Customer Relationship Management: Emerging Concepts, Tools and Applications

MANAGING CUSTOMER RELATIONSHIPS 3 CHAPTER 1 MANAGING CUSTOMER RELATIONSHIPS RUTH N. BOLTON AND CRINA O. TARASI Abstract The customer relationship management (CRM) literature recognizes the long-run value of potential and current customers. Increased revenues, pro? ts, and shareholder value are the result of marketing activities directed toward developing, maintaining, and enhancing successful company–customer relationships. These activities require an in-depth understanding of the underlying sources of value that the ? rm both derives from customers, as well as delivers to customers.

We built our review from the perspective that customers are the building blocks of a ? rm. In order to endure long-term success, the role of marketing in a ? rm is to contribute to building strong market assets, including a valuable customer portfolio. CRM is an integral part of a company’s strategy, and its input should be actively considered in decisions regarding the development of organizational capabilities, the management of value creation, and the allocation of resources. CRM principles provide a strategic and tactical focus for identifying and realizing sources of value for the customer and the ? m and can guide ? ve key organizational processes: making strategic choices that foster organizational learning, creating value for customers and the ? rm, managing sources of value, investing resources across functions, organizational units, and channels, and globally optimizing product and customer portfolios. For each organizational process, we identify some of the challenges facing marketing scientists and practitioners, and develop an extensive research agenda. Companies are increasingly focused on managing customer relationships, the customer asset, or customer equity.

Need essay sample on Customer Relationship Management: Emerging Concepts, Tools... ?We will write a custom essay sample specifically for you for only $12.90/page

order now

Customer relationship management (CRM) explicitly recognizes the long-run value of potential and current customers, and seeks to increase revenues, pro? ts, and shareholder value through targeted marketing activities directed toward developing, maintaining, and enhancing successful company-customer relationships (Berry, 1983, p. 25; Morgan and Hunt, 1994, p. 22; Gronroos, 1990 p. 138). These activities require an in-depth understanding of the underlying sources of value the ? rm both derives from customers and delivers to them.

The purpose of this chapter is to describe how companies can effectively cultivate customer relationships and develop customer portfolios that increase shareholder value in the long run. We review the extensive literature on customer relationship management, customer asset management, and customer portfolio management, and summarize key ? ndings. The chapter has three major components. First, we de? ne CRM, describe how marketing thinking about CRM has evolved over time, and assess whether CRM principles and systems have improved business performance (to date). Second, we examine (in detail) ? e organizational processes that we believe are necessary for effective CRM: making strategic choices that foster organizational learning, creating value 3 4 RUTH N. BOLTON AND CRINA O. TARASI for customers and the ? rm, managing sources of value (acquisition, retention, etc. ), investing resources across functions, organizational units, and channels, and globally optimizing product and customer portfolios. We describe each process, summarize key ? ndings, identify emerging trends and issues, and predict likely future developments (both theoretical and methodological).

Our concluding remarks make recommendations about areas where further research is needed. Perspective on the Evolution of Customer Relationship Management Current De? nition of CRM After surveying many alternative de? nitions of CRM, Payne and Frow (2005, p. 168) offer the following comprehensive de? nition, which we will use to frame the discussion in our chapter: CRM is a strategic approach concerned with creating improved shareholder value through the development of appropriate relationships with key customers and customer segments.

CRM unites the potential of relationship marketing strategies and IT [information technology] to create pro? table, long-term relationships with customers and other key stakeholders. CRM provides enhanced opportunities to use data and information to both understand customers and co-create value with them. This requires a cross-functional integration of processes, people, operations and marketing capabilities that is enabled through information, technology and applications.

Researchers have emphasized different CRM issues depending on whether they are considering a business-to-consumer or business-to-business context. However, we focus on conceptual and methodological principles that are applicable in both contexts, highlighting noteworthy exceptions. CRM vis-a-vis the Domain of Marketing Marketing theory has frequently provided guidance on how ? rms should react to opportunities, but marketing actions are also able to change the environment and create opportunities (Zeithaml and Zeithaml, 1984).

Marketing—considered as a general management responsibility—plays “the crucial roles of (1) navigation through effective market sensing, (2) articulation of the new value proposition, and (3) orchestration by providing the essential glue that ensures a coherent whole” (Hunt, 2004, p. 22). CRM enhances these capabilities because it is “the outcome of the continuing evolution and integration of marketing ideas and newly available data, technologies and organizational forms” (Boulding et al. , 2005). CRM principles and systems help organizations to focus on the dual creation of value: the creation of value for shareholders (via long-term ? m pro? tability) and the creation of value or utility for customers (Vargo and Lusch, 2004). These objectives are congruent because relationships represent market-based assets that a ? rm continuously invests in, in order to be viable in the marketplace. Strong relationships are associated with customer loyalty and/or switching costs, which create barriers to competition. Thus relationships provide a differential advantage by making resources directed to customers more ef? cient. For example, loyal customers are more responsive to marketing actions and cross-selling (Verhoef, 2003).

Marketers sometimes use the term “customer asset,” but customers and assets do not have identical features. The mind-set associated with “owning” customers is dangerous because customer MANAGING CUSTOMER RELATIONSHIPS 5 relationships must be carefully managed and customer loyalty must be earned (Rust et al. , 2004). However, the customer base is certainly a market-based asset that should be measured, managed, and tracked over time (Bell et al. , 2002). Srivastava, Shervani, and Fahey (1998) discuss how market-based assets, such as customer or partner relationships, can increase shareholder value by accelerating and enhancing cash ? ws, lowering the volatility and vulnerability of cash ? ows, and increasing the residual value of cash ? ows. Their framework links customer relationship management with business performance metrics. Origins in Relationship Marketing The foundation for the development of CRM is generally considered to be relationship marketing, de? ned as marketing activities that attract, maintain, and enhance customer relationships (Berry 1983). Gronroos (1990, p. 138) argues for the importance of relationships in the marketing context. He proposes a de? ition for marketing, namely, that marketing is “to establish, maintain and enhance relationships with consumers and other partners, so that the objectives of the parties involved are met. This is achieved by a mutual exchange and ful? llment of promises. ” However, although the terms “CRM” and “relationship marketing” are relatively new, the phenomenon is not (Gummesson, 1994, p. 5, 2002, p. 295). Marketers have always been preoccupied with defensive strategies aimed at increasing customer retention, thereby increasing revenues and pro? tability (Fornell and Wernerfelt, 1987).

For example, writing in the Harvard Business Review, Grant and Schlesinger (1995 p. 61) argue that the gap between organization’s current and full-potential pro? tability is enormous, and suggest that managers ask themselves: “How long on average do your customers remain with the company? [and] What if they remained customers for life? ” During the same time period, a growing literature has focused on the “service pro? t chain” linking employee satisfaction, customer satisfaction, loyalty, and pro? tability (e. g. , Heskett, Sasser, and Schlesinger, 1997; Reichheld, 1993; Liljander, 2000).

Emergence of Customer Equity and Early Customer Relationship Models This perspective naturally evolved and expanded to consider the management of customer equity or the value of the customer base. Initially, researchers were primarily concerned with the allocation of resources between customer acquisition and retention (Blattberg and Deighton, 1996). Generally, the management of customer equity requires that organizations use information about customers and potential customers to segment them and treat them differently depending on their future long-term pro? ability (Blattberg, Getz, and Thomas, 2001; Peppers and Rogers, 2005; Rust, Zeithaml, and Lemon, 2000). Notably, ? rms must go beyond traditional market segmentation activities, such as customizing offerings (i. e. , goods or services) and ef? ciently managing resources to achieve pro? tability criteria. Instead, ? rms must identify and acquire customers who are not only willing to accept the ? rm’s offer or value proposition—but also provide value for the company when they do (e. g. , Cao and Gruca, 2005; Ryals, 2005). Marketers were quick to recognize that the value of the customer asset (i. e. the value a customer or potential customer provides to a company) is the sum of the discounted net contribution margins of the customer over time—that is, the revenue provided to the company less the company’s cost associated with maintaining a relationship with the customer (Berger and Nasr, 1998). Early applications of CRM systems typically utilized models that predict (rather than explain) future customer behavior or pro? tability. For example, in an early paper, Schmittlein and Peterson (1994) use past purchase behavior—that is, data on the frequency, timing, and dollar value of past purchases—to RUTH N. BOLTON AND CRINA O. TARASI predict likely future purchase patterns. They were able to show that their “customer base analysis” was effective in predicting purchase patterns for different key industrial buying groups. For about a decade, relatively narrow CRM systems coexisted, rather uneasily, with broader, strategically meaningful conceptualizations of CRM as a “strategic bridge between information technology and marketing strategies aimed at building long term relationship and pro? tability” (Ryals and Payne 2001, p. ). Modelers frequently applied CLV concepts in direct marketing, database marketing, or electronic commerce contexts (Ansari and Mela, 2003; Bult and Wansbeek, 1995; Elsner, Krafft, and Huchzermeier, 2004). 1 Progress was made toward identifying which variables are the “best” predictors of customer lifetime pro? tability (in a given study context). For example, Reinartz and Kumar (2003) compare traditional models that consider frequency, timing, and monetary value with models that show how managerial decision variables in? ence the pro? tability of customers over time—and show that the latter are superior. Nevertheless, most applications (to date) have relied on estimates of current customer pro? tability, rather than future customer pro? tability. Customer Relationship Management and Business Performance Marketing Metrics The challenges of applying CRM principles were exacerbated as managers and researchers turned their attention to “metrics” or the measurement of the impact of marketing on business performance (cf. Lehmann, 2004).

Most popular measures of current CRM systems are outcome measures: number of acquired customers, “churn” as a percentage of the customer base (the inverse of the customer retention rate), the dollar value of cross-selling, the percentage increase in customer migration to higher margin products, changes in individual customer lifetime value (CLV), and so forth. Any single outcome measure provides an incomplete and (often) short-run assessment of the ? rm’s success at creating value for both customers and shareholders (Boulding et al. , 2005).

Most dangerously, optimizing a small number of outcome measures may lead to core rigidities (Atuahene-Gima, 2005; Leonard-Barton, 1992) that undermine the organization’s core capabilities and lead to business failure. For example, there are numerous stories of ? rms that have focused on customer acquisition at the expense of customer retention activities or vice versa. One way to assess the impact of marketing on business performance is to forecast the lifetime value of individual customers under alternative scenarios, aggregating across customers, and identifying the “best” set of scenarios or set of organizational actions.

This approach seems “doable” but it can be challenging to move from the calculation of individual customers’ lifetime revenues to individual customers’ pro? tability. For example, Niraj, Gupta, and Narasimhan (2001) demonstrate this method for an intermediary in a supply chain, such as a distributor, where costs are incurred at each step in the supply chain and there is heterogeneity in purchasing characteristics. Initial Failure of CRM “Systems” A constructive distinction is often missing in CRM frameworks.

There is a difference between CRM systems—software that integrates relevant customer information (sales, marketing, etc. ) with product and service information—and CRM processes, for example, the cross-functional steps required to ensure customer retention and effectiveness of marketing initiative, such as a continuing dialogue with customers across all contact points and personalized treatment (Day, 2000). In other words, CRM systems are intended to support CRM processes, which are meant to enhance the value of the customer relationship.

MANAGING CUSTOMER RELATIONSHIPS 7 CRM starts from the fundamental assumption that the bounded rationality of humans charged with initiating, maintaining, and building relationships can be supported and enhanced by speci? c organization capabilities, namely, the intelligent utilization of databases and information technology. However, many organizations’ initial experiences were disappointing, especially in the short run. The Economist (2003, p. 16) describes the experiences of ? ancial services organizations and pessimistically observes that: The three year economic downturn has cooled even Wall Street’s ardor for fancy new IT [information technology] gear. . . . The problem is that most IT projects are lengthy affairs and notoriously “back loaded. ” . . . Few things in technology have promised so much and delivered so little as “customer (or client) relationship management” (CRM) software. In implementing CRM, insiders reckon that four out of ? ve such projects fail to deliver the goods.

These failures typically arose from a narrow application of CRM principles. For example, Rigby, Reichheld, and Schefter (2002) identi? ed four situations that independently and together result in failed CRM systems: (1) implementing CRM without having in place a clear customer strategy, (2) assuming that CRM has to match organizations’ current practices, and not enhance them, (3) assuming that CRM technology and not CRM strategy matters, and (4) using CRM to stalk, not to woo customers.

In other words, many so-called CRM systems used technology (both hardware and software) to optimize the usage of information within functional silos, without a relational orientation, creating obstacles to organizational learning and the dual creation of value. Thus, it is not particularly surprising that they identi? ed solutions that were suboptimal—and even unpro? table—in the long run. More Nuanced Approaches to Evaluating CRM Systems and Technology Research has established that CRM systems can improve intermediate measures of business performance.

For example, Mithas, Krishnan, and Fornell (2005) study the effect of CRM applications on customers and ? nd out that the use of CRM systems positively impacts customer satisfaction, both directly and through improved customer knowledge. Despite this fact—and the common belief that more and better customer knowledge can only bene? t a ? rm and its customers—the ? nancial return on large investments in CRM technology has been questioned. For example, as Reinartz, Krafft, and Hoyer (2004, p. 93) report, commercial studies “provide some convergent validity that approximately 70 percent of CRM projects result in either losses or no bottom line improvements. ” Contrary to such reports, their own empirical investigation indicates that companies that implemented CRM processes performed better not only in relationship maintenance but also in relationship initiation. A critical issue for many organizations is that the adoption of CRM technology is fraught with implementation challenges, including information technology design, procedure, and process issues, dif? ulties in maintaining accurate and current information, obstacles arising from interfaces that are not user friendly, and so forth (e. g. , Johnson, Sohi, and Grewal, 2004; Meuter et al. , 2005; Morgan, Anderson, and Mittal, 2005; Winer, 2001). For this reason, we must distinguish between technology-driven implementation—which results in user frustration—and customer-driven implementation—which has high user involvement; the latter resulted in successful operational CRM systems.

A recent study by Jayachandran and colleagues (2005) estimates an interaction effect showing that customer relationship performance for a diverse sample of businesses is enhanced 8 RUTH N. BOLTON AND CRINA O. TARASI by organizational information processes when a high level of technology is used. In other words, technology use for customer relationship management—by moderating the in? uence of organizational information processes on customer relationship performance—performs a supportive role only. They show that effective organizational information processes (i. e. effective communication, information capture, and information integration, as well as access and use of information) enhance the effectiveness of CRM technology in achieving business success. CRM Principles and the Role of Organizational Capabilities and Processes After more than twenty years of research on CRM, the accumulated evidence indicates that the application of CRM principles yields positive ? nancial outcomes. In their introduction to the Journal of Marketing’s special section on CRM, Boulding and colleagues (2005) argue that CRM improves business performance in a wide variety of industry settings.

A striking example is described in a case study by Ryals (2005), showing that a business unit was able to achieve a 270 percent increase in business unit pro? ts above target by implementing some straightforward CRM procedures. Why do ? rms experience such widely varying degrees of success from applying CRM? The implementation of CRM systems or technology alone is doomed to fail, because the collection of the data does not imply the existence of useful information that will be disseminated and acted upon appropriately.

Boulding and colleagues (2005) argue that, holding ? xed the level of CRM investment, the effectiveness of CRM activities depends on (a) how CRM is integrated with the existing processes of the ? rm and (b) the ? rm’s preexisting capabilities. In other words, organizations that have already developed learning capabilities and effective information processes are more likely to improve their business performance by adopting CRM systems. They are able interpret information correctly and act on it in a manner to increase value for both the customer and the ? m. In a recent Harvard Business Review article, Gulati and Oldroyd (2005) observe that the implementation or CRM systems must serve the purpose of getting closer to customers, and that in order to succeed the company as a whole has to engage in a learning journey—learning about the customer and about the business and how its way of doing business can be improved. If this activity is regarded as a departmental or functional responsibility, CRM efforts will fail.

The authors identify four stages in the evolution of a successful CRM implementation: communal coordination (gathering information); serial coordination (gaining insight from customers’ past behavior); symbiotic coordination (learning to predict future customer behavior) and integral coordination (real time response to customer needs). This evolutionary and transformational process takes time, resources, and patience, but the implementation of each of the stages should provide visible end results.

Harrah’s started this process under Gary Loveman’s leadership in 1998 and, after a constant evolution that took more than seven years and involved all employee levels, it enjoyed impressive growth compared to competitors. Furthermore, the deep understanding of the customer provided new levers for future growth (Gulati and Oldroyd, 2005; Gupta and Lehmann, 2005). In summary, marketing science and practice has moved away from simplistic evaluations of investments in CRM technology or systems to consider the role of ? ms’ preexisting capabilities and organizational processes. For this reason, the remainder of this article frames our discussion of what we know about CRM in terms of ? ve interrelated organizational processes: making strategic choices that foster organizational learning, creating value for customers and the ? rm, managing sources of value (acquisition, retention, etc. ), investing resources across functions, organizational units, and channels, and globally optimizing product and customer portfolios. We discuss how each MANAGING CUSTOMER RELATIONSHIPS Figure 1. 1 Customer Relationship Management Processes Figure 1. 1 Customer Relationship Management Processes Strategic Choices Dual creation of value Managing sources of value Investments across functions Global optimization = Existing relational processes Table 1. 1 Processes Strategic choices Dual creation of value Customer portfolio management • Organizational information processes • O rganizational learning • Creating value for customers • Valuing customers • Acquisition • Retention • Increased margins from relationship expansion activities (e. . , product usage, cross-selling) • Divestment • Employee selection and training • Service quality • Customer management effort • Managing customer contacts • Customer equity models • Segmentation • Matching product portfolio and customer portfolio • Risk/return management Allocation of resources across functions, channels, and organizational units Global optimization models process in? uences the effectiveness of CRM, and describe its challenges. The processes and their relationships are depicted in Figure 1. 1; subtopics are listed in Table 1. 1.

We begin by describing research regarding how organizations’ strategic choices in? uence the effectiveness of CRM in enhancing business performance, which provides a conceptual rationale for our framework. Strategic Choices In a recent executive roundtable discussion, executives from IBM, Yellow-Roadway, Luxottica Retail (Lens Crafters and Sunglass Hut), McKinsey & Company and Cisco Systems stated that that there were immense opportunities for the transformation of organizations through the integration of business processes and the use of technology to generate competitive advantage, cost saving ef? iencies and an enhanced customer experience. Executives in Europe and North America strongly believe that successful organizations require a cross-functional process-oriented approach that positions CRM at a strategic level (Brown, 2005; Christopher, Payne, and Ballantyne, 1991; Payne and Frow, 2005). This notion is consistent with empirical evidence showing that ? rms’ prior strategic commitments (as opposed to their general market orientation) have impressive effects on the performance of their CRM investments in a retailing context (Srinivasan and Moorman, 2005). 10 RUTH N.

BOLTON AND CRINA O. TARASI Organizational Learning Based on extensive ? eld interviews, Payne and Frow (2005) identify ? ve key cross-functional CRM processes: a strategy development process; a value creation process; a multichannel integration process; an information management process; and a performance assessment process. They argue that an organization’s strategy development process—a precursor for subsequent processes—requires a dual focus on its business strategy and customer strategy, and that how well the two interrelate will fundamentally affect the success of its CRM strategy.

In particular, organizational information processes—information reciprocity, information capture, information integration, information access, and information use—relevant to CRM can play a vital role in enhancing business performance (Jayachandran et al. , 2005). This observation should not be surprising because the primary outcome of the adoption of CRM technology is the generation of an enormous database describing customer pro? les, sales, costs, operations, and so forth. If intelligently processed and interpreted, these data can provide information regarding the value of customers and the effectiveness and ef? iency of marketing actions (Berger et al. , 2002). Each customer interaction is (or should be) part of an iterative learning process both from the customer and the company points of view (Ballantyne, 2004). Challenges Our review of prior research suggests two fruitful areas for future research. First, marketing scientists and practitioners have acknowledged that CRM technology alone cannot sustain a competitive advantage. The failure of many ? rms to reap economic rewards from investments in CRM technology is a symptom of an underlying problem, namely, how to create a coordinated strategy that integrates business processes and enerates an enhanced customer experience (i. e. , the creation of value for customers), competitive advantage, and cost saving ef? ciencies (i. e. , the creation of value for the ? rm). The value a company has to offer to its customer is derived not only from the quality of its offerings but also from its relational characteristics and supplier characteristics (Crosby, Gronroos, and Johnson, 2002; Menon, Homburg, and Beutin, 2005; Storbacka, Strandvik, and Gronroos, 1994). For this reason, appropriate organizational structures and processes for a given ? m are likely to depend on its business environment (i. e. , they will be contingency-based). Thus, there is a critical need for more research on how CRM principles can guide strategic choices that improve business performance in different business contexts, thereby bridging the functional silos that exist in many organizations. Otherwise, ? rms will be unable to pro? tably exploit innovations in technology and business processes—for example, radio frequency identi? cation technology. Second, ? ms’ experiences in implementing CRM technology have shown that transforming data into useful information—especially learning from past experience—is challenging for many organizations. Ambler (2003, p. 21) points out a paradox: “Marketing is the means whereby a company achieves its key objectives,” but quantifying the results of marketing actions is extremely challenging. CRM systems can provide the tools for accurately measuring marketing outcomes, where “clarity of goals and metrics separate the professional from the amateur” (Ambler, 2003, p. 17). Gupta and Lehmann (2005) have suggested a set of metrics that is based on a pro? ability tree and is suitable for strategic decision-making. It is important to recognize that different metrics are required for different purposes. Hence, research is required to identify metrics linked to future pro? tability because, without making sense of the interrelationships of marketing variables, it will be impossible for marketing to evolve from a function in a company to a guiding principle (Hunt MANAGING CUSTOMER RELATIONSHIPS 11 2004). In addition, research is required to show how metrics can be used to manage value creation for customers and for the ? rm.

Furthermore, at an implementation level, research is required to develop “interlocking” metrics that coordinate decision making at strategic and tactical levels, as well as decision making across channels and organizational units. Dual Creation of Value Dual creation of value requires that the ? rm simultaneously create value for customers and value for shareholders. First, we discuss how to create value for customers. Second, we consider how managers can assess the value of individual customers or segments, and then aggregate them to calculate the value of the customer base to the ? m. We identify the research challenges associated with each task. Creating Value for Customers A common trait of many studies is a focus on measuring CRM’s impact on the end results, such as pro? ts and shareholder value, without studying the relations among processes and connections among variables (Boulding et al. , 2005). Return on investment is certainly a measure of success, but—without a profound understanding of how relational processes can operate effectively—success from CRM initiatives is elusive. Although the speci? cs will be unique to each ? m, prior research provides a conceptual framework for understanding how relational processes create value for customers. Speci? cally, research on the antecedents of service quality, customer satisfaction, trust, and commitment provide insights for managers (Berger et al. , 2002; Rust, Lemon, and Zeithaml, 2004). Relationships with Consumers Research on CRM is a natural evolution of marketers’ longstanding interest in understanding how relationships with individual customers are created, built, and sustained over time (Bhattacharya and Bolton, 2000).

It began with investigations of how customers formed their assessments of products (goods and services). This research stream is extensive; therefore an extensive discussion of the antecedents of customer assessments (e. g. , perceived service quality and customer satisfaction) as well as the implicit bonds (e. g. , legal, economic, technological, knowledge, social, etc. ) (Liljander and Strandvik, 1995) is beyond the scope of this section. Notably, customer satisfaction literature developed around the idea that satisfaction is in? enced by the difference between expectations and experience (Oliver, 1980, 1999). Service quality literature developed along parallel lines (cf. , Parasuraman, Zeithaml, and Berry, 1985, 1988). For example, Boulding and colleagues (1993) brought together two streams of service quality research in showing that both expectations as predictions (expectations about what will happen) and normative expectations (expectations about what should happen, often based on communications from the service provider) are important in determining perceived service quality.

This stream of literature is extremely useful in helping researchers build theory-based models of customer behavior (Bolton and Lemon, 1999). Business-to-Business Relationships Researchers focusing on CRM principles have been especially interested in interorganizational relationships because—until the recent advent of electronic commerce with its potential for 12 RUTH N. BOLTON AND CRINA O. TARASI precise (one-to-one) targeting of marketing activities to customers—business-to-business (B2B) relationships have been the most fruitful context for the application of the principles of customer relationship management.

This stream of research has tended to have a strategic orientation, re? ecting the notion that a coherent set of cross-functional activities is required to create, build, and sustain relationships (Ford, 1990). 2 Two important focal constructs in understanding interorganizational relationships are trust and commitment (Morgan and Hunt, 1994). For example, Anderson and Weitz (1992) consider how commitment depends on self-reported and perceived “pledges” (i. e. , idiosyncratic investments and contractual terms), communication, and relationship characteristics.

Their research is particularly noteworthy because they studied 378 dyads—that is, pairs of manufacturer and industrial distributors—so that they were able to model the antecedents and consequences of each party’s perception of the other party’s commitment. Recent research has extended our knowledge of interorganizational relationships through studies of organizational norms, contracting, opportunism, and so forth (Heide and Weiss, 1995; Kalwani and Narayandas, 1995; Kumar and Corsten, 2005; Narayandas and Rangan, 2004; Wuyt and Geyskens, 2005). B2B decisions are especially complex because multiple eople participate in the purchase decision (e. g. , purchasing manager, end user, decision maker), and interactions occur at multiple levels (e. g. , contract level, organizational unit level, ? rm level). This research stream is very helpful in building theory-based models of organizational buying behavior. Most prior research has been conducted at the enterprise level, using key informants; future research is required that uses information obtained from multiple informants as well as from multiple levels within the buying organization (Bolton, Lemon, and Bramlett, 2004).

Using Customer Assessments of Relationships to Explain Behavior Numerous studies have shown that self-reports of customer assessments (such as satisfaction) can explain customer behavior. Bolton (1998) models the duration of the customer–? rm relationship at the individual level. She ? nds that prior cumulative satisfaction is weighed more heavily than satisfaction from recent events, and that satis? ed customers have longer relationships, and generate greater revenues and pro? ts (for contractual relationships). However, Verhoef (2003) ? nds that, if customer assessments primarily re? ct cognition (without an affective component), it may prove dif? cult to predict customer retention or share of the wallet. At the aggregate level, Gruca and Rego (2005) use data from the American Customer Satisfaction Index and Compustat to show that customer satisfaction plays a major role in increasing cash ? ow and enhancing its stability. Challenges CRM systems operate at the customer–? rm interface, and ? rms frequently use information from customers to create and deliver valuable offerings to them. Customers are likely to be willing to reveal private information if they derive “fair” value from exchanges with the ? m. However, ? rms may behave opportunistically (extracting all economic surplus), creating mistrust among customers, so that they act strategically when they provide information or participate in transactions with the ? rm (Boulding et al. , 2005). For example, customers might retaliate against perceived unfairness by providing inaccurate information, generating unfavorable word of mouth, switching to the competition, or boycotting the ? rm. Consequently, successful implementation of CRM principles requires that ? rms carefully consider issues related to privacy and fairness (Boulding et al. , 2005).

Additional research is required on how these constructs in? uence business performance in the long run. MANAGING CUSTOMER RELATIONSHIPS 13 Mediating constructs, such as perceived fairness, satisfaction, and commitment, are important precursors of customer behavior. Moreover, prior research has shown that self-report measures obtained from survey data can be used to predict customer behavior (e. g. , Bolton and Lemon, 1999). Researchers have also used survey measures as proxies for consumer behavior, assuming that the antecedents of the proxy are identical to the antecedents of the target variable.

However, there is a signi? cant body of literature that shows otherwise (Chandon, Morwitz, and Reinartz, 2005; Morwitz, 1997; Morwitz and Schmittlein, 1992; Seiders et al. , 2005). For example, Mittal and Kamakura (2001) analyze the in? uence of satisfaction on behavioral intentions and actual behavior and ? nd that the effect of satisfaction on behavioral intentions is nonlinear with decreasing returns, whereas its effect on behavior is nonlinear with increasing returns. For this reason, marketers must be cautious about using only survey data to study how relational processes create value for customers.

Hence, there is also a need for additional research to develop more longitudinal models of customer behavior (Bolton, Lemon, and Verhoef 2004). Value of Customers to the Firm Customer Valuation The value of the customer asset (i. e. , the value that the customer provides to a company) is the sum of the customer’s discounted net contribution margins over time—that is, the revenue provided to the company less the company’s cost associated with maintaining a relationship with the customer (Berger and Nasr, 1998). Naturally, a company cannot perfectly predict the cash ? ws associated with an individual customer, but it can calculate the expected value of the cash ? ows (adjusting for risk) associated with an individual customer conditional on the customer’s characteristics, the company’s planned marketing actions and environmental factors (Hogan et al. , 2002). For example, Pfeifer and Bang (2005) propose a model of calculating the mean CLV taking into account the fact that customers have not completed their purchasing cycle and therefore any mean calculation of their value is inaccurate because it does not include future purchases.

They use a nonparametric method to compute mean CLV across all customers, to be used as guidance for the appropriate level of investment in customers. Gupta, Lehmann, and Stuart (2004) propose forecasting CLV by decomposing it into three underlying sources: customer acquisition (i. e. , trial), retention (repeat purchase behavior), and gross margins (in? uenced cross-buying, cost structure, etc). They demonstrate that the basic calculations are relatively straightforward.

Research has shown that the CLV framework can be used to generate estimates of the future pro? tability of individual customers—given certain marketing actions and competitive conditions—and to identify optimal allocations of resources (cf. , Jain and Singh, 2002; Kumar, Ramani, and Bohling, 2004). In contrast, substantial empirical evidence—using rigorous holdout sample procedures—indicates that measures of the past pro? tability of individual customers are poor predictors of future customer pro? tability (Campbell and Frei, 2004; Malthouse and Blattberg, 2005).

Forecasting Sources of CLV To ensure accuracy, it is recommended that estimates of the revenue sources of CLV should be broken down to the customer or cohort or segment level (rather than the ? rm level). Customer-level forecasts of each source are preferable for ? ve reasons (Gupta and Lehmann, 2005, pp. 7–9). First, customer-level pro? tability can be decomposed into its underlying sources—customer acquisition, 14 RUTH N. BOLTON AND CRINA O. TARASI retention, and margin—which are amenable to managerial action. Second, by preparing forecasts of each underlying source (rather than extrapolating ? m-level historical data), managers can explicitly account for changes over time in the underlying sources of pro? tability, thereby identifying turning points. For example, a ? rm might discover that its constant earnings over the past few years are the net result of increases in customer acquisition rates and decreases in margins. Further analysis might reveal that customer acquisition will slow down, causing a decline in future earnings. Third, projected customer revenues can take into account any effects of cross-selling (which increase margins) and word-of-mouth. Fourth, the effect f a planned marketing action will be different for each CLV source: acquisition, retention, and margins (Bolton, Lemon, and Verhoef 2004). For example, Thomas and Reinartz (2003) show that the amount of direct mail sent has an effect on cross-buying opposite to that on purchase frequency. Fifth, without considering customers’ migratory behavior, customers will be undervalued since they are considered lost when they switch to competition and they are accounted for as new customers when they switch back (for a model of accounting for switching behavior see Rust et al. , 2004). To calculate CLV and identify the most pro? able customers, the company must forecast the cost to serve a customer as well as revenue sources. As Kaplan and Narayanan (2001) point out, the cost to serve customers can vary dramatically: 20 percent of customers who are most pro? table can account for 150 percent to 300 percent of pro? ts, while the 10 percent who are least pro? table may lose 50 percent to 200 percent of pro? ts. Under these conditions, it is necessary to measure the real pro? tability of customers and (if necessary) take corrective actions to forestall losses (either by “? ring” the unpro? table customers or by adopting solutions to make the relationship pro? able). Firm Valuation Recent research has shown that the CLV framework (i. e. , using forecasts of acquisition, retention, and margins) can be used to calculate the value of the ? rm’s current and future customer base. Gupta, Lehmann, and Stuart (2004) use publicly available information from annual reports and other ? nancial statements to calculate a customer-based valuation of ? ve companies. They compare their estimates of customer value (post-tax) with the reported market value for each of the companies. Their estimates are reasonably close to the market values for three ? rms, and signi? antly lower for two ? rms (Amazon and eBay). They infer that these two ? rms either are likely to achieve higher growth rates in customers or margins than they forecast, or they have some other large option value that the CLV framework does not capture. Challenges Berger and colleagues (2002) discuss four critical and interrelated actions required of ? rms that wish to understand how their actions affect the value of their customer assets: (1) create a database; (2) segment based on customer needs and behavior; (3) forecast CLV under alternative resource allocation scenarios; and (4) allocate resources.

Although the challenges of creating an integrated database cannot be overestimated, they are primary related to cost and implementation issues. In contrast, forecasting customer-level CLV is a signi? cant technical challenge for four reasons. First, the forecasts should re? ect changes in customer behavior in response to changes in organizational decisions and the environment. To make CLV calculations tractable prior research has made strong implicit assumptions about customer behavior and marketing programs (e. g. , Berger and Nasr, 1998; Blattberg and Deighton, 1996; Dwyer, 1989; Rust et al. 2004). For example, researchers frequently assume ? xed marketing programs, deterministic retention rates, MANAGING CUSTOMER RELATIONSHIPS 15 and stable switching patterns among competitive offerings. Additional research is required to relax these assumptions in practical situations. For example, Lewis (2005) estimated a structural dynamic programming model that accounts for the effects of marketing variables, past purchasing activity, consumer expectations of future promotions, and preference heterogeneity on consumer behavior regarding online grocery purchases.

The model was used to simulate customer response to marketing programs over an extended time period, thereby providing an estimate of customer value that is directly connected to organizational decisions. He found that, relative to a holdout sample, the simulation-based forecasts outperformed standard methods in terms of absolute error and were better able to account for variation in long-term values in a heterogeneous customer base. He was also able to estimate the long-term consequences of alternative pricing and promotion strategies.

Second, different customers will value the same product differently, and they will have different acquisition rates, retention rates, and margins (due to cross-buying); therefore, forecasting models must account for customer heterogeneity (cf. , Chintagunta and Prasad 1998; Schmittlein and Peterson 1994). Third, it will be necessary to allocate costs to individual customers. In direct marketing contexts, ? rms are able to assign the costs of direct communication, delivery of the product, and promotions to individual customers (Berger and Nasr-Bechwati, 2001; Dwyer, 1989; Keane and Wang, 1995).

However, in many industries, ? rms must create methods for accurately attributing the indirect costs of marketing actions to individual customers or customer segments. Berger and colleagues (2002) point out that cost allocation can be particularly challenging for ? rms that invest in programmatic efforts, such as service improvement efforts or investments in physical infrastructure. A fourth challenge is to understand and incorporate competitive effects on customer acquisition and retention. Accounting for competitors’ acquisition campaigns might explain customer behavior in most markets.

Optical scanner data provide competitive information in retail environments, but information about competitive behavior is seldom available in other contexts. Managing Sources of Value Organizations can manage sources of value by acquiring and retaining the most desirable customers; expanding relationships through the stimulation of usage, upgrades, and cross-buying; improving their overall pro? tability by adjusting prices or managing costs; and managing the customer and product portfolios. Since not all customers are equally pro? table, investments in customers should be based on their pro? potential, as illustrated in Table 1. 2. Firms should acquire customers in the upper-right quadrant and divest customers in the lower-left quadrant. Vulnerable customers may defect to competitors unless the ? rm develops an appropriate marketing program to retain them; free riders should receive lower product quality and higher prices. These strategies require the ? rm to develop marketing programs targeted at individual customers or segments that in? uence acquisition, retention, and margins (via cross-buying), thereby maximizing CLV and value for customers.

Marketers have developed a substantial body of knowledge about how ? rm actions in? uence customer behavior. A useful summary of this literature is provided by Bolton, Lemon, and Verhoef (2004), who identify six categories of marketing decision variables that can be used to in? uence customer behavior and CLV: price, service quality programs, direct marketing promotions, relationship marketing instruments (e. g. , rewards programs), advertising communications, and distribution channels. In the following paragraphs, we brie? summarize some key considerations concerning how these marketing actions in? uence each source of value. 16 RUTH N. BOLTON AND CRINA O. TARASI Table 1. 2 Comparison of Value of Customers to the Firm with Value to Customers LOW Value to Customers HIGH Value of Customers LOW Value of Customers Vulnerable Customers Lost Causes HIGH Value to Customers Star Customers Free Riders Source: Gupta and Lehmann (2005), p. 44. Customer Acquisition Customer acquisition is a ? rst step in building a customer base. Targeting, acquiring, and keeping the “right” customers entails a consideration of ? with current ? rm offering, future pro? tability, and contribution to the overall business risk. Many ? rms do not employ appropriate criteria to identify pro? table customers and their marketing programs are broadly communicated to potential customers who may or may not be pro? table. Consequently, customer acquisition can be a costly and risky process—especially because new customers may not represent a good ? t for the organization’s value proposition, a phenomenon that can often occur if acquisition is done outside previously targeted segments. Customer–product ? becomes important because campaigns aimed toward new customers—that change the positioning of a product—can alienate existing customers. Mittal and Kamakura (2001) discuss the nature of the relationship (or ? t) of the customer and the brand, ? nding that customers with different characteristics have different satisfaction thresholds, and, therefore, different probabilities of repurchase. 3 This leads to the more general observation that customer acquisition in? uences the diversity of the customer portfolio—thereby in? uencing business risk—but this aspect of CRM is rarely studied in marketing (Johnson and Selnes, 2005).

Lack of focus during acquisition activities is very likely to result in adverse selection—whereby the prospects that are least likely to be pro? table are mostly likely to respond to marketing efforts. For credit companies, the problem is particularly worrisome because they must verify the suitability of all respondents, thus incurring screening costs. Cao and Gruca (2005) address the problem of adverse selection by using data from a ? rm’s CRM system to target prospects likely to respond and be approved. This approach increases the number of customers who are approved while reducing the number of “bad” customers.

Their analysis is post facto and the marketing message is not altered, but their results show 30 percent to 75 percent improvements compared to traditional models that take into account either response likelihood or approval likelihood but not both. This method can be extended to new customer acquisition and better targeting of costly promotions to migrate customers to higher levels of lifetime value. Customer Retention Even though the optimal mix of marketing programs is unique to each business model, customer retention is often easier and cheaper than customer acquisition, especially in stable markets with low growth rates.

An organizational emphasis on customer retention also makes sense when discount rates are low (Gupta and Lehmann, 2005). Hence, customer retention has received considerable attention from marketers. In fact, many organizations have considered the management of CLV as equivalent to the management of customer retention, and have ignored the MANAGING CUSTOMER RELATIONSHIPS 17 contribution of other sources of CLV. 4 Research con? rms that consumers with higher satisfaction levels and better price perceptions have longer relationships with ? rms (e. g. , Bolton, 1998).

In a B2B context, suppliers who have long-term relationships with customers are able to achieve signi? cant sales growth and higher pro? tability through differential reductions in discretionary expenses (Kalwani and Narayandas, 1995). However, customer retention and defection are complex processes (Akerlund, 2005). Relationship Expansion Organizations can increase CLV and gross margin per customer by stimulating increased product usage or cross-buying (cf. , Hogan et al. , 2002). However, marketing programs designed to expand relationships with customers have received much less attention than programs for retaining customers.

Customer loyalty and cross-buying may be simultaneously determined in some contexts. However, in a direct mail context, Thomas and Reinartz (2003) have shown that cross-buying is a consequence, and not an antecedent, of loyalty behaviors. Nevertheless, the effectiveness of a ? rm’s customer retention and cross-selling efforts will certainly be jointly in? uenced by the organization’s capabilities and systems. A few studies have investigated how service organizations can expand their relationships with customers by increasing usage or cross-buying of additional services (e. . , Bolton and Lemon, 1999; Kamakura et al. , 2002; Kamakura, Ramaswami, and Srivastava, 1991; von Wangenheim, 2004; Verhoef, Franses, and Hoekstra, 2001). They typically show that experiences with currently owned products (goods or services) are an important predictor of cross-buying. Customer Divestment Although organizations may have customers who are unpro? table to serve (“free riders”), ? ring customers or refusing to serve them is seldom necessary. Instead, organizations can offer a less attractive value proposition to some segments (e. g. by raising prices or offering lower product quality). In addition, marketing campaigns can be designed to attract pro? table customers and be unappealing to less desirable customers. Another option is to ? nd a way to make the latter group pro? table by changing the ? rm’s business model. For example, IBM wanted to focus on Fortune 1000 companies, but could not ignore less pro? table relationships with small business. Hence, they developed a dealer network that could serve the medium and small businesses in a pro? table way. Challenges Many ? ms use the predicted value of the customer asset (also known as customer lifetime value or CLV) to allocate resources to customer or customer segments, thus accurate calculations are important. CLV predictions should be based on forecasts of revenue sources and costs to serve—based on a particular set of marketing actions and an environmental scenario—where multiple forecasts are possible. Dynamic models to forecast the sources of CLV are required for four reasons. First, CLV is often considered a ? xed value, when it is actually in? uenced by and in? uences marketing strategy (Berger et al. 2002). For example, certain service attributes or marketing variables—such as price or quality—may become more (or less) important to customers as the duration of the relationship lengthens (Boulding et al. ,1993; Mittal, Katrichis, and Kumar, 2001; Mittal, Kumar, and Tsiros, 1999). Consequently, dynamic models are required to re? ect the evolution of customer 18 RUTH N. BOLTON AND CRINA O. TARASI preferences and behaviors over time—so that the path-dependent nature of organizational decisions is explicitly recognized (Rust and Chung, forthcoming; Bolton, forthcoming).

There are established streams of research that model customer acquisition and retention, but there are fewer dynamic models that describe how relationships are expanded by stimulating usage, cross-buying, and word-of-mouth (WOM)—and how these sources affect CLV. Furthermore, customer behaviors are not (typically) considered to be jointly determined within a system of equations. For example, Hogan, Lemon, and Libai (2003, 2004) assess the impact of customer loss due to WOM on product adoption and examine the underestimated effectiveness of advertising due to failure to account for WOM.

Subsequently, von Wangenheim and Bayon (forthcoming) propose a model for including the effect of customer referrals CLV calculations. We believe that much more work is required to build comprehensive, dynamic models of the multiple sources of CLV to produce accurate estimates of CLV, especially in light of the in? uence of socialization and networks on future behavior (see Hakanson and Snehota, 1995). Second, forecasts of sources of CLV will depend on competitors’ activities—and these activities will change over time. Current CRM models devote little attention to competitors and their in? ence on a customer’s relationship with the target ? rm (for a notable exception see Rust, Lemon, and Zeithaml, 2004). Failure to account for competitive effects in a dynamic manner will impair the accuracy of estimating the impact of the marketing actions (Rust et al. , 2004). Third, it is necessary to forecast the implications of marketing actions for the long and intermediate term, as opposed to the short term (Lewis, 2005; Reinartz, Thomas, and Kumar, 2005; Rust and Verhoef, forthcoming). For example, Dekimpe and Hanssens (1995) estimate the long-term effect of marketing activity (speci? ally, media spending) on sales using persistence modeling based on time-series observations. The long-term advertising effect is a combination of consumer response, competitive reaction, and ? rm decision rules effects. The study shows that an advertising medium with lower short-term impact can have a higher long-term effect. Thus, their example demonstrates that traditional approaches can underestimate the long-term effectiveness of marketing expenditures. In subsequent work, they also show that the strategic context is a major determinant of marketing effectiveness and long-term pro? ability (Dekimpe and Hanssens, 1999). Fourth, it is interesting to observe, that—from a customer portfolio management perspective—the goal of CRM is to invest in customer relationships to maximize value to the customer and (aggregate) value for the ? rm. Maximizing the duration of a speci? c customer–? rm relationship or the CLV of an individual customer may not be appropriate. This issue arises whenever the ? rm makes decisions about which customers to acquire, retain, or divest—as well as how to create a portfolio of customers with desirable risk/return characteristics.

In other words, decisions about individual customers cannot be made without considering the optimal characteristics of the entire customer portfolio. Allocating Resources Within and Across Functions, Channels, and Organizational Units Berger and colleagues (2002, p. 51) recommend that “? rms should manage their customers like they manage their assets: by making pro? table investments in value-producing areas. ” Marketers have been especially interested in methods for allocating resources between customer acquisition and retention to maximize return on investment.

Unfortunately, many CLV calculations have been characterized as “undervaluing long term customers and over-evaluating prospects” (Hogan et al. , 2002), which can lead to misallocation of resources. In mature markets, customer retention is cheaper and easier and has more impact than customer MANAGING CUSTOMER RELATIONSHIPS 19 acquisition (Berger and Nasr, 1998; Gupta and Lehmann, 2005; Gupta, Lehmann, and Stuart, 2004; Jain and Singh, 2002; Reinartz, Thomas, and Kumar, 2005), yet overbidding on the future can shift the attention from retention to acquisition.

Customer acquisition is vital in a growing market because it assures the future growth of the company; yet, in a mature market, retaining customers most often offers the best return on investment. The problem of ?nding the equilibrium between investing in acquisition versus in retention is exacerbated by the fact that even though customer acquisition and retention are not independent processes, data limitations have frequently led marketers to treat them as such. 5 Thomas (2001) ? nds that naive predictions can lead to overinvestment in certain customers (e. g. due to incorrectly estimating the impact of add-on selling). The adoption of a long-term perspective implies maximization of neither acquisition rate nor relationship duration, but maximization of the pro? tability of the relationship over time (Reinartz, Thomas, and Kumar, 2005). Strategic models have emerged to help ? rms allocate resources across diverse organizational actions that in? uence customer equity. For example, Rust and colleagues (2004) develop a comprehensive strategic model that links strategic investments (e. g. , in quality, advertising, loyalty programs, corporate citizenship) to customer equity de? ed as the sum of current and future customer lifetime values. They account for competition (via switching probabilities) and customer heterogeneity. Their comprehensive model represents an important step toward understanding the complex effect of strategic changes. However, most research has focused (more narrowly) on resource allocation within speci? c functional areas, including employee selection and training, service quality, customer management effort, multiple channels, customization at the customer, cohort or segment level, loyalty or rewards programs, and the management of customer contacts and processes.

We brie? y summarize these literature streams below. Employee Selection and Training The “service–pro? t chain” links service operations, employee assessments and customer assessments to ? rm pro? tability (Heskett et al. , 1994). For example, Schlesinger and Heskett (1991) describe a “cycle of failure” that occurs when ? rms minimize employee selection effort and training, so that employees are unable to respond to customers requests, and (consequently) customers become dissatis? ed and do not return—yielding low pro? t margins. A signi? ant stream of research has focused on a single link in the chain: the relationship between employees and customers. For example, Reichheld (1993) recommends that “to build a pro? table base of faithful customers, try loyal employees. ” Subsequently, there have been numerous studies of the relationships among employee performance, satisfaction, organizational citizenship behaviors, service climate, and customer satisfaction (de Jong, Ruyter, and Lemmink, 2004; Donovan, Brown, and Mowen, 2004; Gruen, Summers, and Acito, 2000; Netemeyer et al. 1997; Netemeyer, Maxham, and Pullig, 2005). The service–pro? t chain also provides an integrative framework to guide ? rms’ investments in operations, employee selection and training, and customer management. Researchers have modeled components of the service–pro? t chain in different industry contexts, such as banking (Loveman, 1998; Roth and Jackson, 1995) and retailing (Rucci, Kim, and Quinn, 1998). Notably, Kamakura and colleagues (2002) develop a comprehensive approach to the service–pro? chain, incorporating a strategic model estimated with structural equation modeling and an operational analysis based on data envelopment analysis. They were able to identify ways for bank branches to achieve superior pro? tability. Interestingly, they discovered that bank branches must be operationally ef? cient (in terms of deploying employees and technology) and must achieve high customer retention to be maximally pro? table. 20 RUTH N. BOLTON AND CRINA O. TARASI Service Quality The marketing literature has linked service quality to pro? tability in six ways: as a mediator of key service attributes (e. . , responsiveness), through direct effects of service quality on pro? tability, offensive effects, defensive effects, links between perceived service quality and purchase intentions, and via customer and segment pro? tability. Zeithaml (1999) provides an excellent summary of this vast literature, so we do not review it in this chapter. In an early paper, Rust, Zahorik, and Keiningham (1995) provide a framework for evaluating service quality improvements. They illustrate its application and show how it is possible to spend too much (or too little) on quality.

Subsequently, Rust, Moorman, and Dickson (2002) consider how ? nancial returns from quality improvements arise from revenue expansion, cost reduction or both. On the basis of their empirical work, they conclude that ? rms that adopt primarily a revenue expansion emphasis perform better than ? rms that adopt a cost reduction emphasis or a combination strategy. Customer Management Effort Bowman and Narayandas (2004) investigate how increasing product quality and the effort dedicated to customer management in? uence customer satisfaction and pro? s. They ? nd that customer delight “pays off,” but there are diminishing returns on customer management efforts. Moreover, the presence of a viable competitor provides a benchmark for comparison, as well as resulting in lower margins and lower share of wallet. A competitor’s customer management effort negatively in? uences customer perceptions of employee performance and responsiveness. However, the focal ? rm’s customer management effort is twice as important (in terms of the magnitude of the effect) as competitors’ actions.

The size of the customer matters in three ways: margins increase with customer size (nonlinear relationship with decreasing returns); the responsiveness of share of wallet variables to satisfaction decreases with customer size; and larger customers are more demanding, and thus have a lower baseline for both satisfaction and performance assessment. Multiple Channels The advent of e-commerce has resulted in a proliferation of businesses that use multiple channels to reach their customers. If there is no “overlap” in customers across channel



Get your custom essay sample

Let us write you a custom essay sample

from Essaylead

Hey! So you need an essay done? We have something that you might like - do you want to check it out?

Check it out