Customer Retention Management in the B2C Electronic Business

long range planning

Long Range Planning 36 (2003) 517–532

www.lrpjournal.com

Customer Retention Management in the B2C Electronic Business Bernd W. Wirtz and Nikolai Lihotzky

Businesses competing in the internet economy are turning their attention and resources towards increasing the retention of their customers and users. This paper examines how customer retention can be achieved and which strategy best fits which business model. The authors based their analysis on theoretical reasoning and interviews with B2C executives. Their findings differentiate the most appropriate strategies for particular business models. For example, building customer trust and convenience are the most appropriate for commerce-based businesses while the offer of free services are better suited to content, context and connection business models. 쎻 c 2003 Elsevier Ltd. All rights reserved.

Introduction The strategic imperative for customer retention In 2000 the 10 most frequently visited Business-to-Consumer (B2C) internet companies spent a conservatively estimated $3 billion on marketing.1 While some of this spending can be attributed to brand building, particularly for newer players, the bulk of the money went into acquiring new customers. Sustainable profitability, however, only results from repeat purchases which in turn can only be achieved through building a loyal customer base.2 Businesses competing in the internet economy are therefore turning their attention and resources towards increasing the retention of their customers and users.3 In May 2000 a Jupiter Media Metrix survey found that 82 per cent of e-commerce executives stated that the top priority for the business over the next few months would be to increase the number of new customers: by mid-2001 an exploratory study showed how the situation evolved with 86 per cent of B2C business executives stating that customer retention took priority over new customer acquisition. To achieve their objective of customer retention, however, companies must understand the 0024-6301/$ - see front matter 쎻 c 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.lrp.2003.08.010

specific environment of the electronic marketplace, where the barriers to customers switching are lower compared with the traditional economy and vendors are more vulnerable to customer defection. These lower switching barriers result primarily from the intensification in market transparency in the electronic business, an effect further increased by the advent of smart price agents, from the absence of physical distance between customers, suppliers and competitors, and from the lack of personal vendor–customer relationships. Should an internet business succeed in retaining its customers, on the other hand, the benefits are great: the costs of acquiring a web customer may be high but the relative saving potential from retaining a customer is high as well. Second, the ease of extending existing services and product lines into new ones enables web business to take advantage of cross-selling opportunities to loyal customers. Third, referrals, another important benefit from loyal customers, occur at a higher rate in the internet economy than in the traditional market because they can be made with much less effort, often with a single mouse click. Some internet businesses have therefore long realised the importance of customer retention. This results in a wide array of customer retention tools already implemented in B2C markets today. However, as we will point out, there is no single approach to customer retention that is applicable across the wide varieties of business models found on the internet today. Rather, we theorise that certain retention tools are more suitable for specific business models than for others. So how, exactly, can customer retention be achieved in a challenging environment where market transparency is great and every competitor is just a mouse click away? And which retention strategy is appropriate for which B2C business model? In academic research, to our knowledge, these questions remain unanswered. The few publications that approach the topic of customer retention in electronic business underline its importance and provide some insights into possible measures, but none takes a business model perspective and none provides empirical results on specific retention measures. To approach the subject of successful customer retention in the B2C electronic business we employed a comprehensive research process consisting of four steps combining theoretical and empirical elements. Starting from a foundation of theory and literature research we observed the 29 most frequently visited B2C websites to assess their customer retention measures. Owing to their large loyal customer base, we consider these companies as best-practice in customer retention. Based on these findings we conducted 15 exploratory expert interviews with B2C electronic business marketing executives to deepen our understanding of customer retention management in this market. From the results of the first three research steps we developed a set of research hypotheses which were then analysed in a confirmatory quantitative empirical study based on a large sample of B2C internet businesses. The article will first discuss the suite of retention tools that was developed from our theoretical and exploratory empirical research. Second, we will introduce the 4-C-internet business model typology which is used to classify the different business models found on the web today. The third part of the article describes the set of criteria which will be used to assess the viability of a retention strategy for a business model. In the fourth part a theoretical alignment of retention strategies and business models based on these criteria is presented and subjected to quantitative empirical verification. The final part of the paper presents conclusions from our findings. The logic of the paper as well as the initial theoretical alignment between retention strategies and business models are shown in Figure 1.

Customer retention strategies While in the electronic business the question of customer retention is particularly important, the environment for it differs somewhat from the offline world. The internet economy presents companies with new threats as well as new and different opportunities.4 Our theoretical analysis and exploratory empirical research consisting of best practice website observation and executive interviews led to the finding of seven particularly relevant retention strategies widely and successfully used by B2C electronic businesses today, which we will review in this section. These are: trust 518

Customer Retention Management

Figure 1. Conceptual research framework

building, community, convenience, free services, individualisation, contractual agreements and technical integration. According to Morgan and Hunt, trust plays a major role in the success of business relationships. High levels of trust lead to relationship commitment and to increased co-operation. They also encourage functional conflict and reduce uncertainty in a relationship.5 Additionally, trust is subject to a so-called echo effect, meaning that trust leads to more trust unless misused, making trust itself a major driver of relationship stability. In the anonymous electronic environment, however, achieving trust in the beginning of a relationship is more difficult than in the traditional business world because the customer has much less information available to assess the trustworthiness of a supplier. In addition to the lack of interpersonal interaction between buyer and seller, a potential customer receives no guidance on assessing the economic viability of a supplier. In the traditional business world a customer can draw conclusions from the location, size and design of a store, whereas in the electronic business such clues are largely missing. Moreover in most online transactions the customer carries the risk of upfront payment, while in the offline world an immediate exchange of payment and goods or services is the norm. Therefore online customers are not only less informed about their suppliers than offline customers, they also carry the fulfilment risk. To resolve this dilemma, electronic businesses have a number of options. In addition to the trust building effect of a strong brand, certification and guarantees by independent third parties, such as the trusted-shops-seal can be employed to increase trust. Other options are flexible return policies, prompt and customer-orientated customer service and a state-of-the-art complaint management.

Trust leads to more trust unless misused, making trust itself a major driver of relationship stability Virtual communities, such as discussion forums, news groups and online chats provide the customer with a platform for mutual exchange.6 Interaction occurs multi-directionally between customers and is usually centred on a specific topic, leading to credible communication. Virtual

Long Range Planning, vol 36

2003

519

communities represent an internet-specific development. Customers develop strong emotional bonds to virtual communities which result from the development of social relationships between the community members. This sense of belonging results in loyalty towards the community. For a company in the B2C electronic business that operates a community, this loyalty towards the community also results in increased loyalty towards the operating company, thereby enhancing customer retention. The internet is well established as an efficient channel for information, communication and transaction. By using the internet, both suppliers and customers can benefit from transaction cost savings. The magnitude of the transaction cost savings on the customer side depends to a large extent on the convenience of an offering. Convenience, in turn, is driven by factors such as fast page loading, clear presentation and simple, intuitive navigation. Amazon.com provides an excellent example of convenience with its ‘One-Click’ ordering function which the company found so powerful that it had it patent-protected. For retention purposes, customers will prefer offers which support transaction cost savings through high levels of convenience. Because the variable reproduction and distribution costs of digital information products tend to zero many electronic business companies offer services which are free of charge.7 Free e-mail accounts, downloads, greeting cards and online games are just a few examples. Their retention effect is based on increased customer satisfaction resulting from receiving a value without a corresponding cost or payment.8 Additionally, by bundling several free services, a company can achieve multiple customer bonding—retaining a customer through several separate relationships.9 Yahoo! is a very good showcase of how integrated free services work together to help increase customer loyalty.

The added value that individualisation offers leads to increased customer satisfaction while creating an effective switching barrier. One of the most frequently mentioned customer retention measures in the electronic business is individualisation of products and services. There are a number of reasons why individualisation is promising. First, individualisation caters to the overall trend for individual products and services.10 Second, comparison of differentiated personalised offerings is difficult if not elusive. Thus market transparency is reduced and search costs in the case of switching are increased. Finally, and most importantly, the added value that individualisation offers leads to increased customer satisfaction while creating an effective switching barrier. The added value can result from reduced transaction costs, particularly search and communication costs, or can take the form of an improved core value due to an offering customised to the exact need of the customer. Two major forms of individualisation exist: personalisation, where the customer initiates and manages individualisation information; and tailoring, where the supplier takes on that role.11 Personalisation usually requires the customer to expend time and effort. According to transaction cost theory, the time and effort required for providing the individualisation information represents a specific investment into a relationship. Depending on the type of individualisation, this specific investment takes an ex ante or an ex post form. Ex ante specific investments occur when customers fill out customer profiles or individualise a personal home page, such as myYahoo! The other form of specific investments, ex post, occurs over time during the ongoing relationship. An example might be the creation of an online address book that the customer fills with data over time. Another example is the supplier tailoring its offer by using a customer profile based on past transactions. In this case, the customer does not knowingly invest into the relationship, rather the benefit of appropriate tailored offerings is generated as a side effect of repeated transactions. Nevertheless this benefit only exists within the context of the ongoing relationship. Contractual agreements are another form of customer retention. Usually they take the form of subscription-like contracts with minimum duration or quantities. The retention effect results from the customer’s expected costs in case of breach of contract or from upfront payment for services. 520

Customer Retention Management

A contractual agreement represents a specific investment in a relationship from the customer’s perspective, which will only yield its benefits within the context of this relationship. Since the contractual agreement is entered consciously and upfront the investment is of ex ante-specificity. The Wall Street-Journal-Online, for instance, offers annual subscriptions which include an automatic renewal clause. Finally, customer retention can be achieved through technical integration of an interface to a supplier’s offering into the customer’s work station. A frequently found method is the set-up of a specific web page as the starting page in the customer’s internet browser software. Each use of the browser then automatically guides the customer to the corresponding offering. Alternatively (or additionally) browser plug-ins, which can take the form of an additional tool bar or menu item, can be used to extend the browser’s functionality. The ultimate form of technical integration is the provision of proprietary access software required to use a specific supplier’s offering. AOL provides a good example of this method. The time and effort required by the customer to remove the integration elements makes all described forms of technical integration a switching barrier. Additionally, many internet customers do not have the necessary skills required to modify browser and computer settings and, thus, leave the settings as they are. The seven identified retention strategies are used by some of the most successful companies on the internet resulting in superior customer retention for them, which is evidenced by the fact that they are among the most frequently visited offerings on the internet and by statements from their executives in our in-depth interviews. While many internet companies vanished from the landscape during the shake-out in 2000/2001 because of their lack of attention to customer retention, the observed best-practice companies survived owing, in part, to their strong, loyal customer base. From this exploratory empirical evidence and from the theoretical perspective just discussed, we conclude that the use of these seven retention strategies successfully contributes to enhancing customer retention. At this point, however, we would like to re-emphasise our initial premise that not all retention strategies are suitable for all of the business models found on the internet. To approach the important question of suitability we now turn to a discussion of different business model types.

Not all retention strategies are suitable for all of the business models found on the internet. Today’s internet business models The evolution of the internet brought about the development of a large number of new and innovative business and revenue models in both the business to business (B2B) and B2C markets. In order to develop insights about specific business models, it is helpful to apply a basic classification scheme. We propose the use of the 4C-Net-Business-Model framework because of its straightforward yet comprehensive approach. The 4C-Net-Business-Model framework differentiates between the four basic internet business model types: Content, Commerce, Context, and Connection.12 Firms specialising in the content-orientated business model generate revenues through the collection, selection, compilation, distribution, and/or presentation of online content. Its value proposition to the end-user is to provide convenient online access to information, education or entertainment content in a visually attractive form. The commerce-orientated business model pertains to the arrangement (i.e. bringing supply and demand together), negotiation and payment and delivery aspects of trade transactions using electronic media. Its objective is the support or even substitution of traditional transaction phases through electronic internet processes. Context business models relate to the aggregation and structuring of the information existing on the internet, thus increasing market transparency.13 Their objective is to reduce complexity and provide orientation for the user through navigation services. Businesses with a connection business

Long Range Planning, vol 36

2003

521

model provide the physical and virtual network infrastructure which enables users’ participation in networks, either on a physical level (interconnection) or a virtual level (intraconnection). While it is true that many internet ventures employ several business models at once (in hybrid and integrated versions), expanding their scope as they mature, usually the focus lies in one of those four areas. The four basic internet business models, their objectives, their revenue model(s) as well as their further breakdown into sub-models (variants) along with examples are presented in the 4C Net Business Model Framework shown in Figure 2.14 Clearly, the internet business models differ significantly in their value propositions and, thus, their revenue sources. Considering these differences, it seems not without reason to assume that suitable customer retention strategies may also differ by business model. Matching customer retention strategies and internet business models To approach the task of aligning retention strategies to the business model types we applied a framework structured around a set of criteria along which we assessed the business models as well as the retention strategies. Matching high levels on one or several criteria for both retention strategy and business model were taken as an indication of an initial theoretical fit between the two. We will now briefly introduce the criteria used in the framework, the assessment of the business models along these criteria and then discuss each retention strategy with the corresponding business models in turn. Business models can be described and differentiated along a long list of criteria. Timmers, for instance, employs three basic categories of criteria: first, the revenue sources of a business model; second, the business architecture for product, service and information flows including a description of the market participants; and third, an account of the business model’s benefits for the market participants.15 While the thorough specification of a business model along these items is clearly outside the scope necessary for analysing retention management, some elements within the three categories appear well suited for this analysis. The revenue model is relevant because it defines how a business model is financed and thus drives which customer behaviour is desired to maximise long-term profit. Many customer retention measures benefit from economies of scale and network effects, therefore the question of usage intensity as one element of the business architecture becomes important. And finally, certain customer retention measures function through providing the customer with a net benefit, from which follows that a business model’s value proposition or core offering is a relevant assessment criterion. In this context it can be safely assumed that cus-

Figure 2. 4C Net business model framework

522

Customer Retention Management

tomer retention measures which provide complementary benefits to those of the core offering will be particularly effective. The relevance of revenue model, usage intensity and core offering for the assessment of customer retention strategies is also immediately apparent, as all three elements are directly related to the customer or the customer relationship. Other elements regularly found in business model descriptions, such as the procurement model or the production model have no direct relevance for the customer. Revenues, however, ultimately must be earned from the customer or at least with customer participation. The customer determines the usage intensity of an offering and the offering must be in line with demand.16 The important influence of these three elements on decisions concerning customer retention management was also confirmed in the expert interviews. We will now turn to a brief description of the three criteria and their individual categories. Two dimensions are useful for differentiating and structuring revenue models, the directness of the revenue stream, and the transaction dependency of the revenue stream. Direct revenues are obtained from the end-customer directly whereas indirect revenues flow between a supplier and an intermediary and are therefore only indirectly paid for by the end-customer. The transaction dimension determines whether an end-user or customer has to become active on a website, for instance by clicking on a link or by placing an order, in order for revenues to flow or whether the revenue stream is maintained without such action. Table 1 provides an overview of the revenue model classification including examples of typical revenue types from each category. Usage intensity in turn is a function of the three metrics customer volume, usage frequency and duration of each usage. Business model specific characteristics can be determined for each of the three measures. Certain business model types are generally frequented by more customers due to their specific core offering. Every user of the internet, for instance, must frequent a connection business model as it represents a basic condition of internet access. The question which other types of business models users visit during their internet use, however, remains open. At the same time, the average number of customers per supplier is determined by the number of suppliers in the relevant market. While the average customer volume is thus determined at the supply side of the market, the average usage frequency and duration are largely based on the usage habits of each individual customer. These usage habits tend to be specific to each business model type. Search engines (as a type of context business model) are visited frequently but only for very short durations, because they represent a springboard for destination sites and the offerings of other suppliers. Visits to destination sites are comparatively less frequent but last longer. Usage habits are also relevant for the analysis because both usage frequency and duration tend to correlate with the level of indirect transaction-independent revenues. In addition to the measure of unique visitors, page visits and length of stay are increasingly becoming important measures of reach used for the determination of banner pricing.17 As far as customer retention management is concerned, these measures represent different dimensions of the goal of customer retention.18 In electronic B2C markets a wide variety of service offerings exists. In order to structure internet business models along the line of their value proposition to the customer, it is useful to define different categories of core offerings. Commonly five such categories are used, four of which are

Table 1. Internet revenue models

Transaction-dependent

Transaction-independent

Long Range Planning, vol 36

Direct revenues

Indirect revenues

Sales revenues Connection fees Download fees Usage fees Activation fees Subscription fees

Commissions

2003

(Banner) Advertising Data mining revenues Sponsorship

523

pertinent to customer retention. They are: information, entertainment, communication and transaction. A possible fifth category, education, currently represents a rare special case which is not relevant with regard to customer retention. An information offering fills the customer’s need for up-to-date relevant information that can be retrieved when and where the customer wants. The offering of entertainment is closely related to that of information as it also offers content, but in this case the content diverts and amuses the customers. A communication offering enables interaction between two or more participants. This offering also includes physical connection. Communication offerings are therefore concerned with the provision and use of network-based electronic communication platforms, either free of charge or for a fee. The fourth category of service offerings, transaction, is concerned with the selling of products and services over electronic networks. Frequently the use of this service offering results in transaction cost savings, making the service offering more attractive in turn.19 As mentioned above, the first step of developing a preliminary theoretical alignment of customer retention strategies and internet business models consists of assessing each business model against the just presented alignment criteria. Content business models The vast majority of content business models today are financed through indirect transactionindependent revenues, mainly banner advertising and sponsorship. However, we begin to see the content businesses pursuing direct revenue models. The Wall Street Journal Online, for instance, offers premium content for a paid subscription. This represents an example of a transactionindependent direct revenue model. Sciencedirect.com is a direct transaction-dependent revenue model. At Sciencedirect, non-subscribers can download individual articles against credit card payment. As evidenced by the introduction of pay content on a wide scale in early 2002 by Europe’s leading online service, T-Online, there is a trend for companies pursuing a content business model to transition their free, indirectly financed content services to a direct revenue model. Their success will in part depend upon the availability and acceptance of secure and convenient micro payment methods. For the time being, for content business models indirect, transaction-independent revenue models dominate with direct revenue models growing. With respect to usage intensity, content business models score average in both frequency and duration. Their customer volumes tend to be lower than average, due to a large number of suppliers, as shown by a 2002 empirical study by Wirtz and Becker.20 In terms of their core offerings content business models have a clear strength in information and entertainment. Commerce business models With respect to commerce business models, the majority are financed by direct, transactiondependent revenues. While for most commerce businesses sales revenues are the primary source of income, for a smaller subset indirect, transaction-dependent revenues are also relevant. This is the case for businesses, such as online auctioneer eBay, where revenues mainly consist of commissions. Due to the large number of companies following a commerce business model, such suppliers tend towards a less than average-sized customer base. They are also neither visited very frequently nor for very long durations. Their core offering, clearly, is transaction. Context business models Context business models such as search engines are almost exclusively financed through indirect, transaction-independent revenues, mainly banner advertising and sponsorship. Another potential source of revenue is the sale of qualified user data.21 This opportunity exists because context business models have a very large customer base. Because of their important navigation function, context business models are visited very frequently but only for a short duration. The core offering represents a special form of information—the presentation of locations that offer the content that the customer is seeking. 524

Customer Retention Management

Connection business models The core offering of connection business models is primarily communication-orientated. While interconnection business models allow for the physical communication between an end-user and the internet, intraconnection models offer communication services within the internet, in a way that a customer can enter into communication with other participants. Another typical function is the organisation of discussion forums and online chat groups. Thus, particularly intraconnection business models also enable an entertainment function. Finally the most frequently used internet function of all, e-mail, represents an important offering of the intraconnection business model. Therefore connection business models benefit from high usage frequencies and long usage durations. Customer numbers are average. In order to assess the revenue models of the connection business models a differentiation between intra- and interconnection business models is necessary. Intraconnection business models are primarily financed through indirect, transaction-independent revenue sources. Interconnection business models, on the other hand, raise revenues directly helped by the fact that their service offering represents a necessity of internet use. Both transactionindependent revenues, such as set-up fees, subscriptions and flat rate charges, and transactiondependent revenue forms, such as connection and online fees as well as combinations of transaction-dependent and transaction-independent revenues are commonly employed. The assessment of the four business models along the alignment criteria is summarised in Figure 3. The second analysis step centres around assessing which of the alignment criteria are relevant for each of the seven retention strategies and then matching the corresponding business models that score high in those criteria. An initial fit between a retention strategy and one or several business model(s) is then subjected to empirical testing using univariate analysis of variance (ANOVA). In this section we therefore present our hypotheses concerning viable matches along with the results from empirical testing using the findings from our online survey. For the survey we invited 1500 e-business companies via email to fill in an online questionnaire on customer retention. The questionnaire included questions about the business model type as well as questions on the use of the seven retention strategies. In the fiercely competitive business that the online business represents, our empirical work is based on the premise that only meaningful business model retention strategy combinations are successful in the long-term. As our empirical work was performed in the winter of 2001, after the major shake-out in the internet world, we consider our empirical results proscriptive in the sense that they represent successful combinations of retention

Figure 3. Business model assessment

Long Range Planning, vol 36

2003

525

strategies and business models. For a more detailed discussion of the empirical base and the analysis methods used, please refer to Appendix A. Trust building In order for trust to be relevant in a customer’s decision for or against a given supplier, the customer must be vulnerable to trust abuse. Thus, trust building should be particularly relevant to business models with a high perceived risk from the customer’s perspective. This perceived risk is closely correlated with a business model’s revenue model. A customer’s perceived risk is higher for transactions which are subject to direct transaction-dependent revenues, where the customer usually pays up-front, than for transactions where up-front payment is not required. If a business succeeds in increasing its customers’ trust, the customers’ perceived risk within the business relationship is reduced. We have argued trust building to be well suited for business models employing direct revenue models. Direct revenue models are most frequently encountered in commerce and connection business models. The risk perception in connection business models can be assumed to be lower than in commerce models because of lower overall advance payments. In addition, in commerce business models the period between payment and reception of goods or services is longer than in connection models. Therefore we propose the focus on trust building to be in commerce business models. Empirically, in the ANOVA test, the relevance group (consisting of commerce business models only) scored an average value of 0.165. The comparison group consisting of the remaining three business models achieved a score of ⫺0.256. The corresponding p-value was 0.026 and, thus, significant at the 5 per cent level. Our theoretical perspective is therefore supported by empirical evidence: based on the evidence and our reasoning we suggest that trust building is appropriate for customer retention in commerce business models more so than in other business models.

If a business succeeds in increasing its customers’ trust, the customers’ perceived risk within the business relationship is reduced. Community Since communities exist in a wide variety of forms, they can offer communication, entertainment, and to a lesser extent information. Communities should be well suited for customer retention if the core offering of the operating business and the core offering of the community are in line. Studies have shown the possibility for interaction in communities to result, on average, in a twofold increase in usage frequency of an offering and tripled usage duration. This would make communities well suited for business models that are financed through indirect, transaction-independent revenue models. The use of communities for customer retention appears appropriate if a business model is financed by an indirect revenue model because communities support usage duration and frequency and the level of revenue in an indirect revenue model frequently is dependent on usage duration and frequency. These criteria are best matched by connection business models but to a certain extent also by content and context business models. Content business models are mostly financed indirectly and pursue information and entertainment offerings. Context business models are also indirectly financed but only fulfil an information function. Therefore community is a viable retention measure for these three business models. Again, the empirical findings support this hypothesis: the relevance group with connection, content, and context business models achieves a score of 0.344 for the community strategy. By contrast, the remaining business model, commerce, achieves ⫺0.173. The results are significant at the 1 per cent level as indicated by a p-value of 0.008. Both the empirical evidence and our theoretical logic lead to the recommendation that communities should be used for customer retention by connection, content and context business models rather than by commerce business models. 526

Customer Retention Management

Convenience Convenience is a promising customer retention measure if an offering is transaction orientated because one of the key benefits of transaction models on the web is transaction cost reduction. Convenience further lowers the costs in a transaction and therefore supports the transaction model’s key benefit. Another relevant criterion is a direct revenue model because customers will be more transaction cost sensitive due to the higher absolute level of visible costs. Both criteria, transaction orientation and direct revenue, are only fulfilled for the commerce business model, which leads to the conclusion that convenience might be a suitable retention measure for this business model. Referring to the results of the empirical study we observe that commerce scored 0.203 on convenience while the comparison group has a negative score of ⫺0.276. As the corresponding p-value of 0.008 indicates significance for this result at the 1 per cent level, we find this hypothesis clearly supported and conclude that Convenience is appropriate for customer retention in commerce business models rather than for others. Free services Many of the free offerings are subject to network effects (e.g. e-mail, games) and seem well suited for business models with high customer volume. Given high customer volume, the relatively high first copy costs can be distributed over a large number of relationships resulting in decreasing marginal costs.22 Additionally, free services offering communication (e-mail, greeting cards, SMS) should be appropriate as customer retention measures in business models whose core offering is also communication-related. Finally, some free services such as online games produce their value by offering entertainment which makes them a good complement to business models that also offer entertainment. Free services therefore appear to be appropriate for all business models with the exception of commerce. The empirical findings point to a score of 0.299 on free services for the relevance group consisting of content, context and connection business models and to a negative ⫺0.117 for the commerce model. The p-value of 0.035 indicates a significant result at the 5 per cent level. The hypothesis is therefore supported and our conclusion is: connection, content and context business models can enhance their customer retention by offering free services better than commerce business models can.

Free services appear to be appropriate for all business models with the exception of commerce. Individualisation Individualisation should be relevant to those business models in which the customer benefits most from individualisation. Communication core offerings appear well suited because communication is frequently individually-orientated. This perspective points to connection business models. More importantly, however, customers will only make the specific investments in terms of time and effort to personalise an offering if they plan to use the offering frequently because otherwise the investment will not pay off for them.23 Thus, high usage frequency of a business model is an important criterion for the suitability of individualisation as a retention strategy. Usage frequency is high in context and connection business models. Our initial thinking was therefore that individualisation should be primarily used by context and connection business models. Empirically, however, the proposed relevance group of context and connection business models scores 0.309 on individualisation, while the comparison group receives a score of ⫺0.059 with a p-value of 0.122 indicating a non-significant result at our 5 per cent threshold; the hypothesis cannot be maintained. Given the high overall effectiveness of individualisation supported by both expert opinion and theoretical logic, we come to the conclusion that individualisation is an appropriate strategy to enhance customer retention across all business models.

Long Range Planning, vol 36

2003

527

Contractual agreements Clearly, contractual agreements can only be an appropriate retention strategy for business models based on direct revenue models. Contracts based on a fixed minimum duration are well suited for transaction-independent direct revenue models while minimum quantity-type contracts are appropriate for transaction-dependent direct revenue models. Similar to individualisation, a contractual agreement with a supplier represents a specific investment by the customer, which he or she will only enter into if frequent usage in the future is expected, making usage frequency another relevant criterion. Based on these criteria, contractual agreements are apt for the business models commerce, connection, and, to a lesser extent, content. With regard to commerce models it must be said, however, that minimum quantity and minimum duration contracts are rather unusual. In content business models we observe a growing trend towards suppliers trying to establish direct revenue models. The interconnection business model benefits from the fact that fees for telecommunication services are long established in the consumers’ mind.24 We therefore conclude an emphasis on contractual agreements for customer retention to be in the content and connection business models. The empirical values were 0.412 and ⫺0.122 for the relevance group (content and connection) and comparison group, respectively. The corresponding p-value is 0.02. The result is therefore significant at the 5 per cent level. Based on our theoretical thinking and the empirical evidence we therefore conclude that contractual agreements are appropriate for customer retention in content and connection business models. Technical integration Due to the relatively high investments needed for the development of the necessary integration software elements, technical integration is only appropriate for business models with high customer volume. Customers, on the other hand, will only accept technical integration if it reduces their effort in using a service, hence, if they expect frequent usage. Both measures are particularly high in context business models. Even though they only have average customer volumes, connection business models also seem apt for technical integration due to their high usage frequency. The empirical findings support this hypothesis. Values for technical integration are 0.659 for the relevance group (context and connection business models) and ⫺0.120 for the comparison group (content and commerce models). The p-value of 0.001 indicates a highly significant result at the 1 per cent level. According to theoretical analysis and the empirical results, context and connection business models should pursue technical integration to enhance customer retention more so than the other two business models.

Customers will only accept technical integration if it reduces their effort in using a service. Summary of findings and conclusions Customer retention in the transparent electronic business market is particularly important.25 Due to low information costs among other factors, the barriers to customers switching dwindle. At the same time, suppliers can use a variety of new measures to counter this and create new switching barriers. So far, both management and scientific literature have covered electronic business in the B2C and B2B markets in a quite general way including, but not limited to, the topic of customer retention. This article takes a more differentiated perspective and analyses the use of customer retention measures depending on the pursuit of one of four business models. The empirical analysis shows that companies in the B2C electronic business use the instruments of customer retention adequately with respect to their primary business model. Most combinations of customer retention measure and business model, that seem appropriate from a conceptual perspective, occur with above-average frequency in the empirical data. The article provides businesses with a first orien528

Customer Retention Management

tation framework to assess the suitability of a set of customer retention measures in accordance with a given business model. The empirically supported recommendations presented in this article should be viewed as initial research findings. Our empirical research is based on the premise that the companies surveyed are successful in customer retention. While there is indication that this is the case (they all survived the internet shake-out of 2000/2001), more detailed analysis, particularly with regards to the actual effectiveness of each retention strategy is desirable and necessary to confirm our initial findings. Bearing this limitation in mind, the findings should still be viewed as noteworthy as theoretical reasoning and empirical findings tally well, indicating a certain level of validity. With the notable exception of individualisation, both our theoretical reasoning and the empirical evidence obtained through surveying successful internet businesses point to the fact that the use of retention strategies should be viewed in a differentiated way based on the primary business model an internet company employs. Trust building is most appropriate for commerce business models; Communities appear particularly successful for content, context and connection business models. Convenience is a promising retention strategy for commerce business models. Free services should be employed by content, context and connection business models. Individualisation stands out among the seven discussed retention strategies as it appears to be universally applicable across all business models. We attribute this to the fact that individualisation is considered a particularly powerful retention strategy as it both locks the customer into specific investments while increasing customer satisfaction through relevant offerings. Customer retention through contractual agreements is suitable for content and connection business models. Finally, technical integration is best used by context and connection business models. Figure 4 summarises the alignment between customer retention strategies and business models, as evidenced by the empirical findings along with the key rationale for each match. Even given the limitations discussed above three remarkable conclusions remain. First, our research has shown us that internet companies have realised the importance of customer retention and have initiated measures to counter the low switching barriers. Second, there are very strong indications that internet firms realise that different business models call for different retention strategies, which was our initial premise. Third, it appears that internet firms are doing the right

Figure 4. Alignment of retention strategies and business models

Long Range Planning, vol 36

2003

529

things to increase customer retention as indicated by the strong match between theory and actual practice. For the academic reader we have turned the imporant area of research of customer retention and customer loyalty on to a new playing field, the internet, where it appears to be particularly important. However, the surface is barely scratched. Important issues remain to be answered. Among them is the important question of the actual effectiveness in increasing customer retention through appropriate strategies. Second, it appears to be worthwhile also to consider the customer perspective as we have taken a pure supplier-based approach. Finally, our initial research is limited to the business-to-consumer markets, leaving the potentially large research field of business-tobusiness untapped.

Acknowledgements We would like to thank two anonymous LRP reviewers and Professor Charles Baden-Fuller for their appreciated remarks.

Appendix A. Empirical approach and findings From a database containing more than 2500 B2C electronic business companies, we extracted a sample of 1500. They were contacted by e-mail and asked to fill out an online questionnaire. The questionnaire consisted of two sections. The first section inquired about overall data of the responding company and individual. To ensure the answering individuals had sufficient insight into their companies’ strategies, we asked them to answer a question concerning their role in the company. All of the respondents included in the analysis were either the company’s principal head of marketing/sales, head of business development, or in another top management position. It is therefore safe to assume the answering individuals’ competence in responding to our questions.

Table 2. ANOVA results between relevance and comparison groups Retention measure

Trust building (H1)

Business models in:

Average values in

ANOVA

Relevance group Comparison group

Relevance group Comparison group

p-value

Content Context 0,165 Connection Community (H2) Content Context Commerce 0,344 Connection Convenience (H3) Commerce Content Context 0,203 Connection Free Services (H4) Content Context Commerce 0,299 Connection Individualisation (H5) Context Content 0,309 Connection Commerce Contractual agreements Content Commerce 0,412 (H6) Connection Context Technical integration Context Content 0,659 (H7) Connection Commerce ∗

Commerce

significant at the 5% level;

530

∗∗

⫺0.256

0.026∗

⫺0.173

0.008∗∗

⫺0.276

0.008∗∗

⫺0.117

0.035∗

⫺0.059

0.122

⫺0.122

0.02∗

⫺0.120

0.001∗∗

significant at the 1% level.

Customer Retention Management

In the second section questions concerning the use of the seven identified retention strategies were covered. For this purpose several indicators for each retention strategy were included. Some 143 questionnaires were returned, resulting in a gross response rate of 9.53 per cent which is a good value for online research. After elimination of responses with missing data 122 sets of data were used in the empirical analysis. To control for non-response bias, as conventional in marketing science, the Armstrong/Overton test was used. T-tests revealed no significant differences between early and late respondents. There is therefore no indication of non-response bias. To test our alignment hypotheses, we used univariate analysis of variance (ANOVA). For each hypothesis a relevance group consisting of the business model(s) for which appropriateness of the retention measure was hypothesised and a comparison group consisting of the remaining business model(s) were built. Based on an exploratory factor analysis factor values for the usage intensity of the seven customer retention measures were calculated. In the online questionnaire several indicators were used to measure the customer retention activities. They were tested for reliability and validity using Cronbach’s Alpha and an exploratory factor analysis. The commonly used quality requirements were met. Table 2 contains the resulting average factor values for the usage intensities for both, the relevance and comparison groups and the p-values of the corresponding analysis of variance between the two groups. The average factor value across all business models is standardised to zero for each retention measure. Negative average values are therefore indicative of a less-than-average use and positive values indicate an above average use of a retention measure.

References 1. Estimate based on 10 most frequently visited web properties as published by Media Metrix. Marketing expenditures taken from annual reports where available. 2. Reichheld and Schefter point out for electronic business that ‘you cannot generate superior long-term profits unless you achieve superior customer loyalty.’ F. Reichheld and P. Schefter, E-Loyalty, Harvard Business Review 78(4), 105–113 (2000). 3. In the context of the internet economy we distinguish between users and customers: users do not generate direct revenues for a provider whereas customers pay directly for the services they use. 4. J. Hagel III and J. F. Rayport, The coming battle for customer information, Harvard Business Review 75(1), 53–65 (1997). 5. R. Morgan and S. D. Hunt, The commitment–trust theory of relationship marketing, Journal of Marketing 58(3), 20–38 (1994). 6. R. Williams and J. Cothrel, Four smart ways to run online communities, Sloan Management Review 41(Summer), 81–91 (2000). 7. D. Yoffie and M. Cusumano, Judo strategy: the competitive dynamics of internet time, Harvard Business Review 77(1), 70–81 (1999). 8. E. Anderson, C. Fornell and D. Lehmann, Customer satisfaction, market share and profitability: findings from Sweden, Journal of Marketing 58(3), 53–66 (1994). 9. B. W. Wirtz, Reconfiguration of value chains in converging media and communications markets, Long Range Planning 34(4), 489–506 (2001). 10. S. Ghosh, Making business sense of the internet, Harvard Business Review 76(2), 127–135 (1998); K. T. Rosen and A. L. Howard, E-Retail: gold rush or fool’s gold?, California Management Review 42(3), 72– 100 (2000). 11. J. F. Rayport and B. J. Jaworski, e-Commerce, McGraw-Hill, Boston (2001). 12. B. W. Wirtz and A. Kleineicken, Gescha¨ftsmodelltypologien im Internet, Wirtschaftswissenschaftliches Studium (WiST) 29(11), 628–635 (2000). 13. S. Kaplan and M. Sawhney, E-Hubs: the new B2B marketplaces, Harvard Business Review 78(3), 97– 103 (2000). 14. B. W. Wirtz, Electronic Business, 2nd edition, Gabler, Wiesbaden (2001). 15. P. Timmers, Electronic Commerce—Strategies and Models for Business-to-Business Trading, Wiley, Chichester (1999). 16. J. F. Rayport and B. W. Wirtz, Vergessen wir das ‘E’ und kehren zum Business zuru¨ ck, um erfolgreich zu sein, Frankfurter Allgemeine Zeitung 69, 30 (2001). 17. A. Afuah and C. Tucci, Internet Business Models and Strategies, McGraw-Hill, Boston (2001).

Long Range Planning, vol 36

2003

531

18. B. W. Wirtz and N. Lihotzky, Interneto¨ konomie, Kundenbindung und Portalstrategien, Die Betriebswirtschaft (DBW) 61(3), 285–305 (2001). 19. E. Brynjolfsson and M. D. Smith, Frictionless commerce? A comparision of Internet and conventional retailers, Management Science 46(4), 563–585 (2000). 20. B. W. Wirtz and D. Becker, Erfolgsrelevanz und Entwicklungsperspektiven von Gescha¨ftsmodellvarianten im Electronic Business, Wirtschaftswissenschaftliches Studium (WiSt) 31(3), 142–148 (2002). 21. J. Calkins, M. Farello and C. Smith Shi, From retailing to e-tailing, The McKinsey Quarterly 36(1), 140– 147 (2000). 22. C. Shapiro and H. R. Varian, Information Rules: A Strategic Guide to the Network Economy, Harvard Business School Press, Boston (1999). 23. O. Williamson, The Economic Institutions of Capitalism, Free Press, New York (1985). 24. The connection business model’s two sub-models intra-connection and inter-connection differ significantly from each other with regards to their revenue models. However, they are frequently pursued together as evidenced by the examples AOL, Tiscali, and T-Online, which makes it useful to analyse them collectively. 25. M. Larsson and D. Lundberg, The Transparent Market, St. Martin’s Press, New York (1998).

Biographies Prof. Bernd W. Wirtz holds the Deutsche Bank Chair for Strategic Management, University of Witten/Herdecke, Germany and is director of the Euro Lab of Electronic Commerce & Internet Economics (ecLab). [email protected] Dr. Nikolai Lihotzky, MBA Vanderbilt University, is a senior strategic controller at Henkel in Du¨ sseldorf, Germany. [email protected]

532

Customer Retention Management