Rooted in Japanese folklore, the tale of the three monkeys (saru) who clasp both hands over eyes, ears, or mouth, thus not seeing (mizaru), not hearing (kikazaru), or not speaking (iwazaru) evil, can be traced back as far as the late Muromachi period (1333–1568). Nowadays, this story stands for human behavior of playing blind, deaf, and dumb in order to avoid recognizing unpleasant aspects of reality (adapted from Campbell & Noble,1993).

It is well known that complaint management offers many potential benefits to companies. By effectively soliciting, handling, and analyzing customer complaints, firms can ensure high levels of customer satisfaction and loyalty (e.g., Fornell, 1981; Smith, Bolton, & Wagner, 1999) and, in turn, increase their market share and reduce their expenditures for offensive marketing (e.g., advertising) (Fornell & Wernerfelt, 1987). Thus, investments into complaint management may yield very high returns (e.g., Rust, Subramanian, & Wells, 1992; TARP, 1986).

However, there is ample evidence that many firms do not manage complaints effectively. Specifically, after a dissatisfying experience, customers are often unable or reluctant to complain due to a lack of appropriate complaint channels (e.g., Andreasen, 1988; Richins, 1987). Moreover, about half of the customers who do complain are reported to be dissatisfied with complaint handling (e.g., Estelami, 2000; Grainer, 2003). This often leads to a “double deviation” effect (Bitner, Booms, & Tetreault, 1990, p. 80), i.e., even more negative customer feelings about the organization after the complaint. In addition, many firms do not use the information inherent in complaints to initiate systematic improvements (e.g., Brown, 1997; Fornell & Westbrook, 1984). Against this background, ineffective complaint management is likely to contribute to persistently high or even increasing levels of dissatisfaction and defection in the marketplace (e.g., Fornell, Johnson, Anderson, Cha, & Bryant, 1996).

These facts strongly indicate the existence of a paradox: Despite substantial potential benefits, “many examples of poor complaint management” (Rust, Zahorik, & Keiningham, 1996, p. 182) can be observed in the marketplace. There seem to be significant organizational barriers which make it difficult for companies to implement an effective complaint management (see also remarks by Cook & Macaulay, 1997, p. 39).

Previous research has largely neglected such barriers (see Fornell & Westbrook, 1984 as well as Gilly, Stevenson, & Yale, 1991 for studies that at least partially address this issue). Against this background, our study seeks to provide an understanding of this phenomenon. Drawing on psychological and organizational theory, we introduce the concept of defensive organizational behavior towards customer complaints. This notion refers to organizational behavior which avoids contact with dissatisfied customers, dissemination of complaint-related information within the organization, and responsiveness to complaints. This behavior parallels that of the three monkeys in the above mentioned Japanese tale who deny the existence of evil by trying to cover their eyes (see no evil), ears (hear no evil), and mouths (speak no evil).

Besides providing a theoretical basis, we also develop a conceptualization of defensive organizational behavior towards customer complaints (in the future referred to as DOB), thereby identifying the different facets and types of DOB that may exist in a firm. Based on this conceptualization, we conduct a large scale empirical study that analyzes determinants and outcomes of DOB. This study is based on a dyadic data set in which data obtained from firms are matched with assessments obtained from these firms’ complainants.

We feel that studying this phenomenon can make a significant contribution to our academic understanding of complaint management. The relevance of this topic is also emphasized in a review article by Dellande (1995, p. 35) who states that “more research effort is needed to better understand [...] why a firm might disregard or only superficially address consumer dissatisfaction at the expense of the firm’s long term profit.” Besides being theoretically interesting, our study is also relevant from a managerial perspective. It provides insight into an important organizational phenomenon and guides managers on how to reduce this powerful impediment for the implementation of an effective complaint management.

Theoretical background

Our point of departure for developing a theoretical explanation for the phenomenon of DOB is the theory of defense mechanisms coming from the field of individual psychology. In accordance with previous research, we then transfer this concept to the organizational level.

Introduced by the groundbreaking work of Freud (1894/1962, 1926/1959) on psychoanalysis and expanded by the research of Freud (1936/1946), the theory of defense mechanisms was originally developed to explain a person’s efforts to avoid recognizing sexual or aggressive desires. Contemporary research in personality and social psychology, however, is more in accordance with Fenichel (1945) who broadened the role of defense mechanisms to include the protection of self-esteem (e.g., Baumeister, Dale, & Sommer, 1998; Cramer, 2000).

Most human beings tend to hold overly favorable views of themselves (e.g., Mabe & West, 1982). Therefore, in case “an internal or external event occurs that clearly violates the preferred view of self [...], it is necessary for the self to have some mechanism [...] to defend itself against the threatening implications of this event” (Baumeister et al., 1998, p. 1082). Thus, defense mechanisms “are habitual and unconscious strategies used to deny, distort, or counteract sources of anxiety and to help maintain an idealised self-image [...]. They are learned and incorporated into patterns of acting because they are rewarding in the respect that tension and pressures are decreased by their use” (Oldham & Kleiner, 1990, p. 1). Psychologists have identified various types of defense mechanisms (e.g., Laughlin, 1970). Among those, isolation, denial, projection, rationalization, and repression are particularly relevant to our study.

Isolation relates to the creation of “a mental gap or barrier between some threatening cognition and other thoughts and feelings” (Baumeister et al., 1998, p. 1099). By keeping a psychological distance from unpleasant issues, individuals minimize the perceived impact of these issues. Oldham and Kleiner (1990, p. 2) describe this type of defense mechanism as the attempt to “flee from [...] problems [in order to] achieve some protection.”

Denial represents the refusal to recognize facts of the environment that may damage self-esteem (e.g., Oldham & Kleiner, 1990). For example, people “dispute or minimize information that threatens their self-esteem, [...] discount bad feedback [or] selectively forget material that is disagreeable or esteem-threatening” (e.g., Baumeister et al., 1998, p. 1112). Also, through denial, individuals “seek to disclaim knowledge and responsibility, to reject claims made on them, and to disavow acts and their consequences” (Brown & Starkey, 2000, p. 105).

Projection describes the rejection to accept own bad traits, faults, or shortcomings which, in turn, results in seeing others as having the same bad traits, faults, or shortcomings (e.g., Baumeister et al., 1998; Newman, Duff, & Baumeister, 1997). In addition, the perceptions of these bad traits, faults, or shortcomings of other people tend to be even exaggerated (e.g., Newman et al., 1997; Oldham & Kleiner, 1990).

Rationalization reflects the “attempt to justify impulses, needs, feelings, behaviors, and motives that one finds unacceptable” (Brown & Starkey, 2000, p. 106) by substituting the real reason for these phenomena with a logical and socially accepted reason. In doing so, people can avoid criticism and disapproval and make themselves feel better (e.g., Oldham & Kleiner, 1990).

Repression refers to the exclusion of unpleasant thoughts or feelings from consciousness (e.g., Brewin & Andrews, 2000). For example, individuals tend to repress the memory of an embarrassing incident or forget to perform an unpleasant duty (Oldham & Kleiner, 1990).

While defense mechanisms protect from mental pain, they may also cause negative consequences such as chronic avoidance of action, blame, and change (Baumeister & Scher, 1988; Cramer, 2000).

In this study, we transfer the theory of defense mechanisms to the organizational level, thus looking at typical employee behavior in a firm (which ultimately leads to corresponding organizational behavior) rather than at behavior of specific individuals. It is widely accepted that this theory can also be applied in an organizational setting. For example, Brown and Starkey (2000, p. 104) note: “Like individuals, the [...] organization seeks to maximize self-esteem [...]. [O]rganizational concepts of self are maintained by a variety of defenses that are engaged in order to avoid psychic pain and discomfort, allay or prevent anxiety [...], and generally support and increase self-esteem.”

Important research in organization theory in which the concept of defense mechanisms plays a central role has been provided by Argyris (1985, 1990). According to his perspective on organizational learning, employees strive to be in control of the context in which they operate, struggle to win, and try to minimize negative feelings in themselves or others. Thus, whenever they are confronted with an “issue that contains significant embarrassment or threat, they act in ways that bypass [...] the embarrassment or threat” (Argyris, 1990, p. 25). Thereby, employees advocate their own position in order to be in control and to win and save their own face or that of others (Argyris, 1985, 1990). Because these defensive actions are frequently used, they become organizational defensive routines, i.e., “actions [...] that prevent [...] the organization from experiencing embarrassment or threat” (Argyris, 1990, p. 25).

Organizational defensive routines are typically not openly practiced but rather covered with further organizational defensive routines which, for example, hide these behaviors from supervising managers (Argyris, 1985, 1990). Thus, they do not only protect from a potential embarrassment or threat but also “prevent [...] from identifying and getting rid of the causes of the potential embarrassment or threat” (Argyris, 1990, p. 25). Hence, they lead to blockage and distortion in upward communication and can be described as antilearning, overprotective, and self-sealing (Argyris, 1985, 1990). They also lead to denial of responsibility for mistakes and problem solving (Ashforth & Lee, 1990) and ultimately cause employees “to act in ways that are counterproductive to the formal goals or objectives of their organization” (Argyris, 1990, p. 45).

Conceptual framework and constructs

Our unit of analysis is a company and its complaining customers. The framework (see Fig. 1) encompasses three domains including defensive organizational behavior towards complaints (which represents the focal construct of the study) as well as its organizational antecedents and its consequences with respect to customer post-complaint reactions.

Figure 1
figure 1

Framework and constructs

Full size image

Following our theoretical discussion and in line with literature on organizational behavior (e.g., Ashforth & Lee, 1990) and complaint management (e.g., Fornell & Westbrook, 1984), we argue that individuals in organizations perceive complaints as a source of threat to self-esteem, reputation, autonomy, resources, or job security. Thus, to protect themselves against this threat, they exhibit different types of defensive behavior towards complaints. In the following, we first categorize, introduce, and explain these different types of DOB and then introduce the constructs in the two other domains of our framework.

Complaints are highly relevant sources of market information (e.g., Kasouf, Celuch, & Strieter, 1995). Thus, to study different types of DOB, we adopt a behavioral perspective on market orientation which is characterized by an information processing view on organizations (e.g., Jaworski & Kohli, 1993; Moorman, 1995). More specifically, following the conceptualization of market orientation by Moorman (1995) and in line with measurement literature (e.g., Edwards, 2001; Law, Wong, & Mobley, 1998), we regard the complex phenomenon of DOB as a construct that is determined by the following three facets: DOB with respect to complaint acquisition, DOB with respect to complaint transmission, and DOB with respect to complaint utilization.

Defensive organizational behavior with respect to complaint acquisition

The first category of DOB refers to the widespread absence of an effective solicitation and registering of customer complaints in business practice. More specifically, there is evidence that many companies neither actively seek feedback from dissatisfied customers (e.g., Plymire, 1990; Rust et al., 1996) nor do they react in a friendly way when confronted with such feedback (e.g., Best, 1981; Tax & Brown, 1998).

Isolation from complaints (DOB1)

The defense mechanisms ‘isolation’ (keeping a psychological distance from unpleasant issues) and ‘denial’ (refusing to recognize threatening facts) contribute to a theoretical explanation for the fact that “most firms tend to avoid [complaints] rather than solicit them” (Estelami, 1999, p. 166). Specifically, companies often do not provide communication channels that would enable customers to complain in a cost-effective, easy, and uncomplicated way (Fornell, 1981; Rust et al., 1996). Also, many firms tend to discourage customers from complaining by not communicating their responsiveness to complaints or by not informing where, how, and to whom customers can complain (e.g., Kendall & Russ, 1975). On an individual level, employees “tend to personalize complaints, seeing them as personal attacks, so they [...] prefer to avoid the issue” (Plymire, 1990, p. 51). This is in line with Berry (1995, p. 99) who stresses that “[t]he natural temptation is to avoid customers carrying bad news.”

Hostile behavior towards complainants (DOB2)

This type of DOB especially relates to the defense mechanisms ‘isolation’ (keeping a psychological distance from unpleasant issues), ‘denial’ (refusing to recognize threatening facts), and ‘projection’ (rejecting to accept own bad traits, faults, or shortcomings and exaggerating perceptions of other individuals). Complaints are unpleasant or threatening issues, because they symbolize customer problems and, in turn, potential negative consequences for employees. Thus, when confronted with complaints, staff often react in a hostile manner, especially when they have caused the corresponding problem (e.g., Rust et al., 1996). For example, they frequently deny responsibility for registering complaints (e.g., Ashforth & Lee, 1990; Best, 1981) and blame complainants for the failure (Best, 1981; Tax & Brown, 1998). In addition, they often treat complaining customers rudely and become increasingly angry as the dispute progresses (e.g., Best, 1981; Tax & Brown, 1998). This is supported by Menon and Dubé (2000) who find that a person’s expression of anger (e.g., a complaint) naturally causes similarly hostile and aggressive responses in other people.

Defensive organizational behavior with respect to complaint transmission

The second category of DOB alludes to the common lack of an effective intra-organizational transmission of customer complaints to complaint managers and senior managers (e.g., Fornell & Westbrook, 1984; Gilly et al., 1991).

No (or biased) transmission of complaints to complaint managers (DOB3)

This type of DOB is particularly linked to the defense mechanisms ‘projection’ (rejecting to accept own bad traits, faults, or shortcomings and exaggerating perceptions of other individuals), ‘rationalization’ (justifying unpleasant facts with a logical and socially accepted reason), and ‘repression’ (excluding threatening issues from consciousness). The fact that the intra-organizational transmission of unpleasant information is subject to considerable suppression or distortion is confirmed by various empirical studies (e.g., O’Reilly & Roberts, 1974; Wilensky, 1967). In particular, employees are shown to be reluctant to pass information along to the rest of the firm in a complete and accurate way, if the content could lead to unfavorable consequences (e.g., punishments) for them (e.g., Kaufman, 1973; Read, 1962). With respect to complaints, it is important to note that “customers often lodge complaints with the nearest employee” (Tax & Brown, 1998, p. 84). In many cases, this employee is at least partly responsible for the cause of the complaint. Thus, customer contact personnel may often be averse to forward complaints to complaint managers in a complete and accurate manner (Gilly et al., 1991; Tax & Brown, 1998). In line with this, an empirical study by Ross and Gardner (1985) reports that many complaint managers believe that they receive incomplete information about complaints.

No (or biased) transmission of complaints to senior managers (DOB4)

This type of DOB can be best illustrated by the defense mechanisms ‘denial’ (refusing to recognize threatening facts) and ‘repression’ (excluding threatening issues from consciousness). According to Argyris (1990), employees (e.g., complaint managers) frequently attempt to save their own face or that of other organizational members (e.g., senior managers). This behavior can lead to the creation of “organizational black holes in which information from below gets lost” (Argyris, 1990, p. 23). In support of this, work in social psychology indicates that individuals are reluctant to communicate messages that are perceived as unpleasant for the recipient (so-called MUM effect; e.g., Rosen & Tesser, 1970). Also, studies of organizational behavior report that group members tend to “appoint themselves as mindguards to protect the leader [...] from adverse information that might break the complacency they shared about the effectiveness and morality of past decisions” (Janis, 1977, p. 340). Evidence for blocked or distorted transmission of complaint data to senior managers can also be found in complaint literature. An empirical study by Fornell and Westbrook (1984) shows that complaint managers become increasingly reluctant to transmit complaint data as the relative number of complaints increases. These authors argue that this is partly due to “the diminished stature of the bearer of bad tidings” (p. 69). For example, senior management may be unwilling to devote attention to communication about complaints and even put the blame on the person trying to transmit the information. Thus, the empirical finding by Kasper (1984) that many senior managers are not systematically informed about complaints is not very surprising.

Defensive organizational behavior with respect to complaint utilization

The third category of DOB relates to the prevalent lack of an effective utilization of customer complaints in terms of complaint handling (e.g., Best, 1981), complaint analysis (e.g., Brown, 1997), and decision making (e.g., Fornell & Westbrook, 1984).

No (or inadequate) handling of complaints (DOB5)

This type of DOB is closely linked to the defense mechanisms ‘isolation,’ ‘denial,’ and ‘repression’. Complaint literature provides considerable support for the presence of this type of DOB. For example, complainants often do not receive any organizational response at all or only after a long period of time (e.g., Kauchak, 1991; Rust et al., 1996). In addition, a significant number of companies seem to offer redress only when they are legally bound to do so (e.g., Halstead, Dröge, & Cooper, 1993) and tend to provide less redress than customers expect to receive (e.g., Best, 1981).

No (or inadequate) analysis of complaints (DOB6)

The theoretical explanation for this type of DOB is essentially based on the defense mechanisms ‘isolation,’ ‘denial,’ and ‘repression.’ Moreover, Argyris (1990) argues that firms tend to cover defensive behaviors with further defensive routines which, in turn, prevents them from identifying the causes of the potential embarrassment or threat. In our context, these mechanisms can keep firms from regularly and systematically analyzing the reasons for complaints and identifying the root causes of customer dissatisfaction. This reasoning is supported by complaint literature that points to the widespread absence of an effective complaint analysis (e.g., Best, 1981; Brown, 1997; Kendall & Russ, 1975).

No (or inadequate) use of complaint information in decision making (DOB7)

Finally, we assume that senior managers may use defense mechanisms such as ‘isolation,’ ‘denial,’ ‘rationalization,’ and ‘repression’ to preserve self-esteem. This is supported by the concept of organizational defensive routines which argues that senior managers (like other individuals in organizations) try to defend their position and to save their face when confronted with unpleasant issues (e.g., complaints as symbols of own wrong decisions in the past). Thus, senior managers may not use aggregated complaint data when making marketing decisions.

This is in line with organizational research on the phenomenon of ‘groupthink’ (i.e., a collective pattern of cognitive defenses to support and justify past decisions) in senior management teams. Symptoms of groupthink include, among others, collective rationalization to discount negative feedback and the illusion of invulnerability which causes senior managers to fail to respond to clear warnings of danger (Janis, 1977, 1982). Additional support for this type of DOB is provided by research on ‘escalation of commitment’. This term describes the tendency of senior managers responsible for a wrong decision in the past to become overly committed to this incorrect course of action (e.g., Staw, 1981). Moreover, this type of DOB can also be explained by firms’ inability to promote active unlearning (i.e., discarding previously beneficial, but now dysfunctional organizational knowledge and practices) (e.g., Hedberg, 1981). Complaint literature also presents evidence for this type of DOB. Because complaints are highly symbolic, “they might imply failure or inadequacies of previous marketing decisions and constitute criticism of the individuals responsible for the problematic policies or programs” (Fornell & Westbrook, 1984, p. 69). Thus, senior managers tend to be reluctant to use complaint data in their decision making process (e.g., Best, 1981; Brown, 1997; Fornell, 1981).

Table 1 presents a summary of the different types of DOB including examples as well as selected references and corresponding underlying defense mechanisms.

Table 1 Overview of defensive organizational behavior towards complaints
Full size table

Antecedents of defensive organizational behavior towards complaints

In line with research on mechanisms that influence the behavior of marketing personnel (e.g., Hartline, Maxham, & McKee, 2000) and with research on barriers to market-oriented behavior (e.g., Harris, 1998), we suggest that the presence of DOB is affected by the firm’s human resource management (HRM) and culture. This is also in accordance with two recent complaint management studies that provide evidence for an impact of a firm’s internal environment (in terms of HRM and culture) on the behavior of complaint-handling staff (Homburg & Fürst, 2005; Maxham & Netemeyer, 2003).

We define the supportiveness of HRM with respect to complaint management as the degree to which personnel-related activities of an organization favor effective solicitation, handling, and analysis of complaints. These activities encompass both leadership behavior and training. Highly supportive leadership behavior covers, for example, the communication of the benefits of an effective complaint management to employees, the setting of a good example in terms of high customer orientation in general and effective complaint management in particular, and the performance evaluation of employees based on the achievement of customer-focused goals (e.g., de Ruyter & Brack, 1993; Kennedy, Goolsby, & Arnould, 2003). With respect to training, this construct covers activities that aim at assuring employees’ sensitivity to the importance of complaints as well as employees’ abilities to deal with dissatisfied customers (e.g., de Ruyter & Brack, 1993; Maxham & Netemeyer, 2003).

Following the widely accepted definition of organizational culture by Deshpandé and Webster (1989, p. 4), we define the customer orientation of corporate culture as the degree to which shared values, norms, and behaviors in an organization favor a customer-oriented thinking and acting of employees. More specifically, our conceptualization refers to an organizational orientation that puts the customer’s interest first (Deshpandé, Farley, & Webster, 1993). The high importance of this construct is stressed by Parasuraman (1987, p. 41): “Perhaps the most precious asset that [...] firms can acquire is a single-minded dedication on the part of all its employees toward satisfying its customers.”

Consequences of defensive organizational behavior towards complaints

As discussed above, DOB is caused by mechanisms that aim at protecting from a potential threat. However, we have also described that DOB can have substantial negative consequences. Such negative consequences may include unfavorable effects on complaint satisfaction and on perceived complaint-based improvements.

We define complaint satisfaction as the degree to which the firm’s complaint-handling performance meets or exceeds the complainant’s expectations (Gilly & Gelb, 1982). From the customer’s viewpoint, the occurrence of a problem followed by the firm’s complaint handling can be regarded as a specific transaction experience. Thus, complaint satisfaction represents a particular form of transaction-specific customer satisfaction (e.g., Smith & Bolton, 1998).

Perceived complaint-based improvements is defined as the degree to which the complainant has the impression that the firm makes an effort to avoid the problem in the future. A longitudinal study of complainants’ evaluations by Maxham and Netemeyer (2002) provides empirical evidence for the relevance of this construct. These authors find that customers reporting two similar failures of a firm adopt a particularly critical perspective on this company. Based on this result, Maxham and Netemeyer (2002, p. 67) recommend to firms: “Do not make the same mistake twice.”

Hypotheses development

As mentioned above, an important goal of this study is to detect factors that influence the newly introduced concept of DOB. Thus, in the first subsection, we develop hypotheses related to the antecedents of DOB. Moreover, our study seeks to identify the consequences of this new concept. Hence, corresponding hypotheses are developed in the second subsection.

Hypotheses related to antecedents of DOB

We first address the link between supportiveness of HRM with respect to complaint management and the prevalence of DOB. There is substantial evidence that customer-oriented leadership behavior positively affects employees’ customer-oriented behavior in general (e.g., Jaworski & Kohli, 1993) and customer-oriented treatment of complaints in particular (e.g., TARP, 1986). Thus, leadership behavior such as communicating the benefits of an effective complaint management to employees, setting a good example in terms of customer orientation in general and complaint management in particular, and evaluating staff performance based on the achievement of customer-focused goals is likely to reduce the tendency of employees to exhibit DOB. This is also in line with instrumental conditioning theory (e.g., Skinner, 1938) and social learning theory (e.g., Bandura, 1977). Moreover, customer-oriented training is reported to enhance employees’ customer-oriented behavior in general (Schlesinger & Heskett, 1991) and customer-oriented complaint handling in particular (Tax & Brown, 1998). Thus, training activities that aim at improving staff abilities to deal with dissatisfied customers may decrease the prevalence of DOB. Therefore, we predict:

H1: The supportiveness of HRM with respect to complaint management has a negative impact on the prevalence of DOB.

Work on defensive organizational behavior points to the “possibility that organizational culture is a ‘meta-cause’ of much defensive behavior. Specifically, the shared system of values, assumptions, and norms may well [...] influence the tendency to avoid action, blame, and change” (Ashforth & Lee, 1990, p. 631). This is in line with empirical studies reporting that the more employees perceive their firm to be customer-oriented, the stronger is their own customer-oriented behavior (e.g., Kelley, 1992; Siguaw, Brown, & Widing, 1994). Furthermore, complaint research emphasizes the relevance of a customer-oriented corporate culture as a supportive factor for complaint management (e.g., Kasouf et al., 1995). Thus, we hypothesize:

H2: The customer orientation of corporate culture has a negative impact on the prevalence of DOB.

Hypotheses related to consequences of DOB

We have already described that defensive organizational behavior can impair organizational efficiency and effectiveness. More specifically, “while defensiveness may serve the short-run interests of individuals, it tends not to serve their long-run interests nor the short- or long-run interests of recipients or organizations” (Ashforth & Lee, 1990, p. 643).

Hostile employee behavior towards complainants inevitably leads to customer perception of unfair interpersonal treatment (Maxham & Netemeyer, 2003). In addition, a blocked or distorted complaint transmission to complaint managers and a lacking or inadequate complaint handling slow down the complaint process and impede a fair complaint outcome (Gilly et al., 1991; Smith et al., 1999). As customer evaluation of complaint satisfaction is based on perceptions of interpersonal treatment, complaint process, and complaint outcome (e.g., Smith et al., 1999), we hypothesize:

H3: The prevalence of DOB has a negative impact on complaint satisfaction.

Organizational isolation from complaints impedes the acquisition of information about customer problems (Fornell, 1981). Also, a blocked or distorted communication of complaints to complaint managers and a lacking or inadequate analysis of complaints inhibit a complete and accurate identification of root causes of customer problems (Gilly et al., 1991; Kendall & Russ, 1975). Moreover, a blocked or distorted communication of complaints to senior managers and a lacking or inadequate use of complaint information by senior managers in decision making ultimately prevent companies from responding appropriately to market needs by eliminating the root causes of customer problems (Fornell & Westbrook, 1984). In support of this, an empirical study by Johnston (2001) shows a positive link between the quality of the complaint management process and complaint-based improvements. Therefore, we predict:

H4: The prevalence of DOB has a negative impact on perceived complaint-based improvements.

In addition, it might be argued that complaint satisfaction and perceived complaint-based improvements are not only directly influenced by the prevalence of DOB, but also by the supportiveness of HRM with respect to complaint management and the customer orientation of corporate culture, respectively. However, in our view, customers’ complaint-related perceptions (i.e., complaint satisfaction and complaint-based improvements) are solely directly affected by corresponding complaint-related behavior of staff (i.e., whether personnel solicit, handle, and analyze complaints in a proper way as well as use complaint information in decision making). Thus, whereas we believe that organizational variables such as HRM and corporate culture can have a substantial direct effect on complaint-related behavior of staff (therefore indirectly affecting customers’ complaint-related perceptions), we do not see compelling arguments why these variables (which mostly cannot be directly assessed by customers anyhow) may also directly influence customers’ complaint-related perceptions. Nevertheless, we also analyze this issue empirically by testing whether the prevalence of DOB completely or only partially mediates the effects of the supportiveness of HRM with respect to complaint management and customer orientation of corporate culture on complaint satisfaction and perceived complaint-based improvements.

Methodology

Data collection procedure and sample

In the first phase, we identified a company sample (1,786 firms) based on information from a commercial provider. Our sample was restricted to firms with at least 200 employees and an annual revenue of at least $50 million. It covered industries from both the manufacturing and the service sector. For each of the firms, we tried to identify the manager with primary responsibility for complaint management. This was successful in 1,707 cases. Then, we sent a questionnaire to these individuals and started follow-up telephone calls 3 weeks later. As a result, we obtained 379 useable questionnaires which corresponds to a response rate of 22.2%. In order to assess non-response bias, we examined whether the responding firms and the firms we initially addressed differ in terms of size or industry. The results provide no evidence for non-response bias.

In the second phase, we contacted the responding 379 managers again and asked for a list of ten customers who had complained to the company within the last 3 months and who had been typical with respect to reason for complaint, importance to the firm, and customer type. 110 managers agreed to provide the requested information, resulting in a response rate of 29.0%. Given the high confidentiality of customer information, this can be considered as a satisfactory response. By assessing whether the responding firms differ from the firms we initially contacted in the second phase, we again tested for non-response bias. The test related to size, industry, and the prevalence of DOB. We found no statistical differences with respect to these variables. This indicates that non-response bias is also not a problem in the second phase of our data collection.

In the third phase, we conducted telephone interviews with complainants. Our goal to obtain responses from five complainants per company was achieved for all 110 firms. The intraclass correlation coefficient (ICC) (1) values (e.g., Bartko, 1976) for these responses indicate a good consistency among customers reporting on the same firm. Thus, for subsequent data analysis, we averaged the five customer responses for each firm.

Overall, our data analysis is based on 110 dyads. Each of these dyads consists of a managerial assessment of the antecedents and types of DOB in the focal company and five customer assessments related to their post-complaint responses (more detailed information about the company sample and customer sample is available upon request).

Measure development and assessment

We followed standard psychometric scale development procedures. Scales were developed based on a review of the literature and field interviews with practitioners (a complete list of items is available upon request). Using a seven-point rating scale, we operationalized the antecedents of DOB (i.e., supportiveness of HRM with respect to complaint management and customer orientation of corporate culture) with nine and five items, respectively. The focal construct of our study, prevalence of DOB, was also assessed on a seven-point rating scale. For measuring this construct, we used a total of 22 items, covering the seven types of DOB (see Table 1). The consequences of DOB (i.e., complaint satisfaction and perceived complaint-based improvements) were assessed on a five-point rating scale, using three and two items, respectively.

We applied reflective measurement models to all constructs with the one exception of prevalence of DOB which was measured in a formative way. In line with measurement literature (e.g., Edwards & Bagozzi, 2000; Fornell, Rhee, & Yi, 1991), our decision to use a formative model for measuring the prevalence of DOB is “primarily [...] based on theoretical considerations regarding the causal priority between the indicators and the latent variable involved” (Diamantopoulos & Winklhofer, 2001, p. 274).

As discussed in our theoretical section, the construct of DOB represents a complex organizational phenomenon that is determined by a combination of three different facets (i.e., DOB with respect to complaint acquisition, DOB with respect to complaint transmission, and DOB with respect to complaint utilization). These three facets, in turn, each consist of a combination of different types of DOB (see Table 1) which themselves represent a set of very specific behaviors. Thus, we followed the advice of different authors (e.g., Bagozzi, 1994; Fornell & Bookstein, 1982) who strongly recommend the use of a formative measurement model “when constructs are conceived as explanatory combinations of indicators [...] that are determined by a combination of variables” (Fornell & Bookstein, 1982, p. 292). Moreover, our decision was also based on the list of criteria specified by Jarvis, MacKenzie, and Podsakoff (2003).

Against this background, in the model, the prevalence of DOB is represented as a composite latent construct with three formative indicators (which correspond to the three different facets of this construct). Because, in such a model, the error terms associated with these indicators are not identified, we fixed them at one minus the assumed reliability of the index (i.e., Cronbach’s Alpha), times the variance of the index (Jöreskog & Sörbom, 1982).

For each of the constructs with reflective indicators, a single-factor confirmatory factor analysis was carried out using LISREL 8.54. Information about the results of these analyses is shown in Table 2. The global fit measures (i.e., χ 2/df, GFI, CFI, RMSEA, and SRMR) assess how well the confirmatory factor analysis model reproduces the observed variables’ covariance matrix. Overall, the values of these measures suggest that each model fits the data well (Hu & Bentler, 1999). Also, we evaluated measurement reliability and validity for each factor. The results indicate acceptable psychometric properties. Specifically, with respect to the coefficient alpha, all constructs (except perceived complaint-based improvements) show values higher than the suggested threshold value of 0.8 (Nunnally, 1978). Moreover, each construct manifests a composite reliability greater than 0.8, thus exceeding the recommended marginal value of 0.6 (Bagozzi & Yi, 1988). Furthermore, for all constructs (except supportiveness of HRM with respect to complaint management), the average variance extracted is above the suggested threshold value of 0.5 (Bagozzi & Yi, 1988). In addition, with one exception, all individual item reliabilities are greater than the recommended value of 0.4 (Bagozzi & Baumgartner, 1994). Moreover, the factor loadings of all items were significant on the 0.01 level with completely standardized loadings ranging from 0.55 to 0.89. Furthermore, for each pair of constructs, discriminant validity was assessed based on Fornell and Larcker’s (1981) criterion (see Table 2) and on the chi-square difference test (e.g., Bollen, 1989). Results indicate no problems with respect to discriminant validity.

Table 2 Measure-related information regarding reflective measures
Full size table

Results

We tested the hypotheses using causal modeling by means of LISREL 8.54. The overall fit measures (χ 2/df = 1.99, GFI = 0.95, CFI = 0.99, RMSEA = 0.098, SRMR = 0.098) indicate that the hypothesized model is a good representation of the structures underlying the observed data (Hu & Bentler, 1999). Figure 2 displays the results of the hypotheses testing.

Figure 2
figure 2

Results of the hypotheses testing

Full size image

H1 and H2 suggested that the prevalence of DOB is negatively affected by the supportiveness of HRM with respect to complaint management and the customer orientation of corporate culture, respectively. Both hypotheses are confirmed because each of the parameter estimates is negative and significant at least on the 0.05 level.

H3 and H4 presumed a negative effect of the prevalence of DOB on complaint satisfaction as well as on perceived complaint-based improvements. Because both parameter estimates are negative and significant on the 0.01 level each of the hypotheses is confirmed. It is noteworthy that these two hypotheses include dependent and independent constructs which were measured on different sides of the dyad. We feel that the confirmation of both hypotheses by data “crossing the boundaries of the company” is a strong empirical support for our theoretical reasoning.

Moreover, for both ultimate endogenous constructs of our model (i.e., complaint satisfaction and perceived complaint-based improvements), we conducted tests to examine whether the prevalence of DOB completely or only partially mediates the effects of the exogenous constructs (i.e., supportiveness of HRM with respect to complaint management and customer orientation of corporate culture) on these constructs. Therefore, we introduced direct effects of the exogenous constructs on the ultimate endogenous constructs. As we expected, the parameter estimates for these (newly established) direct effects are all non-significant (p > 0.10), whereas the parameter estimates for the effects hypothesized in our model are all significant (p < 0.10). This provides evidence for the fact that the prevalence of DOB completely mediates the effects of the exogenous constructs on the ultimate endogenous constructs of our model.

In summary, our empirical findings strongly support our theoretical reasoning. A more detailed discussion of our results is provided in the next section.

Discussion

The point of departure for our study was the observation that, despite its widely accepted high importance, there is ample evidence that many firms do not have an effective complaint management. In order to explain this paradoxical situation, we introduced the construct of DOB. Although specific aspects of this phenomenon have been previously mentioned in the literature (e.g., Fornell & Westbrook, 1984; Gilly et al., 1991), to the best of our knowledge, there is no study which systematically addresses this phenomenon. Our study has implications for research and managerial practice alike.

Research issues

We feel that the introduction of the construct of DOB is an important contribution of our study to the marketing discipline. Besides developing a theoretical basis for the prevalence of DOB, we also provide a rich conceptualization and operationalization of this construct as well as an empirical study of its antecedents and consequences. Our research underlines the importance of this construct as we find that the prevalence of DOB significantly affects customer perceptions of a firm’s complaint management. The validity of this result is enhanced by the use of dyadic data. Thus, our study constitutes a fairly comprehensive treatment of this phenomenon. Although defensive behavior in organizations has been discussed to some extent in the organization literature, to the best of our knowledge, there is no comprehensive approach to conceptualizing and empirically studying this phenomenon. This void is also noted in a review article by Ashforth and Lee (1990, p. 642): “It is surprising [...] that the phenomenon of defensiveness has not been systematically explored in the literature [...]. Given the apparent pervasiveness of defensive behavior in organizations and the severity of its effects, it is time that researchers investigate the whys and wherefore of this important phenomenon.” Our research represents a first step towards understanding this important phenomenon in the context of dealing with customer complaints.

Our study has also a number of implications for complaint research. This research stream can basically be divided into two categories. First, there is a large number of studies that analyze customer complaint behavior thus typically adopting a consumer behavior perspective (e.g., Singh, 1990; Smith & Bolton, 1998). Second, there is a small number of studies that focus on complaint management, thereby adopting a company perspective (e.g., Fornell & Wernerfelt, 1987; Johnston, 2001). By simultaneously analyzing mechanisms within a company and resulting customer perceptions, we feel that our study is one of the first to build a bridge between these two research streams (see Homburg & Fürst, 2005 for another example). Complaint research would certainly benefit from additional studies linking aspects of organizational complaint management to customer reactions. Obviously, such studies require a dyadic approach to data collection.

Furthermore, our findings advance complaint research by providing insight into drivers of the phenomenon of DOB. In this context, the explained variance of our focal construct is of particular relevance. The fact that the antecedents in our model explain 68% of the variance of the prevalence of DOB is very encouraging. This finding means that the prevalence of DOB can be largely explained by the two antecedents included in our model.

Moreover, our study sheds light on the mechanisms leading to DOB. Specifically, both the supportiveness of HRM with respect to complaint management and the customer orientation of corporate culture have a significant effect on the prevalence of DOB. As the findings shown in Fig. 2 indicate, the prevalence of DOB is more strongly affected by the supportiveness of a firm’s HRM than of the customer orientation of a firm’s culture.

In addition, we feel that the concept of defensive organizational behavior towards certain stimuli of the environment is also relevant to other areas of organizational research in marketing. For example, future research in the field of market or customer orientation (e.g., Deshpandé et al., 1993; Jaworski & Kohli, 1993) might analyze the prevalence of defensive organizational behavior towards a stronger organizational focus on customer needs. Similarly, research looking at forces that drive the use of market information in firms (e.g., Moorman & Zaltman, 1992) as well as research analyzing organizational learning about markets (e.g., Sinkula, Baker, & Noordewier, 1997) could also draw on this concept.

Finally, we are not aware of a single application of our theoretical basis of individual defense mechanisms in the field of marketing. This theory may, for example, be relevant to areas focusing on interpersonal interaction such as personal selling (e.g., Johnston & Kim, 1994) and service encounters (e.g., Bitner et al., 1990). We hope that our introduction of this theory to the marketing literature will lead to its increasing usage in the marketing discipline.

Managerial implications

On a very general level, our study underlines the high managerial relevance of an effective complaint management. This relevance becomes visible by the fact that a poor solicitation, handling, and analysis of complaints inevitably lead to substantial negative consequences for a company (i.e., unfavorable customer post-complaint reactions). Thus, managers responsible for customer relationship management should focus a lot of attention, energy, and resources on developing an effective complaint management.

Moreover, our study should sensitize managers for the phenomenon of DOB as an important impediment for the implementation of an effective complaint management. Based on our findings, managers are encouraged to systematically reduce the prevalence of DOB within their companies. This can be achieved in two ways.

First, managers may work directly on this phenomenon. Our conceptualization of the prevalence of DOB (i.e., the identification of three different facets and seven different types, respectively) provides managers with a checklist type of structure (see Table 1). Based on this, they can analyze the presence of DOB in their firm and, in turn, initiate activities to reduce this behavior. For example, in a firm that is characterized by a high degree of isolation from complaints (DOB1), managers are well advised to establish appropriate complaint channels, to communicate their firm’s responsiveness to complaints, and to inform where, how, and to whom customers can lodge a complaint. In addition, managers who have identified hostile behavior towards complainants (DOB2) may establish an adequate organizational policy for employees’ behavior towards complainants and regularly observe staff adherence to these guidelines. Moreover, in case of no (or biased) transmission of complaints to complaint managers (DOB3) or senior managers (DOB4), a company could, for instance, develop and monitor an appropriate formal organizational procedure for registering and processing complaints as well as implement a suitable information system for complaint management. Similar actions may be taken if employees do not handle complaints at all or only in an inadequate manner (DOB5). Furthermore, managers who have detected the lack of an appropriate analysis of complaints (DOB6) are well advised to establish and monitor standard operating procedures for systematically analyzing the reasons for complaints and identifying the root causes of customer dissatisfaction. An appropriate information system for complaint analysis could also help to examine complaints at an aggregate level. Moreover, in a firm in which aggregated complaint data is not used in decision making (DOB7), senior managers should be frequently reminded of the relevance of this form of customer feedback. On a more general basis, managers should sensitize staff to the existence and the problematic consequences of the different types of DOB. Also, they could threaten sanctions against employees who repeatedly show this behavior.

Second, managers may also work on the antecedents of DOB. In this context, the supportiveness of a firm’s HRM with respect to complaint management is particularly important. Therefore, managers seeking to reduce the existence of DOB should lay stress on implementing training activities that aim at assuring employees’ sensitivity to the importance of complaints and employees’ abilities to deal with dissatisfied customers. By showing leadership behavior such as emphasizing the benefits of an effective complaint management to staff, setting a good example in terms of customer orientation in general and complaint management in particular, and evaluating staff performance based on the achievement of customer-focused goals, managers can further contribute to a decrease in the prevalence of DOB. Like an adequate HRM, a highly customer-oriented corporate culture has also shown to decrease the prevalence of DOB. Thus, firms should also try to create a corporate culture that puts the customer’s interest first.

Finally, our study shows that not only customer satisfaction with complaint handling is under managerial control, but also customer perception of a firm’s effort to avoid problems in the future. Therefore, managers in charge of complaint management should not only focus on an effective handling of complaints, but also ensure that customers get the impression that their complaints stimulate improvement and learning processes within the company. This can, for example, be achieved through systematic feedback to complainants some time after their complaint has been resolved, thereby informing about improvements initiated through their complaint. In addition, customer contact personnel should be instructed to communicate in such a way that customers get the impression that, besides getting the problem out of the world (i.e., resolving individual customer complaints), the company is also interested in understanding and eliminating the underlying causes.