g) This firm chooses the right advertising channel, such as TV or the Internet, balancing future margins and the cost of acquisition for the single customer. 1 2

3

4

5

3

4

5

3

4

5

3

4

5

3

4

5

3

4

5

3

4

5

Appendix Chapter 3: Questionnaire

63

Q17. To what extent does your firm use the following customer characteristics for your segmentation strategy? (1 ¼ no extent; 5 ¼ great extent) Exchange Characteristics (Customer behavior) 1 2 3 (a) Purchase propensity (b) Contribution margin 1 2 3 (c) Cross-buying behavior 1 2 3 1 2 3 (d) Purchase frequency 1 2 3 (e) Last purchase made (f) Type of media used, i.e., advertising channel, such as TV, Web(CPM, CPC, CPA) 1 2 3 (g) Billing method 1 2 3 (h) Tenure (i.e., duration of usage) 1 2 3 Customer Heterogeneity (Sociodemographic characteristics) (i) Age 1 (j) Income 1 (k) Gender 1 (l) Place of residence 1 1 (m) Type of residence

2 2 2 2 2

3 3 3 3 3

4 4 4 4 4

5 5 5 5 5

4 4 4

5 5 5

4 4 4 4 4

5 5 5 5 5

Q18. Please indicate to what extent you agree with the following statements. (1 ¼ no extent; 5 ¼ great extent). (a) Customer demand for new products is highly predictable (b) Occurrence of major technological change is highly predictable. (c) Competitor product design changes are highly predictable. (d) Our competitors are relatively weak. (e) Customers who are identified by this firm as potentially profitable prove to be profitable in the long run. (f) A larger proportion of acquired customers remain profitable in the long term for this firm compared to its competitors. (g) The number of customers who were unprofitable last year and became profitable this year for this firm is greater than the number of customers who were profitable last year but became unprofitable this year.

1

2

3

4

5

1

2

3

4

5

1 1

2 2

3 3

4 4

5 5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

Q19. Please rate the following statements. (1 ¼ much lower; 5 ¼ much higher) (a) Relative to this firm’s main competitors, currently this firm’s profits are (b) Relative to last year, this firm’s profits are …

1 1

2 2

3 3

4 4

5 5

64

3 Customer Analytics for Internal Decision-Making and Control

References Ambler, T., & Roberts, J. H. (2008). Assessing marketing performance: Don’t settle for a silver metric. Journal of Marketing Management, 24(7/8), 733–750. Anand, B. N., Rukstad, M. G., & Paige, C. H. (2001). Capital one financial corp. Harvard Business School – Case Study. Banker, R. D., Potter, G., & Srinivasan, D. (2000). An empirical investigation of an incentive plan that includes nonfinancial performance measures. The Accounting Review, 75(1), 65. Barker, T. (2008). Exploring the differences between accountants and marketers in terms of information sharing. Marketing Intelligence & Planning, 26(3), 316–319. Behn, B. K., & Riley, R. A. (1999). Using nonfinancial information to predict financial performance: The case of the U.S. airline industry. Journal of Accounting, Auditing & Finance, 14(1), 29–56. Bonacchi, M., & Perego, P. (2012). Improving profitability with customer-centric strategies: The case of a mobile content provider. Strategic Change, 20(7–8), 253–267. Bonacchi, M., Kolev, K., & Lev, B. (2015). Customer franchise—A hidden, yet crucial, asset. Contemporary Accounting Research, 32(3), 1024–1049. Brickley, J., Smith, C., & Zimmerman, J. (1995). The economics of organizational architecture. Journal of Applied Corporate Finance, 8(2), 19–31. Carmines, E., & Zeller, R. (1979). Reliability and validity assessment. Beverly Hills, CA: Sage Publications. Casas-Arce, P., Martínez-Jerez, F. A., & Narayanan, V. G. (2011, April). The impact of forwardlooking metrics on employee decision making. Harvard Business School Working Paper 11–106. Casas-Arce, P., Martínez-Jerez, F. A., & Narayanan, V. G. (2017). The impact of forward-looking metrics on employee decision-making: The case of customer lifetime value. The Accounting Review, 92, 31–56. Chapman, C. S. (1997). Reflections on a contingent view of accounting. Accounting, Organizations and Society, 22(2), 189–205. Chenhall, R. H. (2003). Management control systems design within its organizational context: Findings from contingency-based research and directions for the future. Accounting, Organizations and Society, 28(2–3), 127–168. Chin, W. W. (1998). The partial least squares approach to structural equation modeling. In G. A. Marcoulides (Ed.), Modern methods for business research (pp. 295–335). Mahway, NJ: Lawrence Erlbaum Associates. Chin, W. W., & Newsted, P. R. (1999). Structural equation modeling analysis with small samples using partial least squares. In R. Hoyle (Ed.), Statistical strategies for small sample research (pp. 307–341). Thousand Oaks: Sage Publications. Davila, A., & Foster, G. (2005). Management accounting systems adoption decisions: Evidence and performance implications from early-stage/startup companies. Accounting Review, 80, 1039–1068. Deshpandé, R., & Farley, J. U. (1998). Measuring market orientation: Generalization and synthesis. Journal of Market-Focused Management, 2(3), 213–232. Dikolli, S. S., & Sedatole, K. L. (2007). Improvements in the information content of nonfinancial forward-looking performance measures: A taxonomy and empirical application. Journal of Management Accounting Research, 19, 71–104. Dillman, D. A. (2011). Mail and internet surveys: The tailored design method—2007 update with new internet, visual, and mixed-mode guide. Hoboken, NJ: John Wiley & Sons. Dubosson-Torbay, M., Pigneur, Y., & Usunier, J.-C.. (2004). Business models for music distribution after the P2p revolution. Paper presented at the 4th International Conference on WEB Delivering of Music, Barcelona, Spain.

References

65

El-Tawy, N., & Tollington, T. (2008). The recognition and measurement of brand assets: An exploration of the accounting/marketing interface. Journal of Marketing Management, 24, 711–731. Fader, P., & Hardie, B. (2009). Probability models for customer-base analysis. Journal of Interactive Marketing, 23(1), 61–69. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. Foster, G., & Gupta, M. (1994). Marketing, cost management and management accounting. Journal of Management Accounting Research, 6, 43–77. Germann, F., Lilien, G. L., & Rangaswamy, A. (2013). Performance implications of deploying marketing analytics. International Journal of Research in Marketing, 30(2), 114–128. Gleaves, R., Burton, J., Kitshoff, J., Bates, K., & Whittington, M. (2008). Accounting is from Mars, marketing is from Venus: Establishing common ground for the concept of customer profitability. Journal of Marketing Management, 24(7–8), 825–845. Guilding, C., & McManus, L. (2002). The incidence, perceived merit and antecedents of customer accounting: An exploratory note. Accounting, Organizations and Society, 27(1–2), 45–59. Gulati, R. (2009). Reorganize for resilience: Putting customers at the center of your business. Boston, MA: Harvard Business Press. Gupta, S., & Zeithaml, V. (2006). Customer metrics and their impact on financial performance. Marketing Science, 25(6), 718. Haenlein, M., Kaplan, A. M., & Beeser, A. J. (2007). A model to determine customer lifetime value in a retail banking context. European Management Journal, 25(3), 221–234. Haenlin, M., & Kaplan, A. M. (2004). A beginner’s guide to partial least squares analysis. Understanding Statistics, 3(4), 283–297. Holm, M., Kumar, V., & Plenborg, T. (2016). An investigation of customer accounting systems as a source of sustainable competitive advantage. Advances in Accounting, 32, 18–30. Hulland, J. (1999). Use of partial least squares (PLS) in strategic management research: A review of four recent studies. Strategic Management Journal, 20, 195–204. Inglis, R. M. (2008). Exploring accounting and market orientation: An interfunctional case study. Journal of Marketing Management, 24, 687–710. Ittner, C. D., & Larcker, D. F. (1998). Are nonfinancial measures leading indicators of financial performance? An analysis of customer satisfaction. Journal of Accounting Research, 36 (Supplement), 1–35. Jayachandran, S., Sharma, S., Kaufman, P., & Raman, P. (2005). The role of relational information processes and technology use in customer relationship management. Journal of Marketing, 69(4), 177–192. Kober, R., Ng, J., & Paul, B. J. (2007). The interrelationship between management control mechanisms and strategy. Management Accounting Research, 18(4), 425–452. Kumar, V. (2008). Customer lifetime value – The path to profitability. Foundations and Trends in Marketing, 2(1), 1–96. Kumar, V., & George, M. (2007). Measuring and maximizing customer equity: A critical analysis. Journal of the Academy of Marketing Science, 35(2), 157–171. Kumar, V., & Rajan, B. (2009a, September). Nurturing the right customers: By measuring and improving customer lifetime value, you’ll be able to grow your most profitable customers. Strategic Finance, 27–33. Kumar, V., & Rajan, B. (2009b). Profitable customer management: Measuring and maximizing customer lifetime value. Management Accounting Quarterly, 10(3), 1–18. Kumar, V., & Shah, D. (2009). Expanding the role of marketing: From customer equity to market capitalization. Journal of Marketing, 73(6), 119. Kumar, V., Lemon, K. N., & Parasuraman, A. (2006). Managing customers for value: An overview and research agenda. Journal of Service Research, 9(2), 87–94. Kumar, V., Venkatesan, R., & Reinartz, W. (2008). Performance implications of adopting a customer-focused sales campaign. Journal of Marketing, 72(5), 50–68.

66

3 Customer Analytics for Internal Decision-Making and Control

Lee, J.-Y., Sridhar, S., Henderson, C. M., & Palmatier, R. W. (2015). Effect of customer-centric structure on long-term financial performance. Marketing Science, 34(2), 250–268. Loveman, G. (2003). Diamonds in the data mine. Harvard Business Review, 81(5), 109–113. Lyons, K., Messinger, P. R., Playford, C., Niu, R., & Stroulia, E. (2009). Business models in emerging online services. Paper presented at Fifteenth Americas Conference on Information Systems, San Francisco, CA. McManus, L. (2013). Customer accounting and marketing performance measures in the hotel industry: Evidence from Australia. International Journal of Hospitality Management, 33, 140–152. McManus, L., & Guilding, C. (2008). Exploring the potential of customer accounting: A synthesis of the accounting and marketing literatures. Journal of Marketing Management, 24, 771–795. Petersen, J. A., McAlister, L., Reibstein, D. J., Winer, R. S., Kumar, V., & Atkinson, G. (2009). Choosing the right metrics to maximize profitability and shareholder value. Journal of Retailing, 85(1), 95–111. Pfeifer, P. E., Haskins, M. E., & Conroy, R. M. (2005). Customer lifetime value, customer profitability, and the treatment of acquisition spending. Journal of Managerial Issues, 17(1), 11–25. Phillips, P., & Halliday, S. V. (2008). Marketing/accounting synergy: A discussion of its potential and evidence in e-business planning. Journal of Marketing Management, 24, 751–770. Ramani, G., & Kumar, V. (2008). Interaction orientation and firm performance. Journal of Marketing, 72(1), 27. Reinartz, W. J., & Kumar, V. (2003). The impact of customer relationship characteristics on profitable lifetime duration. Journal of Marketing, 67(1), 77–99. Rierson, M., & Lattin, J. (2007). Capital one: Leveraging information-based marketing. Stanford Graduate School of Business (Case M-316). Roslender, R., & Wilson, R. M. S. (2008). Editorial: The marketing/accounting interface. Journal of Marketing Management, 661–668. Rust, R. T., Zeithaml, V. A., & Lemon, K. N. (2000). Driving customer equity: How customer lifetime value is reshaping corporate strategy. New York: Free Press. Ryals, L. (2005). Making customer relationship management work: The measurement and profitable management of customer relationships. Journal of Marketing, 69(4), 252–261. Shah, D., Rust, R. T., Parasuraman, A., Staelin, R., & Day, G. S. (2006). The path to customer centricity. Journal of Service Research, 9(2), 113–124. Smith, R. E., & Wright, W. F. (2004). Determinants of customer loyalty and financial performance. Journal of Management Accounting Research, 16, 183. Song, M., & Thieme, R. J. (2006). A cross-national investigation of the R&D-marketing interface in the product innovation process. Industrial Marketing Management, 35(3), 308–322. Verhoef, P. C., & Lemon, K. N. (2013). Successful customer value management: Key lessons and emerging trends. European Management Journal, 31(1), 1–15. Villanueva, J., & Hanssens, D. (2007). Customer equity: Measurement management and research opportunities. Foundations and Trends in Marketing, 1(1), 1–95. Vogel, V., Evanschitzky, H., & Ramaseshan, B. (2008). Customer equity drivers and future sales. Journal of Marketing, 72(6), 98–108. Widener, S. K., Shackell, M. B., & Demers, E. A. (2008). The Juxtaposition of social surveillance controls with traditional organizational design components. Contemporary Accounting Research, 25(2), 605–638. Wiesel, T., Skiera, B., & Villanueva, J. (2008). Customer equity: An integral part of financial reporting. Journal of Marketing, 72(2), 1–14. Yin, R. K. (2003). Case study research: Design and methods. Thousand Oaks, CA: Sage Publications.

Chapter 4

Customer Equity for External Reporting and Valuation

Not everything that can be counted counts, and not everything that counts can be counted. —Albert Einstein

4.1

Customers as the Most Valuable (Intangible) Asset

Recurring concerns have been expressed by academics, managers, stakeholders and regulators about the increasing irrelevance of the current financial accounting information. From the latter part of the twentieth century to the outset of the twenty-first century, the global economy has moved from an industrial dependence to knowledge- and service-based. Consequently, the demand for informational products and services has replaced the need for many physical products. New business models have thus emerged with significant implications on how organizations create and preserve value (Govindarajan et al. 2018). However, accounting-based financial information is produced and disseminated using corporate reports which have not evolved much over the years in either form or content. Figure 4.1 shows the main drivers of change to financial disclosure in order to make informed decisions and clearly shows that both stakeholder intensity and business model complexities have evolved in the last decades. Firms have moved from single owner scenarios to multi-stakeholder settings, and the competitive advantage business model is now largely determined by intangible assets that are difficult for competitors to imitate (think for example of the Gucci brand, Google’s search engine, Netflix’s customer base). We argue that stakeholders need clear, consistent and comparable information on these strategic assets of companies, and the efficiency of their use. In the 2016 book The End of Accounting, NYU Stern Professor Baruch Lev claimed that over the last 100 years or so, financial reports have become less useful for making capital market decisions (Lev and Gu 2016). Our current financial accounting models cannot capture the principle value creator for customer-centric firms—the intangible assets. Customer Franchise is among one of the most important © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2019 M. Bonacchi, P. Perego, Customer Accounting, SpringerBriefs in Accounting, https://doi.org/10.1007/978-3-030-01971-6_4

67

68

4 Customer Equity for External Reporting and Valuation

Fig. 4.1 What drives accounting information

intangible assets for understanding a firm’s value. Despite its importance, information about customers is often not available in IFRS/US GAAP financial statements.

4.2

Customer Franchise Is Missing in IFRS/US GAAP Financial Statements: How to Value It?

We argue that customer information helps investors understand the fundamental business model of a company and is precisely what they need to make better decisions. In fact, it is not sufficient to provide non-financial indicators such as order backlogs, customer satisfaction, wireless phone metrics, web usage data, brand values, biotech companies’ product pipeline content or firms’ patent attributes. What is needed is to link the non-financial information to the crucial mechanisms by which the examined non-financial variables affect a firm’s earnings and, ultimately, its stock price, and then to summarize them into a direct measure of value. As well explained in Bini et al. (2016) the business model should drive non-financial disclosure and these metrics need to illuminate the value creation process of the firm. We approach the issue by characterizing the business model of subscriptionbased enterprises (SBEs), companies that offer a for-fee-per-period access to products or services. Specifically, we build on the idea that the acquisition and retention of profitable customers is crucial for SBEs to identify the fundamental elements of their business model (e.g. customer base, revenues, service cost per user, and customer turnover), and summarize them into a comprehensive metric that quantifies

4.3 Describing SBEs Business Model Using Customer Metrics

69

its value. We then explore the usefulness of the metric in explaining stock price. We believe that the proposed method identifies important value drivers and provides an easy-to-implement algorithm to summarize these drivers into a measure of considerable usefulness to investors.1 We apply the “SBE” label to companies that structure their operations so that a customer pays a fee for the right to access products or services for a period of time. While pioneered by magazine and newspaper publishers, this business model is spreading to a growing number of industries including, among others, telecom and software. While the approach we propose is applicable to a wide range of business models, we focus on SBEs for several reasons. First, the role of the SBE business model in the modern economy is fast expanding. A recent Economist article titled: Subscribers are the new, new thing in business (Economist 2018). Second, as we mentioned in Chap. 2, the arrival and departure of customers is clearly observable and companies are likely to have systems in place for keeping track of these data. Third, although the customer base is an important asset for such companies, similar to critical asset values in other sectors, it is not recognized on the balance sheet under current IFRS/U.S. GAAP. To fill this information gap, an increasing number of companies are voluntarily disclosing customer base data in financial statements, press releases and conference calls. In spite of the growing interest in the metrics, however, there is little uniformity or consistency in these disclosures, making analysis and valuation challenging. Our study, therefore, informs SBEs of disclosure beneficial to financial statement users and provides an algorithm for summarizing these data into a measure of value.2

4.3

Describing SBEs Business Model Using Customer Metrics

An attractive feature of SBEs is that the acquisition and departure of customers is clearly observable, allowing companies to track the composition and profitability of their customer base.3 Companies employing subscription-based models benefit from 1 An examination of several analysts’ reports for our sample companies revealed that analysts discuss and forecast certain customer metrics. However, we did not encounter a systematic discussion of algorithms used to transform customer metrics into a measure of customer equity. 2 This is consistent with the FASB and IASB guidance for “Management Discussion and Analysis” (“MD&A”) and “Management Commentary” (IASB 2009). In fact, both documents require information that supplements and complements the firm’s financial statements. Specifically, the disclosure required in “Management Commentary” should be future-oriented, understandable, relevant, reliable, and comparable, and should provide an “analysis through the eyes of management” (SEC, Regulation S–K, Item 303). Examples of such information include details about the nature of the business, key resources, risks and relationships, and performance measures and indicators. 3 The subsequent analysis could be extended to non-contractual settings, e.g. restaurants, retailers, or airlines. This, however, requires modeling the probability of repeat purchases which would unnecessarily complicate the analysis (McCarthy and Fader 2018).

70

4 Customer Equity for External Reporting and Valuation

acquiring customers at the lowest possible cost, increasing the monthly average margin per user, and retaining existing customers. Following this rationale, the economics of subscription-based models are driven by four key factors: (1) cost per acquisition, (2) cost of service, (3) average revenue per user (ARPU), and (4) churn. Consistent with this notion, SBEs can identify their business model, albeit very few do, by disclosing the following customer metrics: • Number of subscribers: Number of active customers at the end of the period. • Gross customer additions: Number of new customers that joined the company within the fiscal period. • Net customer additions: Gross number of new customers acquired during the period, less the number of customers deactivated. • Churn rate: Rate of customer attrition, measured as cancellations per user per period. Churn rates are generally presented on a monthly basis. • ARPU: Average monthly service revenue per subscriber. • Cost of service: Average monthly cost per subscriber of providing services and support to existing customers. • Cost per gross addition (CPGA): Average cost incurred to acquire new customers. This measure is used to evaluate how effective marketing programs are in acquiring new subscribers. CPGA is also commonly referred to as subscriber acquisition costs (SAC). We provide an example of sufficient disclosure for an application of the customer-equity model in Table 4.1. Note that Leap Wireless International’s (Leap) disclosure clearly outlines the company’s business model, starting with the cost to acquire a customer (CPGA), the impact of customer acquisition on subscriber Table 4.1 Example of disclosure The following is an excerpt from Leap Wireless International’s 10-Q for the period ending June 30, 2008 (filing date: 8/7/2008) as an example of disclosure of the necessary customer metrics to calculate customer equity Change Amount Percent (%) For the 3 months ended June 30: 2008 2007 Gross customer additions 542,005 462,434 79,571 17.2 Net customer additions 171,171 126,791 44,380 35.0 Weighted-average number of customers 3,162,028 2,586,900 575,128 22.2 As of June 30: Total customers 3,305,251 2,674,963 630,288 23.6 The following table shows metric information for the 3 months ended June 30, 2008 and 2007: Three months ended June 30: 2008 2007 ARPU $ 43.97 $ 44.75 CPGA $ 205 $ 182 CCU $ 21.01 $ 19.87 Churn 3.8% 4.3%

4.4 Valuing SBEs Using Publicly Disclosed Customer Metrics: A. . .

71

growth (gross customer additions), the effect of churn rate on the gross addition and the “leftover” customers (net customer additions). Once the customers are acquired, they provide a certain amount of revenue per month (ARPU) and Leap has to incur some cost to provide the service (CCU). This depiction of Leap’s business model provides a clear picture of the company’s success. Leap acquires a customer with $205 investments. This average customer pays a subscription of $44 and Leap incurs $21, i.e. with a margin of $23. Finally, the average customer has a lifetime of 32 months, calculated as the reciprocal of churn rate (1/3.1%). In other words, with $205 investments in customer acquisition Leap is able to get $736 as the sum of a stream of future cash flow (margin of $23). The information Leap provided, not only describes its business model, but also allows for customer equity calculation, i.e. the major intangible assets of the companies.

4.4

Valuing SBEs Using Publicly Disclosed Customer Metrics: A Parsimonious Model to Estimate Customer Equity

The fundamentals for valuing Customer Equity (CE) have been developed in the customer lifetime value (CLV) literature, which we extend to the accounting field.4 Extant research proposes several methods for estimating CE, which, while analytically elegant, are generally complex and call for numerous inputs. This, in turn, has constrained the empirical examination of CE to very small samples, often individual companies, in very specific settings (Gupta et al. 2004; Schulze et al. 2012). Building on prior work, we refer to two concepts that can be used when evaluating the expected profitability of a firm’s customer base (Villanueva and Hanssens 2007): (a) Current Customer Equity (CEcur): The sum of the future profit margins generated from the customers that have already been acquired by the end of the period (Villanueva and Hanssens 2007: 5). (b) Total Customer Equity (CEtot): The sum of the future profit margins generated from current (CEcur) and future (CEfut) customers of the firm (Hogan et al. 2002a,b; Kumar and Shah 2009).5

4

CLV is the disaggregated measure and CE is the aggregated measure of customer profitability. In essence, calculating CLV involves predicting the future profit margin from each customer and then discounting it by the cost of capital. The sum of the lifetime values of all customers is the CE. 5 CEcur can be considered a special case of CE, where the acquisition of future customers is a zero NPV project. For the remainder of the paper, unless specified otherwise, we use “CEcur” and “customer equity” interchangeably.

72

4 Customer Equity for External Reporting and Valuation

In the marketing literature it is common to estimate the lifetime value of actual and future customers by tracking the evolution of each group of customers acquired during a particular period, typically referred to as “customer cohort” (e.g. Gupta et al. 2004). The general algorithm is as follows: the firm acquires n0 customers at time t0 at an acquisition cost of c0 per customer. Subsequently and over time, customers defect at a fixed defection rate, (1–r), such that the firm is left with n0r customers at the end of period 1, n0r2 customers at the end of period 2, and so on (Fig. 4.2). The value of the firm’s customer base is then estimated as the sum of the discounted customer lifetime values of all cohorts (Berger and Nasr 1998; Gupta and Lehmann 2005; Gupta et al. 2004). The customer equity value therefore could be expressed as: CEtot ¼

1 X k¼0

nk

1 X

mt

t¼k

!

r tk ð1 þ iÞ

tk

 ck

1 ð1 þ i Þk

ð4:1Þ

where t is the unit of time in the analysis; k is the cohort; n is the number of customers; m is the profit margin; r is the retention rate; c is the acquisition cost; and i is the cost of capital. We develop a parsimonious model that aggregates publicly available data into estimates of CEcur and CEtot. Figure 4.3 presents the evolution of the general model: total customer equity (CEtot) can be presented as the sum of the lifetime value of a firm’s current and future customers: CEtot ¼ CE cur þ CE fut

ð4:2Þ

Our implementation of Eq. (4.2) requires forecasting current and future customer equity, using publicly available data. Specifically, to estimate the value of current customers, we need to consider the number of customers at the end of the period, the margin per customer, and the customer retention, accounting for the fact that, over time, customers leave the company. Since the model spans multiple periods, we also require a discount rate. The value of the current customer base (CEcur) can be expressed as: CE cur ¼ n

1 X t¼1

rt m ð1 þ iÞt

ð4:3Þ

where: n ¼ number of customers at the end of the period (customer base). m ¼ margin per customer. r ¼ retention rate (1–churn). i ¼ cost of capital. Since Eq. (4.3) is an infinite geometric series, under the assumption that the margin and retention rates are constant over time, for |r/(1 + i) | < 1 the model could be written as:

Fig. 4.2 Theoretical model for valuing customer base (source: Bonacchi et al. 2015)

4.4 Valuing SBEs Using Publicly Disclosed Customer Metrics: A. . . 73

Fig. 4.3 Parsimonious model to estimate CEcur and CEtot using publicly available data (source: Bonacchi et al. 2015)

74 4 Customer Equity for External Reporting and Valuation

4.4 Valuing SBEs Using Publicly Disclosed Customer Metrics: A. . .

75

  r CE cur ¼ n m ð1 þ i  r Þ

ð4:4Þ

In other words, our measure of CEcur is a margin (m) times a margin multiple that depends on the probability of customer retention and the company’s cost of capital.6 Value of future customers. A challenge with estimating CEfut is forecasting the number of future customers. We use a two-stage approach to model customer acquisition. Step 1—Growth Period: For a finite time horizon, we forecast gross customer additions as the arithmetic mean of the gross addition for the last two quarters. We estimate the length of the growth period as 1/churn (i.e. number of months an average customer stays with the company). We take this approach under the conjecture that a company relying on a subscription-based model will not grow absent innovations in product/services. Thus, we treat churn as a proxy for the degree of innovation, which is the main driver of growth (Thompson 2011).7 Step 2—Steady State: After the Growth Period, we model gross additions by setting them as equal to the number of customers that churn over an infinite time horizon. In other words, absent innovation, the company maintains its customer base, acquiring customers to compensate for the churn. Following the above rationale, future customer equity is the sum of customer equities from the growth and the steady state periods:

r ð1þir Þ

CE fut ¼ growth CE fut þ steady CE fut

ð4:5Þ

where growth_CEfut (steady_CEfut) is the number of new customers acquired in the growth (steady-state) period multiplied by the discounted margin multiple: growth CE fut ¼ growth n*gross ða  cÞaØ

ð4:6Þ

1 steady CE fut ¼ steady n*gross ða  cÞ ið1 þ iÞh

ð4:7Þ

hi

where: growth_n*gross ¼ gross addition during the growth period. steady_n*gross ¼ gross addition during the steady-state period

6

In our model we estimate CEcur over an infinite time horizon. This is desirable since we do not need to specify arbitrarily the number of months that a customer will stay with the company. In addition, the retention rate and discount rate account for future uncertainties by discounting accordingly future profit margins (Gupta et al. 2004). 7 Most of the companies employing a subscription-based business model grow under the recurring revenue model: “Sell it once and collect until it churns.”

76

4 Customer Equity for External Reporting and Valuation

a¼

1 X t¼1

rt r m t ¼ m ð 1 þ i  rÞ ð1 þ i Þ

h

Þ aØ ¼ 1ð1þi (the value of an annuity for h months). i hi

r ¼ (1-churn). h ¼ number of months a customer stays with the firm; [1 / (1 – r)] rounded to an integer. c ¼ cost of customer acquisition. We estimate the gross additions in the growth and steady-state periods as follows: Growth Period: growth n*gross ¼



  ngrosst þ ngrosst1 =2 =3

where the average gross additions for the two prior periods are scaled by three to convert the data from quarterly to monthly. If a company does not disclose gross additions, we model it as the sum of the customers that churn and the net additions for the period: growth n* gross ¼ ððnt þ nt1 Þ=2Þchurn þ ðnt  nt1 Þ=3 Steady State steady n*gross ¼

nt r h þ growth n*gross

h X

! r t churn

t¼1

where: ntrh¼ number of current customers that remain after h months. h X growth n*gross r t¼ number of customers acquired during the growth period that t¼1

remain after h months. Combining CEcur and CEfut, we can express CEtot as: CEtot ¼ CE cur þ growth CE fut þ steady CE fut

ð4:8Þ

Using Eqs. (4.5, 4.6 and 4.7), abstracting from the effect of taxes, and recognizing that, for r 6¼ 1, k X

    r t ¼ r  r ðhþ1Þ =ð1  r Þ ¼ ½r=ð1  r Þ 1  r k ,

t¼1

we can re-write (4.8) as:

4.5 Customer Equity and Stock Returns: Empirical Evidence

CE tot, t ¼ nt a þ growth n*gross ða  cÞaØ þ steady n*gross ða  cÞ hi

77

1 i ð1 þ i Þh

ð4:9Þ

The graphical representation of the model in depicted in Fig. 4.2.

4.5

Customer Equity and Stock Returns: Empirical Evidence

To estimate the value of a firm’s customer base, several inputs are required: number of customers, margin per customer, customer retention rate, and the cost of capital for the firm. The number of customers refers to the active customer base at the end of the fiscal quarter. Margin per customer is measured as the difference between average revenue per customer, ARPU, and cost of service. Similar to the number of customers, most companies that disclose customer-related metrics provide sufficient data to infer ARPU.8 Some companies, however, do not disclose cost of service per customer. In these cases, we estimate the metric by applying to ARPU the ratio of “cost of service” to “service revenue” from the income statement. When companies provide the disclosure by segment (e.g. Post-Paid and Prepaid or U.S. and Latin America), we use the weighted average of the customer metrics. Turning to the customer retention rate, its estimation plays a critical role in the model, as it reflects the likelihood that a customer will defect in the future. Analyses of parametric and non-parametric models to calculate customer lifetime (i.e. how long a customer is expected to stay with the firm and create value) are beyond the scope of this study, so we project the historical churn to the future.9 In practical terms, we derive the probability of the current customer remaining active at time t as (1–churn). For the CEtot model, we also require the cost to acquire a new customer. Customer-acquisition cost, when reported, is expressed as cost per gross addition. However, more than 30 percent of the companies in our sample do not report these data. In these cases, we operationalize the acquisition cost by dividing marketing cost by the gross customer addition. We do this only when it is clear that the cost of customer acquisition is included in the marketing cost. We acknowledge that some of the marketing cost is incurred for retention purpose, but we do not have sufficient information to refine the measure. Several studies, however, show that, in general, customer acquisition costs are significantly higher than customer retention cost (Reichheld and Teal 1996; Thomas 2001; Gupta et al. 2004). The last model input is cost of capital. In theory, cost of capital is a time- and firmspecific measure. In practice, however, there is little consensus on how to measure 8 When a company does not disclose ARPU, we calculate it by dividing subscriber revenues by the weighted average number of customers for the period. 9 Fader and Hardie (2007) and Rosset et al. (2003) provide examples of projecting retention rate.

78

4 Customer Equity for External Reporting and Valuation

cost of capital. For this study, we use a constant discount rate of 12 percent (Frankel and Lee 1998; Gupta et al. 2004). As a robustness test, we repeat the analysis using a time-varying discount rate.10 We focused the main analysis on the future value of current, retained, customers (CEcur). The resulting customer equity model is a simplified version of Eq. (4.1) (Gupta et al. 2004). Under the assumptions that the profit margin and customer churn are constant and the acquisition of future customers is a zero net present value (NPV) project (Gupta and Lehmann 2005), customer equity could be expressed as: CE cur ¼ n

1 X t¼1

  rt r m ¼n m ð1 þ i  r Þ ð1 þ i Þt

ð4:10Þ

where n is the number of active customers at the end of the period (historic customer base); m is the profit margin per customer (revenue minus service cost) for period t; r is the retention rate for period t; i is the cost of capital; and t is the time period.11 Our main analysis focuses on CEcur and not on total customer equity (CEtot) for several reasons. Forecasting the number of future customer acquisitions and their outcomes requires a higher degree of subjectivity, coming from three separate sources: (1) Customer growth: We forecast customer growth using an ad hoc model relying on historical growth. A diffusion model is a natural candidate for estimation of the growth of the customers (Kim et al. 1995; Gupta et al. 2004). Such an approach requires the solution of nonlinear differential equations, and the resulting model is too complex to operationalize for a large sample (Pfeifer 2011). (2) Acquisition cost: The non-random loss of observations is likely to bias the reported results. As discussed previously, more than one-third of companies do not report these data, requiring the use of total marketing costs as a crude proxy, which could be an additional source of bias. (3) Discount rate: Theoretically, the discount rate for future cashflows should be higher than the discount rate used for the current customers’ cashflows. The discount rate is supposed to capture the risk inherent in the customer type: A current customer is more likely to stay with a company through good times and bad. Furthermore, whether or not a company can acquire new customers is strongly impacted by macro and micro economic factors.

10

We derive the time-varying discount rate as 10% + one-year LIBOR. Using this rate instead of the static 12% does not qualitatively affect the results. 11 The constant profit and retention rate assumptions, while not too strong, allow for the generation of a parsimonious model that is easily implementable in practice. Separately, we do not introduce taxes in the model: While the extension is analytically straightforward, the practical implementation presents challenges without contributing to the insights.

4.6 Beyond GAAP: Customer Metrics Reporting

79

In summary, by focusing on the current customers of a company, we obtain a parsimonious and easy-to-implement model of customer equity.12 Despite the fact that our estimate of customer equity does not likely capture the entire customer intangible asset, we believe it is a useful practical valuation tool which provides a summary performance metric which managers and investors can track over time.13 To examine whether CEcur could be used in predicting future market performance, we examine the future returns of companies grouped by the “comprehensive to market value of equity” ratio (Gu and Lev 2011). In particular, we define comprehensive value as the sum of reported BVE and our estimate of CEcur, with the interpretation that comprehensive value (CV) that is higher (lower) than market value of equity (MVE) indicates underpriced (overpriced) stock. In particular, we regress future returns (buy-and-hold market-adjusted returns) on a set of controls and our variable of interest is CV/MVE, which takes a value of 0, 0.5, and 1 when a firm is in the lowest, middle, and highest tercile of the ratio, respectively, for the fiscal quarter.14 It is notable that the coefficient on CVq/MVE10Q,q is significantly positive and economically significant. As an example, the hedge return from buying and holding (selling short) stocks in the top (bottom) tercile of the comprehensive to market value of equity—undervalued (overvalued) firms—yields 36.2, 41.2, and 76.5 percent return for 1, 2, and 3 years after the investment, respectively.

4.6

Beyond GAAP: Customer Metrics Reporting

“Subscribers are the New, New Thing in Business” declared The Economist (April 11, 2018). The magazine says that “Subscription models are seen by many investors and executives as the holy grail, because they promise a recurring stream of revenue. . .. The attractions of subscription businesses are obvious. Firms can predict the future better and build deeper relationships with customers who have less

12

One of this book’s authors applied the model outlined in the previous section in a paper co-authored with Baruch Lev and Kalin Kolev. In this section, we explain how to apply the model previously outlined. For further details, interested readers should refer to Bonacchi et al. (2015). 13 Empirically, Silveira et al. (2012) document that CEcur is a sufficiently close approximation of CEtot. 14 The general model takes the following form:   X RetqþN ¼ γ0 þ γ1 CVq =MVE10Q, q þ γ2 Beta þ γ3 BM þ γ4 logðMVE10Q Þ þ γ5 Accruals N¼1 þ γ6 Momentum þ ϑ where Retq + N is the buy-and-hold market-adjusted return cumulated for 360, 720, and 1080 calendar days, starting 2 days after the 10-Q filing date for the current quarter. The set of controls aims to correct for previously documented determinants of stock returns. A discussion of the general model is available in Doyle et al. (2003).

80

4 Customer Equity for External Reporting and Valuation

incentives to shop around.” Telecom companies, Internet service providers, media and entertainment firms, as well as insurance companies are the traditional, subscription-based enterprises, but the subscription model is fast expanding. Many software producers offer subscription services to customers, and even Procter and Gamble sells detergents to subscribers. Gillette markets razors on the basis of monthly fees, and Rolls-Royce, General Electric and Pratt & Whitney offer “power by the hour subscriptions.” And, as The Economist notes: “Several star firms floating their share this year have subscription models. Dropbox, a cloudstorage firm, listed on NASDAQ on March 23rd and is now worth $13bn. It boasts 500m registered users. . .. Spotify, has 159m users but derives its $27bn valuation from 71m “premium subscribers” who pay to listen without adverts.” “The subscription boom will doubtless continue” concludes The Economist. In this chapter, we argue that customer metrics are useful to understand the business model of a firm in general and of a subscription-based enterprise in particular. To this end, we introduce a model translating the business mechanism of subscription-based enterprises into a single measure of customer equity value. We apply the estimation to a sample of companies that disclose customer-related metrics, and show that the measure of customer equity is positively and significantly associated with future returns. Given the above we advocate for expanding the GAAP disclosure with a formalized and possible standardized Non-GAAP disclosure of customer metrics. This will be useful to estimate Customer Equity. What is needed as a minimum to estimate Customer Equity is listed in Table 4.2. To make the table more meaningful, we look at the customer metrics disclosed by Spotify in their prospectus and value the customer franchise of Premium subscribers as of December 2017.15 The estimated current subscriber value of $2.06 B is conservative, since it does not factor in the value of future customers (CEfut) and is limited to premium customers. But even this conservative value is by far the largest asset owned by Spotify, yet not presented on its balance sheet. Given Spotify’s market capitalization of $30 billion on June 29, its customer franchise multiple is 14.6. As Baruch Lev wrote recently: “Considering all the deficiencies of reported accounting earnings, the customer franchise multiple is a much more meaningful valuation metric than the widely used P/E ratio. Both the customer value and its multiple can be examined over time and across peer companies to assess share over/under valuation for investment purposes.”16 Effective beyond-GAAP disclosure should be standardized and directly linked to performance and most importantly should describe the business model of the 15 The data are retrieved from the FORM F-1 SEC filing available at: https://investors.spotify.com/ financials/default.aspx. In particular, we looked at the filing date of 28 February 2018: http:// d18rn0p25nwr6d.cloudfront.net/CIK-0001639920/6b71c48f-22b3-4c46-a206-5eb425d05e63.pdf 16 Interested readers should refer to Customer Franchise―The Most Valuable Asset: Here Is How to Value It, available from https://levtheendofaccountingblog.wordpress.com/2018/05/18/05-18-18-new-customer-fran chise%E2%80%95the-most-valuable-asset-here-is-how-to-value-it/

References

81

Table 4.2 Example of customer metrics at spotify (1) (2)

Variables ARPU Cost of service

¼ Margin per subscriber (1–2) (4) Churn rate (5) ¼ (1/3) Lifetime (1/3) (6) Number of customers end of the quarter (7) ¼ (3*5*6) Current customer equity (CE)

(3)

Data extracted from FORM F-1 5.24 Pag. 79 3.97 Pag. 76 761/1019 ¼ 74.68% 1.27 5.1% 19.60 months 71 million

Pag. 79 Pag. 79

1767 million euros 2058 million dollars (converted June 29, 2018)

The data are retrieved from the FORM F-1 SEC filing available at: https://investors.spotify.com/ financials/default.aspx. In particular, we looked at the filing date of 28 February 2018: http:// d18rn0p25nwr6d.cloudfront.net/CIK-0001639920/6b71c48f-22b3-4c46-a206-5eb425d05e63.pdf

company. We believe the above-mentioned examples in the subscription-based companies show that customer metrics are able to inform investors of the successful implementation of their strategies.

References Berger, P. D., & Nasr, N. I. (1998). Customer lifetime value: Marketing models and applications. Journal of Interactive Marketing, 12(1), 17–30. Bini, L., Dainelli, F., & Giunta, F. (2016). Business model disclosure in the strategic report: Entangling intellectual capital in value creation process. Journal of Intellectual Capital, 17(1), 83–102. Bonacchi, M., Kolev, K., & Lev, B. (2015). Customer franchise—A hidden, yet crucial, asset. Contemporary Accounting Research, 32(3), 1024–1049. Doyle, J. T., Lundholm, R. J., & Soliman, M. T. (2003). The predictive value of expenses excluded from pro forma earnings. Review of Accounting Studies, 8(2), 145–174. Economist. (2018). Subscribers are the new, new thing in business. The Economist, 414(9083). Fader, P., & Hardie, B. (2007). How to project customer retention. Journal of Interactive Marketing, 21(1), 76–90. Frankel, R., & Lee, C. M. C. (1998). Accounting valuation, market expectation, and cross-sectional stock returns. Journal of Accounting & Economics, 25(3), 283–319. Govindarajan, V., Rajgopal, S., & Srivastava, A. (2018, February 26). Why financial statements don’t work for digital companies. Harvard Business Review. https://hbr.org/2018/02/whyfinancial-statements-dont-work-for-digital-companies. Gu, F., & Lev, B. (2011). Intangible assets: Measurement, drivers, and usefulness. In S. Giovanni (Ed.), Managing knowledge assets and business value creation in organizations: Measures and dynamics (pp. 110–124). Hershey, PA: IGI Global. Gupta, S., & Lehmann, D. R. (2005). Managing customers as investment. The strategic value of customer in the long run. Upper Saddle River, NJ: Wharton School Publishing.

82

4 Customer Equity for External Reporting and Valuation

Gupta, S., Lehmann, D. R., & Stuart, J. A. (2004). Valuing customers. Journal of Marketing Research, 41(1), 7–18. Hogan, J. E., Lehmann, D. R., Merino, M., Srivastava, R. K., Thomas, J. S., & Verhoef, P. C. (2002a). Linking customer assets to financial performance. Journal of Service Research, 5(1), 26–38. Hogan, J. E., Lemon, K. N., & Rust, R. T. (2002b). Customer equity management: Charting new directions for the future of marketing. Journal of Service Research, 5(1), 4–12. IASB. (2009). Management commentary. London: International Accounting Standards Board (IASB). Kim, N., Mahajan, V., & Srivastava, R. K. (1995). Determining the going market value of a business in an emerging information technology industry: The case of the cellular communications industry. Technological Forecasting and Social Change, 49(3), 257–279. Kumar, V., & Shah, D. (2009). Expanding the role of marketing: From customer equity to market capitalization. Journal of Marketing, 73(6), 119. Lev, B., & Gu, F. (2016). The end of accounting and the path forward for investors and managers. Hoboken, NJ: Wiley. McCarthy, D. M., & Fader, P. S. (2018). Customer-based corporate valuation for publicly traded non-contractual firms. Journal of Marketing Research. https://doi.org/10.1509/jmr.17.0102. Pfeifer, P. E. (2011). On estimating current-customer equity using company summary data. Journal of Interactive Marketing, 25(1), 1–14. Reichheld, F. F., & Teal, T. (1996). The loyalty effect: The hidden force behind growth, profits, and lasting value. Boston, MA: Harvard Business School Press. Rosset, S., Neumann, E., Eick, U., & Vatnik, N. (2003). Customer lifetime value models for decision support. Data Mining and Knowledge Discovery, 7(3), 321–339. Schulze, C., Skiera, B., & Wiesel, T. (2012). Linking customer and financial metrics to shareholder value: The leverage effect in customer-based valuation. Journal of Marketing, 76(2), 17–32. Silveira, C. S., de Oliveira, M. O. R., & Luce, F. B. (2012). Customer equity and market value: Two methods, same results? Journal of Business Research, 65(12), 1752–1758. Thomas, J. S. (2001). A methodology for linking customer acquisition to customer retention. Journal of Marketing Research, 38(2), 262–268. Thompson, R. (2011). The Pwc valuation index: Is another tech bubble emerging? London: PwC Valuation. Villanueva, J., & Hanssens, D. (2007). Customer equity: Measurement management and research opportunities. Foundations and Trends in Marketing, 1(1), 1–95.

Chapter 5

Conclusions and Trends to Look Forward

What if you had a weather forecast for everything that happened in your life? —David Kenny (senior vice president of IBM Watson and Cloud Platform)

5.1

Looking Back and Looking Ahead

As illustrated in cases and examples in previous chapters, the explosion of customerrelated data makes the role of data management and forecast much more prominent to guide marketing initiatives and plan firm’s business activities, with huge implications for the modern CFO function. Our empirical evidence confirms that it is essential to ensure high quality of customer data. A common step required in many firms to establish high standards of data quality is to involve various key corporate functions to validate existing sources, removing duplicates by matching records and poor data, and resolving inherent conflicts in data collection or data cleansing. The responsibility to validate customer data should involve the CFO as one of the main user of this type of data for planning and control. Once the right data is generated, CFOs can exploit customer-related information to make sense of customer preferences and meet their expectations. Customer analytics like Customer Lifetime Value (CLV) and Customer Equity (CE) outlined in this monograph should enhance the opportunity to increase revenues with existing customers, but also secure new customers or increase margins. This kind of analytics is fundamental, because it allows differentiating between customers that are worth spending time and budget on from those who are not. In the next three sections, we highlight three trends that will likely shape the landscape of customer accounting in the next decade. First, the ubiquity of consumers getting connected dramatically changes the landscape of e-commerce. A major challenge is to understand and predict customer behaviour for the coming new era of business models that aim to integrate offline with online commerce. Second, while © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2019 M. Bonacchi, P. Perego, Customer Accounting, SpringerBriefs in Accounting, https://doi.org/10.1007/978-3-030-01971-6_5

83

84

5 Conclusions and Trends to Look Forward

the extant accounting literature documents the informational problems associated with financial reporting complexity, firms have at their disposal a variety of disclosure channels beyond financial statements that can be used to shape their information environment and influence investment decisions. We outline the potential role of customer metrics as key leading, forward-looking indicators in enhanced forms of corporate reporting. Finally, we close the monograph by flagging the opportunity to examine how current disruptive innovations in Artificial Intelligence, deep and machine learning will have a profound and radical impact on customer behaviour and related predictive models of marketing performance.

5.2

Linking Online with Offline Commerce

Technology advances have introduced customers to new buying behaviors, and especially the rise of e-commerce has provided retailers with novel avenues to reach those customers. Past best-practices in marketing management relied on demographics to extrapolate key information about the characteristics and behavior of a firm’s customer base. Because consumers’ digital trail can be increasingly collected and analysed in real-time, through the advent of the Internet and mobile device usage, currently firms can get much deeper than audience demographics and down into consumer psychographics. The large amount of online data available provides CMOs and CFOs a much better battery of accurate and timely indicators regarding motivations, needs, and purchase behaviour. It has never been easier to zoom in on what your individual customers are doing, and this is a major reason why enterprise products like Google Analytics and the like are getting easier to use and understand. In their effort to use CLV as a decision-making basis for marketing management, especially companies operating an online store in a non-contractual relationship with customers face the issue of selecting the appropriate CLV model and other predictive customer analytics. The challenge for academic research in the next years will remain to test predictions in various application contexts in online shopping, by exploiting the sophistication of CLV models in marketing literature (Jain and Singh 2010; Ascarza et al. 2017) with more granular customer data at the individual transaction level (refer to these recent studies as examples: Verbraken et al. 2013; Jasek et al. 2018; Mzoughia et al. 2018; Óskarsdóttir et al. 2018). At the same time, the major research issue for retailers remains how to provide convenient service through different channels, with proposed “online to offline” and “offline to online” (O2O) commerce and many kinds of creative service models. O2O Commerce can be defined as providing seamless shopping experience between online commerce and offline bricks-and-mortar with any connected device (e.g. Tsai et al. 2015). Scholars in marketing and accounting will need to address, among others, the following key research questions:

5.3 Enhanced Forms of Corporate Non–financial Reporting

85

• How to integrate new multi-channel services with traditional brick-and-mortar or e-commerce? • How to evaluate whether an O2O service is cost-effective and able to enhance the consumer satisfaction with the proposed new services? • What are the consequences of the integration of O2O services in the organizational architecture? What type of solutions in the allocation of decision rights and incentive systems enable a better fit with the challenges posed by O2O services?

5.3

Enhanced Forms of Corporate Non–financial Reporting

Investors are poorly served by backward-looking accounting methods, as the ‘End of Accounting’ recent book by Lev and Gu argues (Lev and Gu 2016). Deficiencies of reported earnings make it especially hard to communicate the pervasive existence of intangible assets within corporations. The rate of corporate investment in physical capital fell by 35% over the 1977–2012 period, whereas the rate of investment in intangible assets increased by 60% during the same period. Lev and Gu (2016) suggest to improve the current corporate reporting landscape by including an expanded set of non-financial metrics with a higher predictive ability than traditional financial rations. In this respect, customer metrics like CLV, customer equity, churn rate and customer acquisition costs are increasingly the new leading indicators to create and sustain corporate value in the twenty-first century intangible asset-based economy. As we outlined in Chap. 4, effective beyond-GAAP disclosure should be standardized and directly linked to performance and most importantly should describe the business model of the company. We believe the examples presented in the subscription-based companies show that customer metrics are able to inform investors of the successful implementation of their strategies. Among the recent international initiatives to spur enhanced forms of corporate reporting, Integrated Reporting (IR) attempts to merge in one document financial and non-financial data to overcome a potential disconnect in investors’ processing of the two types of information. A salient feature of an IR is ‘integrated thinking’, defined as “the active consideration by an organization of the relationships between its various operating and financial units and the capitals that the organization uses or affects” (IIRC 2013). IR is therefore more than just a reporting framework. It helps to better understand and link together disparate sources and drivers of longterm value by ‘telling the story’ of an underlying business model, both internally and to external financial markets (ACCA 2017). Connectivity is one of the six guiding principles that informs the content and presentation of the IR. An IR should show, as a comprehensive value creation story, the combination, interrelatedness and dependencies between the components that are material to the company’s ability to create value over time (IIRC 2013). Customer metrics like CLV and Customer Equity should enable managers to consistently reach the customers that matter the most and focus on enhancing the connectivity among capitals whether through better targeting, value added services or improvements to customer experience.

86

5 Conclusions and Trends to Look Forward

In the near future, we further predict an increased role of customer metrics because of recent developments in regulations and legal frameworks intended to fix the drawbacks in the format and usefulness of current corporate financial reporting. As a relevant example of this trend, the European Union Directive 2014/95/EU mandates the disclosure of non-financial information for approximately 6000 large companies starting from the financial year 2017 (European Commission 2017). Although customer data is not explicitly included in the new legislations, companies will face a mounting pressure to provide non-financial data related to business operations and ultimately to their impact on the customer base. In their 2016 survey on the state-of-the-art in business reporting, KPMG found that only 41% of the companies examined reported detailed customer information beyond the traditional sales performance based on the financial statement, mostly as single period-data. Measures capturing customer satisfaction and retention were present in 6% of the corporate disclosure investigated. In the medium term, there are increasing opportunities to better align corporate reporting with key value-creation strategies, particularly by exploiting the forward-looking ability of customer data to anticipate future financial performance.

5.4

The Rising Impact of Artificial Intelligence on Modeling Customer Data

Three major “disruptions” have shaped past decades of technological revolution: “Moore’s Law” (1971), fitting increasingly smaller and more powerful transistors on integrated circuits—computers; “Metcalf’s Law” (1995), which revealed the power of networks; and “Digital transformation” of today business environment, where it is possible to apply a “weather forecast attitude” for everything that happens in your life. The digital revolution is possible thanks to progress and exponential increase in computing power (Hyper-computing), large datasets available to train machine learning (from IoT, mobiles, social networks, and so forth), capillary diffusion of robotics. Gartner predicts that by 2020 85% of customer interactions will be managed without a human. Such developments have spurred sophisticated marketing applications relying on Artificial Intelligence (AI), an overarching concept which broadly refers to machines exhibiting intelligence inspired by human biological systems. Under the AI broad label, there are variety of software and algorithm-driven approaches (combined with large amount of data) to simulate human cognitive functions. Using advanced machine learning algorithms will increasingly allow to discover, classify and identify patterns in customer data, a trend commonly referred to as Deep Learning. For example, a bank would be interested to apply Deep Learning to analyze customers’ payment transactions with the objective to identify and anticipate potential fraudulent behaviour. In the hotel sector, machine learning is used in conjunction with advanced statistical methods to produce cutting-edge forecasting and decision

References

87

optimisation to better understand the relationship between price and demand, and generate room rates that dynamically adapt and anticipate market fluctuations. Furthermore, AI have put insurance companies in a better position to assess and manage the financial consequences associated with catastrophic storms. Such disruptive AI techniques would definitely help marketing researchers and managers to overcome the limitations inherent in CLV stochastic models proposed by the marketing literature. In particular, Deep Learning and neural network applications could be helpful to better model churn prediction as an effective and dynamic tool for companies that want to stay competitive in a rapidly growing market (see for example Sifa et al. 2018). We encourage academic researchers from various disciplines in management and computer sciences to cooperate and engage in active collaborations with companies to exploit their rich data environment and ultimately test the validity of customer metrics modelling and their predictive ability. Advances in theoretical knowledge and practical managerial applications in this dynamic area seem warranted for the next decade.

References ACCA. (2017). Insights into integrated reporting. London: Association of Chartered Certified Accountants. Ascarza, E., Fader, P. S., & Hardie, B. G. S. (2017). Marketing models for the customer-centric firm. In B. Wierenga & R. van der Lans (Eds.), Handbook of marketing decision models (pp. 297–329). Cham: Springer. European Commission. (2017, September 30). Non-financial reporting 2017. Available from https:// ec.europa.eu/info/business-economy-euro/company-reporting-and-auditing/company-reporting/ non-financial-reporting_en IIRC. (2013). The international framework. London: International Integrated Reporting Council. Jain, D. C., & Singh, S. S. (2010). Measuring customer lifetime value. In Review of marketing research (pp. 37–62). Bingley: Emerald Group Publishing. Jasek, P., Vrana, L., Sperkova, L., Smutny, Z., & Kobulsky, M. (2018). Modeling and application of customer lifetime value in online retail. Informatics, 5(1), 2. Lev, B., & Gu, F. (2016). The end of accounting and the path forward for investors and managers. New York: Wiley. Mzoughia, M. B., Borle, S., & Limam, M. (2018). A Mcmc approach for modeling customer lifetime behavior using the com-poisson distribution. Applied Stochastic Models in Business and Industry, 34(2), 113–127. Óskarsdóttir, M., Baesens, B., & Vanthienen, J. (2018). Profit-based model selection for customer retention using individual customer lifetime values. Big Data, 6(1), 53–65. Sifa, R., J. Runge, C. Bauckhage, and D. Klapper (2018). Customer lifetime value prediction in non-contractual freemium settings: Chasing high-value users using deep neural networks and smote. Proceedings of the 51st Hawaii International Conference on System Sciences. Tsai, T.-M., Wang, W.-N., Lin, Y.-T., & Choub, S.-C. (2015). An O2O commerce service framework and its effectiveness analysis with application to proximity commerce. Procedia Manufacturing, 3, 3498–3505. Verbraken, T., Verbeke, W., & Baesens, B. (2013). A novel profit maximizing metric for measuring classification performance of customer churn prediction models. IEEE Transactions on Knowledge and Data Engineering, 25(5), 961–973.