Embed Size (px)
Citation preview
Other-customer failureEffects of perceived employee effortand compensation on complainer
and non-complainer service evaluations
Wen-Hsien HuangDepartment of Marketing, National Chung Hsing University, Taichung City,
Taiwan, Republic of China
Abstract
Purpose – The purpose of this paper is to explore how the organizational recovery response toother-customer failure influences the affected customer’s level of satisfaction, unfavorableword-of-mouth (WOM), and repurchase intentions toward the firm.
Design/methodology/approach – Two experimental studies: 2 (complaint versus nocomplaint) £ 3 (employee effort (EE): high versus low versus no) £ 2 (compensation versus nocompensation) were used to test the research hypotheses in a restaurant context.
Findings – In cases of other-customer failure, the outcome valence (failure discontinues versuscontinues) influences how complainants and non-complainants rate their perceptions of satisfactionand subsequent behavioral intentions toward the firm. Customers who perceive that there has beengood EE made to help solve the problem of other-customer failure give higher service evaluations thanthose who perceive little or no EE. Additionally, there are insignificant differences in the rating ofsatisfaction, repurchase intention, or negative WOM from customers in the latter group. Offeringcompensation is not a cure-all. It has the strongest effect on non-complainants in the low EE scenario.
Practical implications – Service managers need to design well-balanced organizational recoverysystems in terms of the outcome of recovery (i.e. compensation) and the way in which the recoveryprocess is delivered (i.e. EE) to the affected customers in response to other-customer failures. Since theinfluence of compensation on service evaluations is largely dependent on customer perceptions ofemployee-effort, providing employees with the appropriate problem-solving skills for working withboth problem-causing customers and the problem-affected customers is a key issue for servicemarketers in cases of other-customer failure.
Originality/value – The paper examines the importance of the as yet under-researched issue of howorganizational recovery responses to other-customer failure influence a customer’s service evaluationsof the firm.
Keywords Consumer behaviour, Customer satisfaction, Complaints, Compensation,Employee behaviour
Paper type Research paper
IntroductionIn many service environments, customer dissatisfaction with the consumptionexperience is derived, at least partially, from the misbehavior of other customers.Bitner et al. (1994), for example, collected 700 incidents from customers of airlines,hotels, and restaurants. They found other-customer misbehavior to be the source of
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-5818.htm
This research was supported by a grant from the National Science Council(NSC97-2410-H-005-029-MY2), Taiwan, awarded to the author.
Other-customerfailure
191
Received 2 July 2008Revised 26 March 2009
Accepted 25 August 2009
Journal of Service ManagementVol. 21 No. 2, 2010
pp. 191-211q Emerald Group Publishing Limited
1757-5818DOI 10.1108/09564231011039286
22 percent of the dissatisfactory incidents. More recently, Grove et al. (1998) in theirinvestigation of the antecedents of both satisfying and unsatisfying incidents thatoccurred at Florida theme parks, concluded that other customers were responsible forthe smallest proportion (14.09 percent) of satisfying events, but the largest proportion(30.95 percent) of dissatisfying events. This problem customer group may be evenlarger in industries in which the customer has greater input into the service deliveryprocess.
What are the specific behaviors of that give rise to dissatisfaction? Examples offrequently mentioned misbehavior include: talking loudly, breaking into line, unrulychildren, smoking, drunkenness, verbal and physical abuse, and so on (Grove and Fisk,1997; Huang, 2008; Martin, 1996; Martin and Pranter, 1989). Other-customer failure issaid to happen when any action by another customer has a negative impact on one’sown service experience (Huang, 2008).
Several studies have shown that unsatisfactory customer-to-customer encountersreflect negatively on the customer’s global evaluation of the service firm (Huang, 2008;Hui and Bateson, 1991; Grove and Fisk, 1997; Grove et al., 1998; Guenzi and Pelloni,2004; Martin, 1996; Moore et al., 2005; Raajpoot and Sharma, 2006). Martin (1996), in hissurvey of restaurants and bowling alleys, found that negative other-customer publicbehavior diminished customer satisfaction with the firm. Harris and Reynolds (2003)reported that the dysfunctional customer behavior lessened the extent of one’s loyaltyand satisfaction toward the service organization. Martin and Pranter (1989) thussuggested that service firms should actively engage in “compatibility management,”the process of attracting a homogeneous clientele to the service environment, thenactively managing both the physical environment and customer interactions, tominimize the frequency of dissatisfying customer-to-customer encounters. However,little is known about what service firms could or should do to mitigate the affectedcustomers’ dissatisfaction when other-customer failure does occur.
Improving customer satisfaction is an important goal in business today. The aim ofthis study is to explore how the recovery response to other-customer failure influencesthe affected customers’ level of satisfaction, negative word-of-mouth (WOM), andrepurchase intentions toward the firm. This will not only assist marketers to build abetter recovery strategy when other-customer failure occurs, but also makes a broadercontribution to the service literature, by providing insight into interpersonalrelationships in customer-to-employee encounters in response to other-customer failure.
Other-customer failureFor more than 20 years, negative interaction between customers in the service settinghas been noted as an important aspect of the service encounter. A number of termshave been used in studies exploring and describing customers’ detrimental manners:Bitner et al. (1994) use the term “problem customers,” Lovelock (1994) uses the word“jaycustomers,” Fullerton and Punj (1997) use the label “consumer misbehavior,” andHarris and Reynolds (2003, 2004) use the phrase “dysfunctional customer behavior.”
Extant research on dysfunctional customer behavior can be grouped in into threedistinct streams. The first includes different types of dysfunctional customer behavior.Zemke and Anderson (1990) grouped a typology of five “customers from hell,”comprising abusive egocentrics, insulting whiners, hysterical shouters, dictators, andfreeloaders. Bitner et al. (1994) identified four problem customer behaviors, from the
JOSM21,2
192
employees’ perspective, in their examination of 774 critical service encounters. Thesewere drunkenness, verbal and physical abuse, breaking company policies or laws, anduncooperative customers. Harris and Reynolds (2004) noted eight fundamental types ofjaycustomer behavior in their look at the hospitality industry, including compensationletter writers, undesirable customers, property abusers, service workers, vindictivecustomers, oral abusers, physical abusers, and sexual predators.
The second stream is related to motives, antecedents, and consequences ofdysfunctional customer behavior. For example, Fullerton and Punj (1993) showed themain drivers of deviant customer behavior to be the consumer’s traits and predispositionwhich includes demographic, psychological, and social influences. Harris and Reynolds(2003) and Wu (2007) explored the downstream effects of dysfunctional customerbehavior. They noted that customer misbehavior had significant negative effects onfellow customers’ service evaluations.
The third stream comprises the fellow customer’s psychological response process todysfunctional customer behaviors. For example, Raajpoot and Sharma (2006) reportedthat mood, expectations, and perceived control over outcome were the three mostimportant factors influencing perception of incompatibility in customer-to-customerinteractions. Huang (2008) discovered that controllability attributions, firmresponsibility, severity of other-customer failure, and perceived employee effort (EE)all contribute to the evaluation process underlying customer dissatisfaction in cases ofother-customer failure.
While there have been a number of studies related to the typology of dysfunctionalcustomer behavior, and the causes and outcomes of dysfunctional customer behavior,little is known about what service organizations should do to relieve the affectedcustomers’ unpleasant feelings when they suffer from other-customer misbehavior.Our goal is to fulfill this gap, focusing on recovery after a service deficiency (or failure)caused by the dysfunctional other customer. In this study, we prefer to use morepertinent term other-customer failure, which refers to actions by another customer,whether intentional or unintentional, that disrupts one’s service experience.
Complaints versus non-complaintsSome customers choose to voice their complaints and seek redress from the firm afterexperiencing other-customer failure, whereas others simply do nothing (Stephens andGwinner, 1998). A number of studies have shown that the most dissatisfied customersdo not complain following a failure (Stephens and Gwinner, 1998; Tax and Brown,1998). Instead, they switch providers and express dissatisfaction by advising others todo the same (Singh, 1990). Why are such customers reluctant to complain? Accordingto Hirschman (1970), customers make a mental judgment of whether “it is worthcomplaining” or “it is not worth complaining” based on their simultaneous assessmentof the probability of success, the effort it takes to complain, and the value of theproduct or service involved. In other words, a dissatisfied customer’s decision to voicea complaint rather than defecting to a competitor depends, in part, on his or her ownestimation of the probability of achieving a positive outcome. The questions that ariseinclude: “should a service firm encourage customers to complain when they suffer fromother-customer misbehavior?” and “are these complainers more or less satisfied thancustomers who do not complain?”
Other-customerfailure
193
Bolfing (1989) found that 67 percent of the complainers in her study engaged innegative WOM, but only 45 percent of non-complainers voiced negative comments toothers. In agreement with these findings, Halstead (2002) found negative WOM to bemore common among customers who voiced complaints. Voorhees et al. (2006) foundthat customers who chose to complain indicated higher levels of negative affect andperceived regret, and were less satisfied, as well as less likely to repurchase in thefuture than those who did not complain. It is assumed the effects in case ofother-customer failure would be similar:
H1. Customers who voice their complaints about other-customer failure to theservice employee are (H1a) less satisfied with the firm, (H1b) less willing torepurchase, and (H1c) more likely to engage in negative WOM than those whodo not complain.
Nevertheless, it has been strongly and paradoxically argued that dissatisfiedcustomers should be encouraged to voice their complaints and seek redress, becausethis provides the firm with an opportunity both to make amends and to correct rootcauses of the service failure (defined as service performance that falls below acustomer’s expectations) (Blodgett and Anderson, 2000; Gronroos, 1990; Voorhees et al.,2006). From a management viewpoint, the questions remain as to how firms shouldbest deal with other-customer failure so as to satisfy both the complainants andnon-complainants, and what is the impact of organizational recovery responses onservice evaluations in these cases? Service recovery involves actions taken by serviceproviders in response to service failures. Previous studies have consistently shown theimportance of service recovery in achieving customer satisfaction in a servicedeficiency situation (Bitner et al., 1990; Smith et al., 1999; Tax et al., 1998). Thus, in thisstudy, this is used as the theoretical basis to provide insight into the factors thatinfluence a complainant or non-complainant’s post-recovery service evaluations incases of other-customer failure.
Organizational recovery responseIt is suggested in the service recovery literature that there are two dimensions criticalto successful service recovery: outcome and process (Bitner et al., 1990; McColloughet al., 2000). The outcome of the service recovery refers to the tangible result deliveredto a dissatisfied customer (i.e. what is delivered), while the process of service recoveryrefers to the manner in which the service firm handles the problem during the course ofservice recovery (i.e. how it is delivered). There is a tendency to look at recoveryevaluations from the perspective of justice (McCollough et al., 2000; Tax et al., 1998).According to justice theory, there are three dimensions of perceived justice thatinfluence how customers evaluate the fairness of a service recovery: procedural,distributive, and interactional justice (IJ) (Blodgett et al., 1997; Smith et al., 1999; Taxet al., 1998).
Procedural justice (PJ) involves the processes, policies, and rules by which recoveryeffort decisions are made. IJ reflects the way in which the customer is treated duringthe service recovery (Smith et al., 1999; Tax et al., 1998). In other-customer failurerecovery situations, PJ is related to the speed with which the customer misbehavior iscorrected or complaints are handled by the service staff, whereas IJ is related to theperceived helpfulness, courtesy, and empathy of the service employees. In short, both
JOSM21,2
194
PJ and IJ are related to how the recovery effort is delivered and are thus closely linkedto customer perceptions of EE made in response to other-customer misbehavior (i.e.service recovery process). Distributive justice (DJ) reflects the perceived fairness of thetangible outcome of the service recovery (i.e. the service outcome). Blodgett et al. (1997)and Tax et al. (1998) have shown that compensation (in the form of discounts, coupons,free gifts, and so forth) is the most important recovery strategy impacting customerperceptions of DJ.
To sum up, there are two dimensions to organizational recovery responses includedin this study: process (EE) and outcome (tangible compensation). We argue that thepost-recovery satisfaction and subsequent behavioral intentions of complainants andnon-complaints are dependent, in large part, on their perceptions of EE to find asolution to the other-customer failure, and to the compensation that the firm offers tomake up for their bad feelings and loss. Specifically, we expect to see interactive effectsbetween these variables.
Perceived EEPerceived EE refers to the amount of energy a customer believes an employee hasinvested on their behalf (Huang, 2008; Mattila and Patterson, 2004; Mohr and Bitner,1995; Specht et al., 2007). Though the firm may view the behavior of other customers asuncontrollable, consumers tend to consider the firm’s management of their guests andtheir guest’s behavior as an important component of the service process, which affectstheir overall service evaluations. Bitner et al. (1990) found that customers often refer toEE when describing highly satisfying or highly dissatisfying service encounters. Forexample, when customers carried the impression that an employee did nothing torespond to potential disruption (e.g. “The hotel staff did nothing to deal with the peoplenoisily partying in the hall at 3 am”), they tended to feel more dissatisfied. Conversely,when the customer carried the impression that the employees actively exerted a greatdeal of effort in response to potentially disruptive others (e.g. “the manager kept an eyeon the obnoxious guy at the bar, to make sure that he did not bother us.”), they feltmore satisfied. These findings clearly suggest that when other-customer failureoccurs (or might occur), customers expect the service employees to keep thedisruptive customer(s) from disturbing others. In other words, a customer’s evaluationof the service is not only affected by the other-customer misbehavior, but also by theemployee reaction.
To sum up, when customers experience other-customer failure, and they perceive EEto be high to correct this problem, they are more satisfied, less likely to indulge innegative WOM, and more willing to come back. Conversely, when customers suffer fromother-customer failure but perceive a low (insufficient) or even lack of EE to help solvethe problem, then these customers become more dissatisfied, more likely to indulge innegative WOM, and less likely to offer future patronage. We therefore propose that:
H2. Customers who carry the impression that there is more EE exerted to helpsolve the problem of other-customer failure will have (H2a) highersatisfaction, (H2b) greater repurchase intentions, and (H2c) lower negativeWOM than those who carry the impression that there is little or no EE.Additionally, for customers in the low and no EE perception category thereare insignificant differences in the rating of (H2a) satisfaction, (H2b)repurchase intention, and (H2c) negative WOM.
Other-customerfailure
195
Offering compensation after other-customer failureEmployee impact on customer satisfaction is not limited only to their behavior duringthe other-customer failure incident. How they interact with the affected customers mayalso have a significant effect on satisfaction. It is suggested that compensation shouldbe made to accommodate the irate customers (Wirtz and Mattila, 2004). Bitner et al.(1990) confirmed that some tangible redress for a customer’s loss would beremembered as more highly satisfying despite the initial deficiency in service delivery.Hoffman et al. (1995) investigated service failures and recovery in the restaurantindustry. They found that compensation had a negative effect on defection rates(percentage of people indicating that they would no longer patronize that restaurant).Davidow and Leigh (1998) showed that compensation had a positive effect onrepurchase intentions and a negative impact on WOM activity. Smith et al. (1999), intheir study of the restaurant and hotel industries, reported a significant relationshipbetween compensation and service encounter satisfaction.
Nevertheless, offering compensation is not a universal antidote. It has been notedthat compensation is less effective in increasing satisfaction when the recovery processis rated poorly. For example, in one study, Blodgett et al. (1997) utilized a scenariowhere tennis shoes wore out too quickly. They inquired into three levels ofcompensation: full exchange, 50 percent discount, or a 15 percent discount. They foundthat respondents receiving less than full compensation (discount of 50 or 15 percent)but who were treated with respect and courtesy were more likely to make futurepurchases and less likely to engage in negative WOM than respondents who received afull exchange but were treated rudely. Subjects in Tax et al.’s (1998) study also reportedthat poor interpersonal treatment lessened the positive effect of compensation.McCollough et al. (2000) showed that passengers felt they were being “bought off”when gate agents provided high levels of compensation, without a corresponding levelof empathy and understanding. To sum up, the extant literature seems to suggest thatwhen the customer perceives a lack of EE, the compensation is not viewed as sincere,which in turn may result in even greater customer dissatisfaction.
On the other hand, it seems likely that there may be a ceiling effect in even the bestrecovery process (Wirtz and Mattila, 2004). In other words, when the customerperceives EE to correct other-customer failure to be high, compensation may not beneeded for them to obtain satisfaction and/or significantly enhance satisfaction.A logical conclusion to this line of thought is that compensation should have thestrongest impact on customers who perceive a medium level of EE (theoperationalization of “low EE” condition in our study). Furthermore, we argue thatit is more effective to offer compensation to non-complainants than complainants.There have been several studies offering some insight into the reasons for this.
Recovery disconfirmation predicts that service recovery expectations do influencepost-recovery perceptions (McCollough et al., 2000). In other words, the lower theexpectation of recovery, the more positive the recovery disconfirmation is. This is thecase because customers may not be expecting compensation from other-customerfailures about which they have not complained. Unexpected compensation may thuslead to positive disconfirmation, and increased satisfaction with the firm. In addition,signaling theory suggests that customers may view voluntary recovery responses ascues that a firm is focused on customer service and concerned about their well-fare,thereby increasing their loyalty to that firm (Kirmani and Rao, 2000). In line with these
JOSM21,2
196
reasoning, we predict that offering compensation will have a greater impact on theservice evaluations of non-complainants than complainants (in the low EE condition).Taken together, we formulate the following hypothesis:
H3. Offering compensation will be more effective at (H3a) increasing satisfaction,(H3b) increasing repurchase intention, and (H3c) lessening unfavorable WOMfor non-complainants in low EE situations. Conversely, offering compensationwill be less effective at (H3a) increasing satisfaction, (H3b) increasing repurchaseintentions, and (H3c) diminishing unfavorable WOM for complainants andnon-complainants in the high (ceiling effect) and no EE situations.
Study 1Research designA 2 (complaint versus no complaint) £ 3 (EE: high versus low versus no) £ 2(compensation versus no compensation) between-subject experimental design was usedto test our predictions.
Subjects were asked to read a written scenario describing an incident ofother-customer failure in a restaurant. The scenario method is appropriate because itallows for greater control over many conditions, removes unmanageable variables thatare present in field conditions, and saves time by summarizing events that mightotherwise unfold over days or weeks (Bitner, 1990). In contrast, asking subjects to recallactual other-customer failure and recovery incidents (retrospective-type method) leadsto increased response bias due to memory lapses and rationalization (Smith et al., 1999).
A restaurant context was chosen as appropriate for this study sincereal-life dysfunctional customer behavior is common in this industry (Harris andReynolds, 2003) and interpersonal customer-to-employee encounters in response toother-customer failure are easily observed in a restaurant.
Sample selection and proceduresConsumers at a large-sized shopping center in Taiwan were recruited as volunteers.Advertisements were posted on bulletin boards at each entrance. Volunteers wereoffered a small gift (about US$5 in value) for participating. The surveys were run on aThursday, Friday, and Saturday, so that both weekday and weekend consumers couldbe polled. Each respondent was given a survey kit consisting of a questionnaire and arandomly chosen scenario. The instructions asked participants to imagine themselvesas the consumer in the scenario. At the end of the survey, respondents were asked tocomplete some demographic information.
A total of 300 individuals took part in the survey. Of these, eight responses wereeliminated from the analysis due to incomplete data and guessing the purpose of thesurvey. The 292 remaining responses were divided into 12 treatment groups, rangingin size from 23 to 25. The average age of participants was 37.8 years; 116 (39.7 percent)of the respondents were male, 176 (60.3 percent) were female; and the average workingexperience was ten years. Of the participants, 55.7 percent had a college degree orhigher; and 78.4 percent reported that they had been bothered by other customers in arestaurant, in a way that was very similar to the story used in our scenarios.
Stimuli developmentIn the scenario, it was stated that the participant was dining with a friend in a quietrestaurant. However, they became irritated by some other loud noisy patrons. In the
Other-customerfailure
197
complaint scenario, the participant complained about the other patrons’ behavior to thefrontline employee, whereas in the no-complaint scenario no complaint was made.
Next, a description of the frontline employee’s effort (high, low- or none) to curb thenoisy behavior of the other patrons was provided. Effort was demonstrated by varyingdegrees of employee persistence, or time spent, in trying to resolve the problem (Mohrand Bitner, 1995). In the high-effort scenario, the frontline employee promptlyapproached the problem customers and asked them three separate times to keep theirvoice down, in an attempt to keep the environment tranquil, whereas in the low-effortscenario, the frontline employee was slower to approach the problem customers andasked them only once to keep their voices down and soon left their table. In theno-effort scenario, no frontline employee action was mentioned. In other words, theemployees did nothing to help solve the problem.
The peaceful atmosphere was restored after the other customer’s noise ceased. In somescenarios, compensation was offered by the restaurant manager. In the compensationscenario, the manger apologized for the other-customer failure and offered a gift of twomugs as compensation, whereas in the no-compensation condition, there was no apologyand no gift. An example of the restaurant scenarios can be seen in the Appendix.
Manipulation checksA four-item, seven-point Likert scale, adapted from Mohr and Bitner (1995), wasemployed to capture the participants’ perceptions of EE (Cronbach’s a ¼ 0.77). Thestatements related to perceptions of EE were as follows: “the employee exerted a lot ofenergy,” “the employee did not spend much time trying to resolve this situation,” “theemployee did not try very hard,” and “the employee put a lot of effort into thissituation.” The high-effort group received a mean rating of 5.62, the low-effort group amean rating of 2.96, and the no-effort group a mean rating of 2.56, F(2, 289) ¼ 224.06,p , 0.001. The contrast between the three levels was significant ( p , 0.05 in all cases).
For the complaint/no complaint against other-customer failure, 85 percent of those whoreceived the complaint scenarios, and 92 percent of those who received the no-complaintscenarios were able to correctly identify the content after reading the scenario (as towhether or not they voiced their complaints to the frontline employee). Similar resultswere found in the cases of compensation/no compensation: 90 percent of those whoreceived an apology and free mugs, and 87 percent of those who did not received anyapology or mugs were able to correctly identify the content (as to whether or notcompensation was offered). Taken together, these results suggest that our manipulationswere perceived as intended.
MeasuresAll multiple-item scales in this study were measured on a seven-point Likert scale.Satisfaction with the service provider’s handling of other-customer failure wasmeasured on a two-item scale (Wirtz and Mattila, 2004): “how did you feel about therestaurant on this particular occasion?” and “how satisfied were you be with thecompany’s handling of the incident caused by other patrons?” (r ¼ 0.80). NegativeWOM intention was measured by the answer to the following three questions(Wirtz and Mattila, 2004): “given what happened, how likely would you be to complainto your friends and relatives about this restaurant,” “given what happened, how likelywould it be that you tell your friends and relatives not to eat at this restaurant?,”
JOSM21,2
198
and “how likely would it be that you warn your friends and relatives not to eat at thisrestaurant?” (unlikely/likely) (Cronbach’s a ¼ 0.72). The repurchase intentionsmeasure included the following three items (Wirtz and Mattila, 2004): “because ofwhat happened, I will never go to this restaurant again,” “If this situation happened tome, I would never go to this restaurant again,” and “given what happened, I would visitthis restaurant again” (Cronbach’s a ¼ 0.87).
Several additional measures were included to ascertain whether the experimentalprocedures worked as intended. These included measures of how realistic the scenariowas, how easy it was for respondents to imagine themselves in the role of the customer,and what they thought the purpose of the research was. In addition, to assess potentialconfounds caused by demand artifacts that might result from the different scenarios,perceived severity of other-customer failure was measured using a three-item scale(Hess et al., 2003): “the noise caused by other patrons was a severe/minor/significantservice problem” (Cronbach’s a ¼ 0.90).
Validity of the experimental proceduresAnalysis showed that participants found the scenario to be realistic and therole-playing easy. The mean rating for scenario realism was 5.63 (with seven indicating“extremely realistic”). When asked to rate how easy it was to imagine themselves asthe customer on a seven-point scale, the mean rating was 5.40. There was no significantdifference ( p . 0.05) in terms of the realism and ease of role-playing among thedifferent treatment groups.
Respondents were asked to evaluate the severity of the other-customer failure.This was done to check whether there were differences in the perceptions of failuremagnitude between the groups, which might confound the results. A 2 £ 3 £ 2 ANOVAtesting was then conducted with perceived severity of other-customer failure as thedependent variable. The analysis revealed that there was no significant difference in therespondents’ severity perceptions for complaints, EE, and compensation.
ResultsCell means are presented in Table I. Because the three dependent variables,satisfaction, negative WOM, and repurchase intentions are correlated (Table II), the useof one MANOVA is more appropriate than the use of separate ANOVAs for eachdependent variable (Tabachnik and Fidell, 1996). In this case, the MANOVA controlsthe experimental error rate. However, separate ANOVAs still have to be used to findout which variables cause differences among the scenarios.
The MANOVA results showed significant main effects for complaints (Wilks’lambda ¼ 0.94, p , 0.001), EE (Wilks’ lambda ¼ 0.70, p , 0.001), and compensation(Wilks’ lambda ¼ 0.92, p , 0.001). No other interactions were significant ( p . 0.1 inall cases). Although none of the possible two- or three-way interactions among thefactors were significant, follow-up univariate analyses revealed significant interactionfor the one three-way interaction for satisfaction and one two-way interaction forrepurchase intentions. These interactions provided initial evidence of differentiationfor the survey conditions. The data were then subjected to separate analyses to testeach specific hypothesis.
H1. Testing for the main effects of complaints showed a significant main effect onthree dependent variables ( p , 0.05). Thus, when customers voiced their complaints
Other-customerfailure
199
about other-customer failure to the service provider, they were less satisfied with therestaurant (M ¼ 3.82 versus 4.17), less likely to offer future patronage (M ¼ 3.91versus 4.48), and more likely to indulge in negative WOM (M ¼ 4.03 versus 3.47) thanthose who did not voice their complaints. Therefore, H1a, H1b, and H1c weresupported.
H2. Perceived EE had significant main effects on the three dependent variables inthe expected directions ( p , 0.001). When consumers perceived employees to exertmore effort at solving the other-customer failure problem, they rated their level ofsatisfaction (M ¼ 4.80 versus 3.79 versus 3.39) and repurchase intention (M ¼ 5.31versus 3.87 versus 3.40) higher, and negative WOM (M ¼ 2.77 versus 4.04 versus4.41) lower than when there was low effort or no effort. The difference in ratings forsatisfaction, repurchase intention, and negative WOM between the low and no EE caseswas insignificant (p . 0.1). Thus, H2a, H2b, and H2c were supported.
H3. As can be seen in Table III, the complaint £ perceived EE £ compensationinteraction had a significant effect on satisfaction, F ¼ 3.56, p , 0.05. Thecomplaint £ perceived EE interaction was significant for repurchase intentions,F ¼ 5.59, p , 0.01. To test H3 directly, we contrasted the means of the “with” and“without compensation” cells for the various scenarios (Table I and Figure 1-3).As expected, we found significant differences between the “with” and “withoutcompensation” scenarios in the no complaint and low EE cases (M ¼ 4.73 versus3.46; t ¼ 3.65, p , 0.001 for satisfaction: M ¼ 4.84 versus 3.78; t ¼ 2.91, p , 0.01for repurchase intentions; and M ¼ 3.23 versus 4.20; t ¼ 22.71, p , 0.01 for
Complaint EE Compensation Satisfaction Negative WOM Repurchase intentions
Yes High With 5.13 (1.52) 3.15 (1.67) 5.00 (1.66)Without 4.04 (1.41) 3.24 (1.05) 4.60 (1.41)
Low With 3.56 (1.38) 4.43 (1.11) 3.52 (1.53)Without 3.39 (1.17) 4.42 (1.60) 3.35 (1.56)
No With 3.97 (1.44) 4.24 (1.31) 3.67 (1.71)Without 2.83 (1.19) 4.84 (1.18) 3.35 (1.27)
No High With 5.15 (1.63) 2.35 (0.97) 6.00 (0.87)Without 4.88 (1.35) 2.35 (1.09) 5.63 (0.97)
Low With 4.73 (1.08) 3.23 (0.91) 4.84 (1.18)Without 3.46 (1.33) 4.20 (1.47) 3.78 (1.41)
No With 3.69 (1.43) 4.12 (1.02) 3.44 (1.26)Without 3.08 (1.15) 4.57 (1.35) 3.16 (1.55)
Note: Standard deviations are given in parentheses
Table I.Cell means forsatisfaction, negativeWOM, and repurchaseintentions in Study 1
Variable 1 2 3
Satisfaction 1.00Negative WOM 20.61 * * 1.00Repurchase intention 0.59 * * 20.58 * * 1.00
Note: Significant at: *p , 0.05; * *p , 0.01Table II.Correlation matrix
JOSM21,2
200
negative WOM). However, we also found significant differences for the “with” and“without compensation” cases in the complaint with high EE cases (M ¼ 5.13 versus4.04; t ¼ 2.64, p , 0.05 for satisfaction), as well as in the complaint with no EE cases(M ¼ 3.97 versus 2.83; t ¼ 2.97, p , 0.01 for satisfaction). These results lend partialsupport to H3a. However, H3b and H3c were supported.
To summarize, the results of Study 1 showed that customers who voiced theircomplaints about other-customer failure to the service employees expressed lowerlevels of satisfaction, repurchase intentions, and higher unfavorable WOM than didthose who did not voice their complaints. When customers felt that employees hadexerted a great deal of effort to correct other-customer failure, they were more satisfied,more willing to offer repeat patronage and less willing to engage in negative WOM. Onthe other hand, when customers felt that the employees spent only a limited amount oftime and energy on solving the problem, they rated their level of satisfaction,repurchase intention low, and were likely to indulge in negative WOM, similar to thosewho received no help from the service employees. As expected, offering compensationin low EE scenarios increased non-complainant satisfaction, repurchase intention anddecreased their negative WOM intentions. We also found that offering compensationenhanced complainant satisfaction in the high and no EE scenarios.
Thus, far, we have focused on obtaining a favorable service outcome – that is, theproblem patron’s misbehavior discontinued after the employees’ high (or low or no)effort in managing the problem. However, if instead of a favorable outcome there was anegative outcome – that is, the problem patron’s misbehavior continued, would be aninteresting question from a theoretical point of view. In fact, this is often the casein many real situations. What difference would this make in the ratings of satisfactionand behavioral intentions for complainants and non-complainants? Study 2 wasdesigned to shed further light on our hypotheses in this type of negative situation.
Study 2Sample and proceduresThe stimuli, procedures, and measures employed in Study 2 were very similar to thoseemployed in Study 1. The only difference was that the problem patrons’ misbehaviorcontinued to disturb the affected customers until they finished their dinner and
MANOVA Univariate F
SourceWilks’lambda F df Satisfaction
NegativeWOM
Repurchaseintentions
Complaint (C) 0.94 5.60 * * 1 4.75 * 16.10 * * 11.80 * *
EE 0.70 18.43 * * 2 28.28 * * 49.26 * * 49.59 * *
Compensation (CPS) 0.92 7.74 * * 1 23.11 * * 5.94 * 7.06 * *
C £ EE 0.96 2.01 2 1.40 1.89 5.59 * *
C £ CPS 0.99 0.52 1 0.07 0.72 0.70EE £ CPS 0.93 0.92 2 0.15 1.14 0.33C £ EE £ CPS 0.97 1.26 2 3.56 * 1.60 0.87
Note: Significant at: *p , 0.05; * *p , 0.01
Table III.MANOVA and ANOVA
results for the dependentvariables in Study 1
Other-customerfailure
201
prepared to leave. Data were collected from 279 shoppers at another large shoppingcenter in Taiwan. Participants were assigned one of 12 scenarios: 2 (complaint versusno complaint) £ 3 (EE: high versus low versus no) £ 2 (compensation versus nocompensation). Cell sizes ranged from 22 to 25. The average age of participants
Figure 1.Interactive effects ofcomplaint, perceivedemployee effort andcompensation onsatisfaction
5.00
4.50
Satis
fact
ion 4.00
4.04
3.39
2.83
3.08
3.46
4.87
3.50
3.00
2.50
Complaint No complaint
Complaint
Without compensation
Employee effort
HighLowNo
5.00
4.50
Satis
fact
ion
4.00
3.50
5.13
3.56
3.97
3.69
4.73
5.15
Complaint No complaint
Complaint
Note: Study 1
With compensation
Employee effort
HighLowNo
JOSM21,2
202
was 40.3 years; 109 (39.1 percent) of the respondents were male, 170 (60.9 percent) werefemale; and the average working experience was 13 years. Furthermore, 60.7 percent had acollege degree or higher; and 84.2 percent reported they had been bothered by othercustomers in a restaurant in a way that was very similar to the story used in the scenarios.
Figure 2.Interactive effects ofcomplaint, perceivedemployee effort and
compensation onrepurchase intentions
6.00
5.50
Rep
urch
ase
inte
ntio
ns
5.00
4.60
3.35
3.353.16
3.78
5.62
4.50
4.00
3.00
3.50
Complaint No complaint
Complaint
Without compensation
Employee effort
HighLowNo
6.5
6.0
5.5
5.0
4.5
4.0
3.5
Rep
urch
ase
inte
ntio
ns
3.0
5.00
3.52
3.673.44
4.84
6.00
Complaint No complaint
Complaint
With compensation
Employee effort
HighLowNo
Note: Study 1
Other-customerfailure
203
ResultsCell means are presented in Table IV. The MANOVA (Table V) produced significantmain effects for EE (Wilks’ lambda ¼ 0.83, p , 0.001) and compensation (Wilks’lambda ¼ 0.93, p , 0.001). No other findings were significant.
Figure 3.Interactive effects ofcomplaint, perceivedemployee effort andcompensation on negativeWOM communications
5.00
4.50
Neg
ativ
e W
OM
4.00
3.24
4.42
4.84
4.57
4.20
2.35
3.50
3.00
2.00
2.50
Complaint No complaint
Complaint
Without compensation
Employee effort
HighLowNo
5.0
4.5
4.0
3.5
Neg
ativ
e W
OM
3.0
3.15
4.43
4.244.12
3.23
2.352.5
2.0
Complaint No complaint
Complaint
With compensation
Employee effort
HighLowNo
Note: Study 1
JOSM21,2
204
H1. The results were found to be inconsistent with this hypothesis, that is the maineffect of complaint on the three dependent variables was insignificant ( p . 0.1 in allcases); there was no significant difference detected between the complainants andnon-complainants satisfaction ratings (M ¼ 3.20 versus 3.33), repurchase intentions(M ¼ 3.33 versus 3.66), and negative WOM (M ¼ 4.58 versus 4.35). Thus, H1a, H1b,and H1c were not supported.
H2. When perceived EE was the independent variable it was found to have asignificant main effect on all three dependent variables ( p , 0.001 in all cases). In otherwords, when respondents perceived that the employees had exerted great effort insolving other-customer failure, they gave a higher rating to level of satisfaction(M ¼ 4.05 versus 3.04 versus 2.71), repurchase intentions (M ¼ 4.16 versus 3.35 versus2.99), and a lower rating to negative WOM (M ¼ 3.89 versus 4.64 versus 4.85), thanwhen they perceived low or no effort. Additionally, the difference in the ratings ofsatisfaction, repurchase intentions, and negative WOM between the low and no EEscenarios was insignificant ( p . 0.1). Therefore, H2a, H2b, and H2c were supported.
H3. To test H3, we contrasted the means of the “with” and “without compensation”groups for the various complaint and perceived EE cells (Table IV). We found the onlysignificant difference between the “with” and “without compensation” conditions to befor the no complaint with low EE cells (M ¼ 3.86 versus 2.27; t ¼ 3.96, p , 0.001
MANOVA Univariate F
SourceWilks’lambda F df Satisfaction
NegativeWOM
Repurchaseintentions
Complaint (C) 0.99 1.14 1 0.35 1.79 2.81EE 0.83 8.46 * * 2 24.57 * * 11.74 * * 15.91 * *
Compensation (CPS) 0.93 7.11 * * 1 18.56 * * 3.19 10.93 * *
C £ EE 0.97 1.39 2 0.50 2.16 0.28C £ CPS 0.99 0.22 1 0.64 0.16 0.74EE £ CPS 0.98 0.76 2 1.50 0.06 0.34C £ EE £ CPS 0.98 0.80 2 1.70 1.75 1.24
Note: Significant at: *p , 0.05; * *p , 0.01
Table V.MANOVA and ANOVA
results for the dependentvariables in Study 1
Complaint EE Compensation Satisfaction Negative WOM Repurchase intentions
Yes High With 4.26 (1.55) 3.74 (1.46) 4.29 (1.49)Without 3.51 (1.64) 4.29 (1.47) 3.74 (1.68)
Low With 3.33 (1.39) 5.01 (1.28) 3.17 (1.47)Without 2.67 (1.21) 4.93 (1.54) 3.04 (1.62)
No With 2.87 (1.46) 4.62 (1.47) 3.17 (1.61)Without 2.53 (1.42) 4.86 (1.13) 2.55 (1.22)
No High With 4.31 (1.53) 3.80 (1.46) 4.40 (1.44)Without 4.10 (1.22) 3.74 (1.78) 4.17 (1.80)
Low With 3.86 (1.39) 3.89 (1.26) 4.17 (1.27)Without 2.27 (1.33) 4.74 (1.30) 3.00 (1.45)
No With 3.00 (1.15) 4.80 (1.50) 3.43 (1.53)Without 2.40 (1.34) 5.12 (1.29) 2.77 (1.15)
Note: Standard deviations are given in parentheses
Table IV.Cell means for
satisfaction, negativeWOM, and repurchase
intentions in Study 2
Other-customerfailure
205
for satisfaction; M ¼ 4.17 versus 3.00; t ¼ 2.91, p , 0.01 for repurchase intentions, andM ¼ 3.89 versus 4.74; t ¼ 22.26, p , 0.05 for negative WOM). These findings suggestthat offering compensation had the strongest effect on non-complainants in the low EEscenario. Therefore, H3a, H3b, and H3c were supported.
General discussionFindings from prior marketing research show that other-customer misbehavior is amajor source of customer dissatisfaction and complaints. The goal of the present studyis to clarify what service firms and their employees can and should do to minimize theaffected customers’ dissatisfaction and negative WOM, so as to enhance the customersrepurchase intentions toward the firm where the other-customer failure had occurred.Several important findings and contributions related to marketing theory andreal-world practices can be drawn.
First, the outcome valence (other-customer failure discontinues as opposed tocontinues) has a strong impact on how complainants and non-complainants rate theirsatisfaction and subsequent behavioral intentions. Customers who had voiced theircomplaints expressed lower levels of satisfaction and repurchase intention, and wereless likely to indulge in negative WOM than did those who did not voice theircomplaints if other-customer failure discontinued (favorable outcome in Study 1).In contrast, when other-customer failure continued (unfavorable outcome in Study 2),both complaining and non-complaining customers gave similar low scores forsatisfaction, repatronage intentions, and high scores for negative WOM. The effects ofoutcome valence on consumer service evaluations have been documented in a numberof marketing studies (Brady et al., 2006; Parasuraman et al., 1985), however, to the bestof our knowledge, this is the first study to highlight the difference that outcome valencemakes on service evaluations for complainants and non-complainants in the context ofother-customer failure.
Second, although firms might view the misbehavior of other patrons asuncontrollable, our findings show that consumers consider management ofother-customer misbehavior an important component of the service process, and thisaffects their overall service evaluations (Bitner et al., 1990). When participants felt theemployees exerted a great deal of effort to help solve such problems, they felt morehighly satisfied, more willing to come back, and less willing to engage in negativeWOM. On the other hand, when participants felt that the employees had not made agood-faith effort, they offered low scores for satisfaction and repurchase intention, andhigh scores for negative WOM, similar to those who did not receive any help from theservice employees. These findings are important and should be distinguished fromthose arrived at in previous studies. In prior research, the different degrees of EE(between high and low) was manipulated and its effect on service evaluations wasinvestigated (Mohr and Bitner, 1995; Sarel and Marmorstein, 1999; Specht et al., 2007).Our study included “no EE” in the experimental design. This kind of research designnot only more accurately reflects real world situations, but also provides more preciseinformation about customer feelings and perceptions when they suffered fromother-customer failure and received (high, low- or no) effort from employees.
Third, as hypothesized, our results show that offering compensation is not thecure-all. This conclusion is consistent with previous service failure and recoveryresearch findings (McCollough et al., 2000; Wirtz and Mattila, 2004). However, our
JOSM21,2
206
results further indicate that offering compensation was only effective in increasingsatisfaction and repurchase intentions, and diminishing negative WOM fornon-complainants in the low EE condition. In short, offering compensation had thestrongest impact on non-complaining customers who received unsolicited help from theservice employees. These findings illustrate that the influence of compensationon post-recovery satisfaction and behavior intentions was largely dependent oncustomer perceptions of employee-exerted efforts for other-customer failure. When EEwas perceived as high, offering compensation seemed to be unnecessary, since it did notfurther enhance satisfaction and favorable behavioral intentions (a ceiling effect).In addition, when EE was perceived as absent, the positive effect of compensation waslessened, that is it did not increase satisfaction and favorable behavioral intentionstoward the firm.
To sum up, in this study we examined the importance of the as yet under-researchedissue of how organizational recovery responses to other-customer failure influence acustomer’s service evaluations of the firm. These findings had not been studiedpreviously, because in much of past work, the focus has been on the antecedentsand classifications of dysfunctional customer types (Grove and Fisk, 1997; Harris andReynolds, 2003), the negative effect of customer interaction on the service experienceand satisfaction (Bitner et al., 1994; Martin, 1996), and why and how theother-customer’s misbehavior influences customer satisfaction toward the firm(Huang, 2008). The primary contribution of this study to the service marketingliterature is that it provides empirical results that shed some light on the responses ofboth complainants and non-complainants to organizational recovery efforts after theoccurrence of an other-customer failure problem. These findings can assist marketersto build better recovery strategies when other-customer failure occurs.
Managerial implicationsSeveral implications for service firms can be drawn from the results of this study. First,our findings suggest that rather than accepting customer-to-customer influences asinevitable, service firms should actively manage customer encounters to ensure that alltheir customers behave appropriately (Martin, 1996). In case of other-customer failure,firms need to react immediately, prior to customer complaints. If they fail to correct theproblem, both complaining and non-complaining customers become highlydissatisfied, and will likely choose to abandon their service relationships with thisfirm. Even though complainants may express lower satisfaction and unfavorablebehavioral intentions than non-complaints, it is still beneficial to encourage dissatisfiedcustomers to voice their complaints, because this gives marketers an opportunity tomake amends, and to identify the root causes of other-customer failure.
Since customer service satisfaction is affected by how employees react to disruptiveother-customer situations, training employees with the appropriate problem-solvingskills is a key issue for service marketers. This can be done by teaching the employeeto show empathy to the affected customer and to be persistent and willing to spendtime trying to resolve the problem. In addition, our findings suggest that if EEs incases of other-customer failure were judged to be insufficient to really help, thisaggravated rather than appeased the affected persons. Thus, being perceived ashaving made a real effort is also critical. Marketers need to design training sessionsthat can be used to explore ways to improve or modify behaviors that contribute
Other-customerfailure
207
to the impressions of not making a real effort to serve (Sarel and Marmorstein, 1999).When a customer believes that the employee has done everything possible to solve another-customer failure problem, their reaction will be much less negative.
It is clear that compensation is not a substitute for a good recovery process. Servicemanagers need to design well-balanced service recovery systems, in terms of both theoutcome and the process. An excellent other-customer failure recovery program needsto emphasize both the outcome of recovery (i.e. compensation) and the way in whichthe recovery process is delivered (i.e. EE).
Future research and limitationsThere are numerous opportunities for future research in this area, some of which aremade evident by the limitations of this study. For example, to maximize internalvalidity, hypothetical scenarios rather than an actual consumption experience wereused as stimuli and the setting involved only a single service category. In futurestudies, other categories and a natural setting could be included. Moreover, our datawere collected in Taiwan, which raises the question of the transferability of ourfindings to other cultural regions. The role of culture in the consumer’s failureattributions and service recovery perceptions needs to be examined, perhaps withrespondents drawn from both individualistic and collectivist cultures. Third, priorevidence suggests that the relative effectiveness of service recovery appears to besituation specific (Ruyter and Wetzels, 2000; Smith et al., 1999). The failure type orthe magnitude of other-customer failure is likely to influence customer evaluationsfor organizational recovery response. For example, Smith et al. (1999) found thatcustomers prefer to receive tangible compensation in the case of outcome failures – theunavailability of a service (e.g. the unavailability of a reserved hotel room because ofoverbooking), but an apology and speedy response in the case of process failures – theinadequate delivery of services (e.g. a hotel desk clerk treats a customer rudely duringthe check-in process). Future research should be done to investigate how the context ofother-customer failure interacts with a firm’s recovery efforts.
References
Bitner, M.J. (1990), “Evaluating service encounters: the effects of physical surroundings andemployee responses”, Journal of Marketing, Vol. 54 No. 2, pp. 69-82.
Bitner, M.J., Booms, B.H. and Mohr, L.A. (1994), “Critical service encounters: the employee’sviewpoint”, Journal of Marketing, Vol. 58 No. 4, pp. 95-106.
Bitner, M.J., Booms, B.H. and Tetreault, M.S. (1990), “The service encounter: diagnosingfavorable and unfavorable incidents”, Journal of Marketing, Vol. 54 No. 1, pp. 71-84.
Blodgett, J.G. and Anderson, R.D. (2000), “A Bayesian network model of the consumer complaintprocess”, Journal of Service Research, Vol. 2 No. 4, pp. 321-38.
Blodgett, J.G., Hill, D.J. and Tax, S.S. (1997), “The effects of distributive, procedural, andinteractional justice on postcomplaint behavior”, Journal of Retailing, Vol. 73 No. 2,pp. 185-210.
Bolfing, C.P. (1989), “How do customer express dissatisfaction and what can service marketersdo about it?”, Journal of Service Marketing, Vol. 3 No. 2, pp. 5-23.
Brady, M.K., Voorhees, C.M., Cronin, J. Jr and Bourdeau, B.L. (2006), “The good guys don’talways win: the effect of valence on service perceptions and consequences”, Journal ofServices Marketing, Vol. 20 No. 2, pp. 83-91.
JOSM21,2
208
Davidow, M. and Leigh, J.H. (1998), “The effects of organizational complaint responses onconsumer satisfaction, word of mouth activity and repurchase intentions”, Journal ofConsumer Satisfaction, Dissatisfaction and Complaining Behavior, Vol. 11, pp. 91-102.
Fullerton, R.A. and Punj, G. (1993), “Choosing to misbehave: a structural model of aberrantconsumer behavior”, Advances in Consumer Research, Vol. 20, pp. 570-4.
Fullerton, R.A. and Punj, G. (1997), “What is consumer misbehavior?”, Advances in ConsumerResearch, Vol. 24, pp. 336-9.
Gronroos, C. (1990), Service Management and Strategy: Marketing the Moments of Truth inService Competition, Lexington Books, Lexington, MA.
Grove, S.J. and Fisk, R.P. (1997), “The impact of other customers on service experiences: a criticalincident examination of ‘getting along’”, Journal of Retailing, Vol. 73 No. 1, pp. 63-85.
Grove, S.J., Fisk, R.P. and Dorsch, M.J. (1998), “Assessing the theatrical components of the serviceencounter: a cluster analysis examination”, The Service Industries Journal, Vol. 18 No. 3,pp. 116-34.
Guenzi, P. and Pelloni, O. (2004), “The impact of interpersonal relationships on customersatisfaction and loyalty to the service provider”, International Journal of Service IndustryManagement, Vol. 15 No. 4, pp. 365-84.
Halstead, D. (2002), “Negative word of mouth: substitutes for or supplement to consumercomplaints?”, Journal of Consumer Satisfaction, Dissatisfaction, and ComplainingBehavior, Vol. 15, pp. 1-12.
Harris, L.C. and Reynolds, K.L. (2003), “The consequences of dysfunctional customer behavior”,Journal of Service Research, Vol. 6 No. 2, pp. 144-61.
Harris, L.C. and Reynolds, K.L. (2004), “Jaycustomer behavior: an exploration of types andmotives in the hospitality industry”, Journal of Service Marketing, Vol. 18 No. 5, pp. 339-57.
Hess, R.L., Ganesan, S. and Klein, N.M. (2003), “Service failure and recovery: the impact ofrelationship factors on customer satisfaction”, Journal of the Academy of MarketingScience, Vol. 31 No. 2, pp. 127-45.
Hirschman, A. (1970), Exit, Voice and Loyalty, Harvard University Press, Cambridge, MA.
Hoffman, K.D., Kelley, S.C. and Rotalsky, H.M. (1995), “Tracking service failures and employeerecovery efforts”, Journal of Services Marketing, Vol. 9 No. 2, pp. 49-61.
Huang, W.H. (2008), “The impact of other-customer failure on service satisfaction”, InternationalJournal of Service Industry Management, Vol. 19 No. 4, pp. 521-36.
Hui, M.K. and Bateson, J.E.G. (1991), “Perceived control and the effects of crowding andconsumer choice on the service experience”, Journal of Consumer Research, Vol. 18 No. 2,pp. 174-84.
Kirmani, A. and Rao, A.R. (2000), “No pain, no gain: a critical review of the literature on signalingunobservable product quality”, Journal of Marketing, Vol. 64 No. 2, pp. 66-79.
Lovelock, C.H. (1994), Product Plus: How Product and Service Equals Competitive Advantage,McGraw-Hill, New York, NY.
McCollough, M.A., Berry, L.L. and Yadav, M.S. (2000), “An empirical investigation of customersatisfaction after service failure and recovery”, Journal of Service Research, Vol. 3 No. 2,pp. 121-37.
Martin, C.L. (1996), “Consumer-to-consumer relationships: satisfaction with other consumers’public behavior”, The Journal of Consumer Affairs, Vol. 30 No. 1, pp. 146-69.
Martin, C.L. and Pranter, C.A. (1989), “Compatibility management: customer-to-customerrelationships in service environments”, Journal of Service Marketing, Vol. 3 No. 3, pp. 6-15.
Other-customerfailure
209
Mattila, A.S. and Patterson, P.G. (2004), “The impact of culture on consumers’ perceptions ofservice recovery efforts”, Journal of Retailing, Vol. 80 No. 3, pp. 196-206.
Mohr, L.A. and Bitner, M.J. (1995), “The role of employee effort in satisfaction with servicetransactions”, Journal of Business Research, Vol. 32 No. 3, pp. 239-52.
Moore, R., Moore, M.L. and Capella, M. (2005), “The impact of customer-to-customer interactionsin a high personal contact service setting”, Journal of Service Marketing, Vol. 19 No. 7,pp. 482-91.
Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1985), “A conceptual model of service qualityand its implications for future research”, Journal of Marketing, Vol. 49 No. 4, pp. 41-50.
Raajpoot, N.A. and Sharma, A. (2006), “Perceptions of incompatibility in customer-to-customerinteractions: examining individual level differences”, Journal of Service Marketing, Vol. 20No. 5, pp. 324-32.
Ruyter, K. and Wetzels, M. (2000), “Customer equity considerations in service recovery:a cross-industry perspective”, International Journal of Service Industry Management,Vol. 11 No. 1, pp. 91-108.
Sarel, D. and Marmorstein, H. (1999), “Managing the delayed service encounter: the role ofemployee action and customer prior experience”, International Journal of Bank Marketing,Vol. 17 No. 6, pp. 286-94.
Singh, J. (1990), “A typology of consumer dissatisfaction response styles”, Journal of Retailing,Vol. 66 No. 1, pp. 57-99.
Smith, A.K., Bolton, R.N. and Wagner, J. (1999), “A model of customer satisfaction with serviceencounters involving failure and recovery”, Journal of Marketing Research, Vol. 36 No. 3,pp. 356-72.
Specht, N., Fichtel, S. and Meyer, A. (2007), “Perception and attribution of employees’ effort andabilities”, International Journal of Service Industry Management, Vol. 18 No. 5, pp. 534-54.
Stephens, N. and Gwinner, K.P. (1998), “Why don’t some people complain? A cognitive-emotiveprocess model of consumer complaint behavior”, Journal of the Academy of MarketingScience, Vol. 26 No. 3, pp. 172-89.
Tabachnik, B.G. and Fidell, L.S. (1996), Using Multivariate Statistics, Harper Collins College,New York, NY.
Tax, S.S. and Brown, S.W. (1998), “Recovering and learning from service failure”,Sloan Management Review, Vol. 40 No. 1, pp. 75-88.
Tax, S.S., Brown, S.W. and Chandrashekaran, M. (1998), “Customer evaluations of servicecomplaint experiences: implications for relationship between delays and evaluations ofservice”, Journal of Marketing, Vol. 58 No. 2, pp. 56-69.
Voorhees, C.M., Brady, M.K. and Horowitz, D.M. (2006), “A voice from the silent masses:an exploratory and comparative analysis of noncomplainers”, Journal of the Academy ofMarketing Science, Vol. 34 No. 4, pp. 514-27.
Wirtz, J. and Mattila, A.S. (2004), “Consumer responses to compensation, speed of recovery andapology after a service failure”, International Journal of Service Industry Management,Vol. 15 No. 2, pp. 150-66.
Wu, C. (2007), “The impact of customer-to-customer interaction and customer homogeneity oncustomer satisfaction in tourism service – the service encounter prospective”, TourismManagement, Vol. 28 No. 6, pp. 1518-28.
Zemke, R. and Anderson, K. (1990), “Customers from hell”, Training, Vol. 27 No. 2, pp. 25-33.
JOSM21,2
210
Appendix. Example scenario: complaint, high effort, failure discontinues, andcompensationIn the following scenario, we are interested in understanding what consumers think aboutrestaurant services. Imagine that the following incident happened to you during your visit to arestaurant. Read the scenario carefully and answer the questions below.
You have decided to go out with your friends for a relaxing dinner on a Friday evening.After entering the restaurant, a hostess seats you near the window. You find the atmosphere inthe restaurant to be a perfect blend of comfort and tranquility.
After a short period, your meal is served. While you are enjoying the delicious food andchatting with your friends about work and life, it gradually comes to your notice that your voicesare being smothered by loud noise from an adjacent table. There are four rowdy and boisterousyoung people seated there who do not seem to mind that you and some other patrons are glaringat them.
The noise makes you uncomfortable and you feel that your wonderful night is ruined.You make a complaint to the frontline employee at the counter. Soon after your complaint, awaitress approaches the problem patrons and politely says to them “Sorry, Sirs, but could youplease keep your voices down?” The waitress tries three separate times to persuade these patronsto be quieter and finally, the normal peaceful atmosphere of the restaurant is restored.
After dinner, you prepare to leave. As you are paying the bill, the restaurant managerapologizes to you and tells you that, to compensate for your bad experience and inconveniencecaused by the noise of the other customers, they would like to offer you two gift mugs.
Corresponding authorWen-Hsien Huang can be contacted at: [email protected]
To purchase reprints of this article please e-mail: [email protected] visit our web site for further details: www.emeraldinsight.com/reprints
Other-customerfailure
211
