Companies are increasingly implementing key or major account management for their most important business customers (Sengupta et al. 1997a). In some companies, these are also called national or global accounts. Key account managers (KAMs) should develop and build long-term relationships with each of their customer accounts. The account manager helps the customer solve operational and strategic problems by providing research and analysis. As part of the deal, the customer commits to large volume purchases of products and services over a long period of time. This is a challenging boundary role suited only for the best sales professionals.

Selected sales management literature has examined the role of the key or national account manager (for an overview, see Weilbaker and Weeks 1997). Numerous conceptual articles and case studies have described the benefits of implementing such programmes, the traits, skills and knowledge required of key account managers, and the environmental conditions under which such programmes would flourish. However, there has not been an integration of the sales management literature with the literature on relationship marketing (Biong and Selnes 1996; Lambe and Spekman 1997a). We agree that such a move is necessary and desirable because relationships between key account managers and their customer accounts are intended to be long-term strategic partnerships resulting in competitive advantage for both.

The relationship marketing literature (Dwyer Schurr and Oh 1987; Morgan and Hunt 1994; Webster 1992 are just a few examples) has a number of constructs, frameworks and empirical generalizations that can be useful to improving the efficiency and effectiveness of KAM relationships. While the relationship marketing literature discusses both vertical buyer–seller and hori­zontal relationships (Bucklin and Sengupta 1993), KAM relationships fall into the vertical category. An important concept in vertical buyer–seller relationships that can be applied to KAM relationships is switching costs, the focus of this paper.

Conceptual framework and hypotheses

Switching costs were popularized by Jackson (1985) in her study of industrial marketing relationships. She defined switching costs as the psychological, physical and economic costs a customer faces in changing a supplier (Jackson 1985, p. 13). When contemplating a switch in suppliers, a customer faces setup costs and takedown costs (Weiss and Anderson 1992). Setup costs include the cost of finding a replacement supplier who can provide the same or better level of performance than the current supplier or the opportunity cost of foregoing exchange with the incumbent (Dwyer et al. 1987). Takedown costs include relationship-specific or idiosyncratic investments made by the customer that have no value outside the relationship with the current supplier and have to be written off (Anderson and Weitz 1992). The net result of switching costs is that they produce inertia for the customer to remain in the relationship with the current supplier.

The main question we seek to investigate in this paper is whether customer switching costs are good or bad for the customer account and the selling firm in key account relationships. Further, what factors lead to the creation of switching costs so that key account managers may exert some influence over the outcomes in these relationships? To this end we develop a conceptual framework (Figure 1) incorporating relevant antecedents and consequences of switching costs based on the existing literature.

Figure 1 Conceptual framework

Data collection

This study is part of a larger study that the National Account Management Association¹ (NAMA) commissioned us to do in spring 1996 on best practices in key account management among its member companies. We began our study with in-depth exploratory interviews of five key account managers (KAMs) and their supervisors with the aim of obtaining face validity for our framework and generating items to operationalize our constructs.

Following the exploratory interviews, a structured survey questionnaire was designed to tap into all the relevant constructs in our framework using multiple item Likert scales wherever feasible. We borrowed items from existing literature where available and formulated our own questions for new constructs. The survey questionnaire, reply-paid envelope, and requests for participation were mailed to 528 NAMA members in February 1996. One hundred and seventy-six completed, usable surveys were returned, giving a respectable 33% response rate. Our respondents were front-line KAMs from a broad cross-section of US firms across many different industries (55% manufacturing, 43% service). Firms represented were large, with median sales of $3 billion and 14,000 employees. Our KAM respondents had a median age of 41 and were well-educated (61% had a bachelors degree, 28% had a graduate degree). They had a long tenure within their firm, median 12 years, 10 in sales and of those, 3 as a KAM.

Each respondent provided data for a specific customer account they were responsible for. If they handled three or more accounts, they were asked to provide data on the account they spent the third largest amount of time with. If they handled two accounts, they were asked to provide data on the account they spent the largest amount of time with. If they handled just one account, they provided data on that account. This procedure was followed to minimize self-selection bias in the reported account.

A reliability analysis was done for each multiple-item variable. Some individual items were dropped to increase reliability using Cronbach's alpha. The final operationalization of variables along with respective Cronbach's alphas is reported in Table 1.

Table 1: Variable operationalizations and reliabilities

We also did an exploratory factor analysis on all 18 items in Table 1 and found that they load cleanly on the underlying 8 factors. This provides some evidence for the convergent and discriminant validity of our measures. Means, standard deviations and correlations for all our eight variables are presented in Table 2.

Table 2: Means, standard deviations, correlations

Results

In order to test our conceptual framework and hypotheses (Figure 1), we used path analysis with ordinary least squares regression. The results are presented in Table 3.

Table 3: Ordinary least squares regressions

In column 2 of Table 3, we regress customer switching costs against the three antecedent variables – seller adaptation, incentives to customers, customer investment – and the control variable, product complexity. We find that seller adaptation and customer investment are strongly and positively associated with customer switching costs (H1, H3), while incentives to customers are weakly and positively associated with switching costs (H2). Contrary to expectation, product complexity was not related to customer switching costs in our sample.

In column 3 of Table 3, we regress KAM subjective performance against customer switching costs and seller firm size. As expected, switching costs were strongly and positively associated with subjective performance (H4). In a similar vein, switching costs were strongly and positively associated with KAM objec­tive performance (H5) in column 4 of Table 3. Seller firm size had a positive significant relationship with both KAM subjective and objective performance in Table 3.

The explained variance in the dependent variables in Table 3 is modest, indicating that our study omitted some variables that also affect switching costs and KAM performance.

Discussion

In this paper, we have demonstrated the importance of customer switching costs in key account relationships. Customer switching costs have a significant positive impact on both aspects of performance – the achievement of seller firm objectives based on objective criteria such as account market share, sales volume and pro­fitability, and the achievement of customer objectives and satisfaction by more subjective criteria. They help to achieve a win–win for both parties in the relationship. This is a marked departure from conventional thinking in adversarial relationships where customer switching costs were considered good for the seller but not for the dependent buyer.

What is the mechanism by which this win–win is achieved? As our results indicate, the biggest factor that gives rise to customer switching costs is customer investment in relationship-specific assets. If this was the only factor at work, it would make the customer dependent on the supplier and the relationship unbalanced, with unfavourable relationship outcomes.

However, our results show that relationship-specific investment by the customer is not the only factor affecting customer switching costs. There is a dependence balancing mechanism at work by way of seller adaptation. Notwithstanding customer relationship-specific investment, the adaptability or flexibility of the seller firm demonstrates to the customer that the seller is not going to be opportunistic or exploitative. This is reassuring to the customer, helps balance the relationship, and further increases the commitment and switching costs of the customer. This reciprocity or balance is the essence of the win–win in key account relationships. The value of balance or mutual dependence, sometimes called interdependence, has been studied in a channels context with results consistent with our own (Gundlach and Cadotte 1994; Kumar et al. 1995).

Interestingly, we found that ‘push’ and ‘pull’ incentives offered to customers do not do much to increase customer switching costs. Customers seem to value adaptation in a long-term relationship more than short-term financial incentives. While adaptation does require investment on the part of the seller firm, this may be a more cost-efficient use of resources than directly offering incentives. The soft-sell of adaptation may yield better results than the hard-sell of push–pull money in key account relationships.

Finally, when we look at the empirically validated antecedents of customer switching costs, we realize that creating these is the responsibility of the entire selling firm, not just the responsibility of the individual KAM. In order to adapt, the selling firm will need to allocate resources from all over the organization in order to solve specific problems that come up. The KAM can bring the requirements to the attention of the selling firm and can do his or her share of internal marketing. Ultimately, a higher level of management within the selling firm must take responsibility for allocating the necessary resources for adaptation or financial incentives.