Effects of service failure on consumer responses across failure types: A moderating role of intimacy

ARTICLE IN PRESS Australasian Marketing Journal ■■ (2015) ■■–■■

Contents lists available at ScienceDirect

Australasian Marketing Journal j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a m j

Effects of service failure on consumer responses across failure types: A moderating role of intimacy Seongun Jeon a, Jonathan S. Kim b,* a b

Business School, Seoul National University, 1, Gwanak-ro, Gwanak-gu, 151-916 Seoul, Republic of Korea Business School, Hanyang University, #412, 17 Haengdang-dong, Seongdong-gu, Seoul, Republic of Korea

A R T I C L E

I N F O

Article history: Accepted 14 December 2015 Available online Keywords: Intimacy Service failure types Buffering effect Betrayal effect

A B S T R A C T

This paper investigates a moderating role of intimacy in two service failure types (outcome and process failure), considering two consumer responses: buffering and betrayal effect. Study 1 employs a 3 (failure type: no vs. outcome vs. process) × 2 (intimacy level: high vs. low) experimental design. Findings show that in case involving outcome failure, a high intimacy group has higher service evaluations (satisfaction, re-patronage intention, and positive word-of-mouth intention) than a low intimacy group, substantiating the existence of buffering effect. In case involving process failure, however, service evaluations are not different depending on the level of intimacy, invalidating the existence of betrayal effect. Study 2 analyzes the effects of intimacy (high vs. low) on service evaluations in the setting of process failure and following failed recovery. The results reveal that betrayal effect is indeed present in times of double deviation, process failure and following failed recovery. Consequently, this research offers service providers practical insights on how to utilize intimacy based on the classification of service failure types. © 2015 Australian and New Zealand Marketing Academy. Published by Elsevier Ltd. All rights reserved.

1. Introduction “You don’t bring me flowers, you don’t send me flowers, you don’t sing me love songs, you hardly talk to me anymore …” This is how a popular love song by Barbara Streisand and Neil Diamond begins. As Barbara and Neil sing, what was so natural in the past counts no more when the relationship is taken for granted. It even applies to the business landscape where service providers often fall in the trap of neglecting regular customers (Blount, 2010). Unfortunately, service providers often forget a well-publicized principle that an increasing customer retention rates by 5% increases profits by 25% to 95% (Frederick and Sasser, 1990). The underlying cause of such an unfortunate event lies in service providers’ excessive reliance on intimacy. Service providers tend to maintain a high level of intimacy with their customers hoping that small mistakes can be understood. However, they get in the habit of asking for tolerance regardless of how customers actually feel. Intimacy is literally the double-edged sword. Consumers can either understand the service failure or feel hugely disappointed about it when they are intimate with their service providers (Jeon et al., 2013; Park, 2007). The former is buffering effect, mitigating feelings thanks to the high level of intimacy (Park, 2007); whereas the latter is betrayal effect, magnifying feelings due to the high level of intimacy (Holloway et al., 2009; Jeon et al., 2013). Such effects can explain

* Corresponding author. Tel.: +82 2 2220 1041; fax: +82 2 2220 1169. E-mail address: [email protected] (J.S. Kim).

conflicting consumer responses after service failure but have not been studied in depth yet. Specifically, previous research has failed to uncover in which circumstances the two different consumer responses, buffering and betrayal effect, would occur. It is thus important to specify service failure types and apply them to the research on the influence of intimacy. Using the categorization from the prior research, we aim to empirically analyze how consumers maintaining different levels of intimacy respond to two types of service failure: outcome and process failure (Smith et al., 1999). Given that when a customer has a high level of intimacy with a certain service provider, the influence of outcome failure that frequently occurs in normal transactions would be mitigated while the impact of process failure that is accompanied by negative emotions would be magnified. Consequently, this research is expected to offer service providers practical insights on when to relax or worry while making use of intimacy based on the classification of service failure types. 2. Theoretical background 2.1. Intimacy Intimacy is one of the most important factors that significantly influence consumer responses but has rarely been explored in the field of marketing. So far, intimacy was defined to be the disclosure of information shared only within that specific relationship (Hansen, 2003). More recently, Brock and Zhou (2012) focused on high levels of closeness and mutual understanding between

http://dx.doi.org/10.1016/j.ausmj.2015.12.004 1441-3582/© 2015 Australian and New Zealand Marketing Academy. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Seongun Jeon, Jonathan S. Kim, Effects of service failure on consumer responses across failure types: A moderating role of intimacy, Australasian Marketing Journal (2015), doi: 10.1016/j.ausmj.2015.12.004

ARTICLE IN PRESS S. Jeon, J.S. Kim/Australasian Marketing Journal ■■ (2015) ■■–■■

2

Table 1 Initial classification of service failure types. Group 1.

Group 2.

Group 3.

Employee response to service delivery system failures

Employee response to customer needs and requests

Unprompted and unsolicited employee actions

1A. Response to unavailable service 1B. Response to unreasonably slow service 1C. Response to other core service failures – –

2A. Response to “special needs” customers 2B. Response to customer preference 2C. Response to admitted customer error 2D. Response to potentially disruptive others –

3A. Attention paid to customer 3B. Truly out-of-the-ordinary employee behavior 3C. Employee behaviors in the context of cultural norms 3D. Gestalt evaluation 3E. Exemplary performance under adverse circumstances

(‘–’: non-exist). Source: Bitner et al. (1990).

customers and suppliers to define intimacy. However, we will adopt following definition from prior research: Intimacy is a customer’s willingness to share personal information and emotion as well as expectations for future transactions based on a long-term relationship with a service provider. Jeon et al. (2013)

This paper will further focus on social and process/psychological intimacy among five types: no intimacy, social intimacy, physical intimacy, process/psychological intimacy, and sexual intimacy (Beetles and Harris, 2009) for its particular service setting, the hair industry. Previous research has shown that intimacy works as a critical moderator when customers experience service failure. Jeon et al. (2013) found that customers were more likely to be disappointed with service failure once they maintain a high level of intimacy with their service providers. Accordingly, service providers were asked to prevent service failure from the beginning as close customers were most satisfied in the absence of service failure. In reality, however, it is almost impossible to prevent the entire service failure as service is meant to be out-of-control for its characteristics (Parasuraman et al., 1985; Wolak et al., 1998) and its external factors (e.g., weather or customer’s ever-changing emotion) being out of control (Tax et al., 1998). Meanwhile, albeit not exactly studied in terms of intimacy, Park (2007) revealed that customers were more thoughtful in handling service failure based on the rapport accumulated through close relationships. That is, service providers could focus on building great rapport with their customers to protect themselves from the detrimental effect of service failure. It is still unlikely that customers would forgive service failure regardless of its kind and severity. Therefore, we should ultimately approach the research on the influence of intimacy by specifying service failure types. We will subdivide the service failure types into outcome and process failure, in which the high level of intimacy between customers and service providers would result in different consumer responses. 2.2. Service failure types Research on service failure types has consistently existed since 1990s. Initially, general types of service failure were analyzed from a customer perspective across hotel, airline, and restaurant industry based on the three groups: employee response to service delivery system failures, employee response to customer needs and requests, and unprompted and unsolicited employee actions. (Bitner et al., 1990) (see Table 1). Following research on service failure (Forbes et al., 2005; Hoffman et al., 1995, 2003; Kelley et al., 1993) was carried out adopting the initial classification scheme by Bitner et al. (1990). Subsequently, classification schemes of retail failures were developed into 11 types (Kelley et al., 1993), and typologies of service

failures in the restaurant industry were identified into 15 types (Hoffman et al., 1995). Then, previous service failure types were grouped into outcome and process failure. Outcome failure is the situation when service providers consequently fail to provide the service that customers want, while process failure means the incident when service providers are flawed in the process of providing service (Smith et al., 1999). More recently, typologies of servicescape failures (Hoffman et al., 2003) and e-commerce/e-tail service failures (Forbes et al., 2005) were also identified. The entire finding is reorganized in Table 2. The standard of categorization combines overlapping failure types such as ‘employee attention failure’ and ‘inappropriate behaviors’ into a single category, ‘employee attitude/ behavior problem’. As shown in Table 2, previous failure types were hardly generalized as they varied depending on the type of service industry until the development of outcome and process failure. The example of outcome failure is the financial loss such as ‘3A. Mischarged’ as a result of employee pricing error (Hoffman et al., 1995; Kelley et al., 1993) or ‘1C. System pricing’ failure due to incorrect pricing information contained in the scanner pricing system (Kelley et al., 1993) and website error (Forbes et al., 2005). Another example of outcome failure is the loss in goods as in the case of ‘3E. Wrong order’, delivery of wrong product or packaging of an incorrect fast food item (Hoffman et al., 1995). This type also includes ‘3F. Lost order’, the customer’s order being misplaced and never fulfilled (Hoffman et al., 1995). On the other hand, process failure is typically accompanied by the loss in symbolic resources such as self-esteem or status. The example of process failure is ‘inattentive service’ that is representative in two different service settings, restaurants and hotels (Smith et al., 1999). Another example of process failure is ‘2A. Special order/ request’ in which service providers fail to cope with customers’ special demands (Forbes et al., 2005; Hoffman et al., 1995, 2003; Kelley et al., 1993), which were present in all of the four studies listed (see Table 2). 2.3. Service evaluations Service evaluations are generally measured in five constructs: sacrifice, service quality, service value, customer satisfaction, and behavioral intentions (Brady et al., 2005). Given that satisfaction is a significant antecedent to behavioral intentions (Brady et al., 2005) and that re-patronage intention and positive word-of-mouth intention pertain to behavioral intentions, service evaluations are due to be analyzed in three dimensions: satisfaction, re-patronage intention, and positive word-of-mouth intention. These are also known to be typically correlated and consistent (Lopez-Lopez et al., 2014). Although a positive relationship is thought to exist between a firm’s service recovery effort and service evaluations (Spreng et al., 1995), this paper initially takes into account how consumers react to service failure based on the level of intimacy with service providers before service recovery is provided. Consumers are expected to show consistency in their reactions.

Please cite this article in press as: Seongun Jeon, Jonathan S. Kim, Effects of service failure on consumer responses across failure types: A moderating role of intimacy, Australasian Marketing Journal (2015), doi: 10.1016/j.ausmj.2015.12.004

ARTICLE IN PRESS S. Jeon, J.S. Kim/Australasian Marketing Journal ■■ (2015) ■■–■■

3

Table 2 Classification of research on service failure types. Failure types Group 1. Employee response to service delivery system/product failure 1A. 1B. 1C. 1D. 1E. 1F. 1G. 1H. 1I. 1J. 1K. Group 2. Response to customer needs and requests 2A. 2B. 2C. 2D. Group 3. Unprompted and unsolicited employee actions 3A. 3B. 3C. 3D. 3E. 3F. 3G.

Kelley et al. (1993)

Hoffman et al. (1995)

Hoffman et al. (2003)

Forbes et al. (2005)

Policy failure Slow/unavailable service System pricing Packaging errors Out of stock Product/service defect Hold disaster Alterations and repairs Bad information Facility/servicescape problems Website system failure

Y Y Y Y Y Y Y Y Y – –

Y Y – – Y Y – – – Y –

Y Y – – Y Y – – – Y –

– Y Y Y Y Y – – Y – Y

Special order/request Customer error Potentially disruptive others Size variation

Y Y – –

Y – – –

Y Y Y –

Y Y – Y

Mischarged Accused of shoplifting Embarrassments (e.g., violation of cultural norms and discrimination) Employee attitude/behavior problem Wrong order Lost order Employee incompetency

Y Y Y

Y – Y

– – Y

– – –

Y – – –

Y Y Y –

Y – – Y

– – – –

(‘Y’: exist, ‘–’: non-exist).

Satisfaction refers to an individual’s positive evaluation of consumption experience, which is subjectively derived (Westbrook, 1980a). Satisfaction can be measured by a function of contentment with the contact person, core service, and institution (Crosby and Stephens, 1987). Oliver (1997) emphasized that satisfaction relies on customers’ subjective judgment or evaluation, which is why it is difficult to be guaranteed. Considering that satisfaction is even more important than service quality in influencing the purchase intention (Cronin and Taylor, 1994), the high level of satisfaction is expected to increase re-patronage intention. In addition, the failure to deal with dissatisfied customers will also result in negative wordof-mouth, thereby losing both existing and potential customers (Yi, 2013). Re-patronage intention is the individual’s determination to repurchase a certain product/service from the same company based on his/her current state and possible circumstances (Hellier et al., 2003). Marketers are well aware that it costs much less to retain old customers than to attract new customers (Gremler and Brown, 1998; Spreng et al., 1995). They also know that satisfaction is a prerequisite for re-patronage intention (Yi, 1990). Accordingly, marketers endeavor to satisfy consumers in order to retain them in the long run. The word-of-mouth is the process by which information about products, services, stores, companies, etc. spreads from one consumer to another. While word-of-mouth can be either positive or negative, marketers attempt to promote positive word-of-mouth which can be defined as “making others aware that one does business with a company or store, making positive recommendations to others about a company, and extolling a company’s quality orientation” (Brown et al., 2005). Collier (1995) reported that customers with negative service experience told nine or ten individuals about their poor service experience, while satisfied customers talked to at most four or five individuals about their good service experience. The useful information gained this way helps customers to decide whether or not to re-patronage a firm (Zeithaml et al., 1993).

Customers are sometimes motivated to a carry out a brand switch based on negative word-of-mouth about a firm through which other firms gain new customers (Maxham, 2001). Undoubtedly, it is more important than ever to promote positive word-of-mouth. 3. Hypotheses As noted earlier, it is required to classify service failure types to overcome the limitation of previous studies (Jeon et al., 2013; Park, 2007) which merely focuses on how intimacy influences customer reaction to service failure without considering specific service failure types. Thus, we expect that three different service failure types (no vs. outcome vs. process) moderated by the level of intimacy (high vs. low) would result in different service evaluations (satisfaction, re-patronage intention, and positive word-of-mouth intention). No failure type is considered a default condition. In case of outcome failure that frequently occurs in normal transactions, we expect buffering effect where a high level of intimacy between customers and service providers mitigates the negative service evaluations. We also postulate that the three dimensions of service evaluations (satisfaction, re-patronage intention, and positive word-of-mouth intention) will be consistent as in Lopez-Lopez et al. (2014). Thus, we posit the following: H1. In case involving outcome failure, a high intimacy group will have higher service evaluations: a. satisfaction, b. re-patronage intention, and c. positive word-of-mouth intention than a low intimacy group. In case of process failure, however, we expect betrayal effect where the high level of intimacy magnifies the negative emotion experienced through service failure, given that customers who experience negative emotion after service encounter tend to have negative service evaluations (Burns and Neisner, 2006; Menon and Dubé, 2000; Smith and Bolton, 2002; Taylor, 1994; Van Dolen et al.,

Please cite this article in press as: Seongun Jeon, Jonathan S. Kim, Effects of service failure on consumer responses across failure types: A moderating role of intimacy, Australasian Marketing Journal (2015), doi: 10.1016/j.ausmj.2015.12.004

ARTICLE IN PRESS S. Jeon, J.S. Kim/Australasian Marketing Journal ■■ (2015) ■■–■■

4

2001; Zeelenberg and Pieters, 2004). We also posit that the three dimensions of service evaluations (satisfaction, re-patronage intention, and positive word-of-mouth intention) will be consistent as in Lopez-Lopez et al. (2014). Thus, we hypothesize the following: H2. In case involving process failure, a high intimacy group will have lower service evaluations: a. satisfaction, b. re-patronage intention, and c. positive word-of-mouth intention than a low intimacy group. 4. Study 1 The purpose of our first study is to examine the influence of service failures on customers’ service evaluations based on the moderating role of intimacy. We attempt to confirm that consumers are more likely to understand outcome failure but feel betrayed by process failure in case they are highly intimate with service providers. 4.1. Methods 4.1.1. Sample and procedure We originally obtained data from a sample of 204 respondents, but later eliminated 18 questionnaires which had incomplete or missing responses, or which failed to pass reality assessment (ratings less than four). Participants for this study were undergraduate and graduate students taking a marketing course during a fall semester at a major university in Seoul, South Korea. Respondents were mostly in their 20s (92.9%) and 51.6% of them were male. The subjects were told to fill out the questionnaires after reading a given scenario describing intimacy level and service failure type in the setting of hair industry. The present research employed a 3 × 2 between-subjects design: service failure type (no vs. outcome vs. process) × intimacy level (high vs. low). Scenarios contained representative examples of failure types from previous research. Examples of outcome failure included ‘lost order’, the coating service originally provided during a perm but never fulfilled (Hoffman et al., 1995), and ‘financial loss’, mischarge due to employee pricing error (Hoffman et al., 1995; Kelley et al., 1993). By contrast, process failure involved ‘inattentive service’, the case in which other customers who arrived later received hairstyling service earlier (Smith et al., 1999), and ‘special order/ request’, the case in which hairstylists repeatedly failed to cope with a regular customer asking for a cup of tea during hairstyling service (Forbes et al., 2005; Hoffman et al., 1995, 2003; Kelley et al., 1993). No failure type was given as a baseline. In addition, a high intimacy situation was depicted as a regular customer constantly receiving hairstyling service from a same hairdresser for a long time. In particular, we described that the customer shared his/her private issues with the hairdresser. On the other hand, a low intimacy situation included a customer who rarely cared about from whom he/she received hairstyling service (see Appendix A). 4.1.2. Measures Each construct was measured on a 7-point Likert type scale anchored by ‘strongly disagree (1)’ and ‘strongly agree (7)’. Reality was measured to confirm whether respondents accepted the given scenario as realistic. Service failure types were also measured to determine whether or not respondents perceived the severity of each scenario (outcome failure vs. process failure situation) similarly. Next, a four-item measure of intimacy, one’s will to share inner feelings in addition to anticipation for transactions to come, was constructed specifically for a hairstyling service (Aaker et al., 2004). They were “I am comfortable sharing detailed information about myself with the hairdresser,” “The hairdresser really understands my preference for hairstyles,” “I am familiar with the range of services this

hair salon offers,” and “I feel comfortable describing what kind of service this hair salon provides to a first-time visitor.” In addition, service evaluations were specifically measured as follows. A three-item measure of satisfaction, a state of being fond of service experience in comparison to prior expectation, was modified for this study with items in the extant literature (Brady et al., 2005; Fornell et al., 1996; Gustafsson et al., 2005; Hui et al., 2004; Maxham, 2001; Voss et al., 1998; Westbrook, 1980b). Items included: “I feel pleased with what has been done in the beauty salon,” “I am delighted with the service provided,” and “The hairdresser provided a satisfactory hairstyling service.” Re-patronage intention, one’s willingness to reuse a product/service from the same company, was measured using a three-item scale adapted from prior research (Berens et al., 2005; Maxham, 2001). Items included: “I am planning to visit this beauty salon again,” “I will continue patronaging this hairdresser for my hairstyling service,” and “The next time I desire a hairstyling service, I intend to re-patronage this hair dresser.” Positive word-of-mouth intention, one’s willingness for voluntary oral recommendation for a product/service to potential customers, was measured with three items by adjusting the original scale of previous research (Maxham and Netemeyer, 2002a, 2002b, 2003). Items included: “I would recommend this beauty salon for hairstyling to my friends,” “I will spread positive words about this beauty salon,” and “If my friends were looking for a new beauty salon, I would tell them to try this beauty salon.” 4.2. Findings 4.2.1. Reality assessment and manipulation check This study conducted reality assessment to find out whether or not respondents accepted the given scenarios as realistic. The findings revealed that t (185) = 59.71 and P = .000, indicating that respondents perceived our scenarios as realistic. The study also conducted manipulation check on the magnitude of two different service failure types: outcome failure and process failure to determine whether or not respondents perceived them as similar in respect of their severity. The results showed that t (114) = .27 and P = .785 with no particular mean difference between outcome failure and process failure, meaning that the impact of two different service failures were virtually same from the consumer perspective. Another manipulation check was done regarding the level of intimacy (high vs. low) to confirm that respondents recognized two different levels of intimacy as clearly different as to their extent. The findings revealed that t (184) = 5.64 and P = .000 with significant mean difference between high and low level of intimacy, indicating that the impact of two types of intimacy were clearly different in eyes of consumers. 4.2.2. Validity and reliability assessment The current study applied an exploratory factor analysis to ensure data validity. When it comes to intimacy, it had only one component explaining 62.45% of the total variance with an Eigenvalue = 1.873. Cronbach’s alpha for intimacy was .674. Another factor analysis was done on service evaluations including satisfaction, re-patronage intention, and positive word-of-mouth intention. As expected, the analysis extracted three factors accounting for 90.07% of the total variance. Cronbach’s alphas were .953, .957, and .910 for satisfaction, re-patronage intention, and positive word-ofmouth intention, respectively. Thus, we concluded that the entire constructs had strong internal consistency suitable for further analyses (see Table 3). 4.2.3. Hypothesis testing We predicted that service evaluations would vary depending on the type of service failure. To be specific, hypothesis 1 proposed that in case involving outcome failure, a high intimacy group would have

Please cite this article in press as: Seongun Jeon, Jonathan S. Kim, Effects of service failure on consumer responses across failure types: A moderating role of intimacy, Australasian Marketing Journal (2015), doi: 10.1016/j.ausmj.2015.12.004

ARTICLE IN PRESS S. Jeon, J.S. Kim/Australasian Marketing Journal ■■ (2015) ■■–■■

5

Table 3 Reliability and validity. Construct

Variables

Satisfaction

I feel pleased with what has been done in the beauty salon I am delighted with the service provided The hairdresser provided a satisfactory hairstyling service I am planning to visit this beauty salon again I will continue patronaging this hairdresser for my hairstyling service The next time I desire a hairstyling service, I intend to re-patronage this hair dresser I would recommend this beauty salon for hairstyling to my friends I will spread positive words about this beauty salon If my friends were looking for a new beauty salon, I would tell them to try this beauty salon

Re-patrongae intention

Positive word-of-mouth

Factor1

Eigenvalue % of variance explained

Table 4 ANOVA of satisfaction.

Failure type Intimacy level Failure type*intimacy level Error Total

df

MS

F

Sig.

2 1 2 180 185

97.135 13.233 5.206 1.647

58.976 8.034 3.161

.000 .005 .045

R 2 = .427 (Adjusted R 2 = .411 ).

6 5.5

Factor 3

.895 .890 .890

Cronbach’s Alphas .953

.824 .867 .878

2.844 31.601

higher service evaluations: a. satisfaction, b. re-patronage intention, and c. positive word-of-mouth intention than a low intimacy group. However, hypothesis 2 anticipated that in case involving process failure, a high intimacy group would have lower service evaluations: a. satisfaction, b. re-patronage intention, and c. positive word-of-mouth intention than a low intimacy group. To examine whether the satisfaction level was different depending on the service failure type and the intimacy level, 3 (no failure vs. outcome failure vs. process failure) × 2 (high intimacy vs. low intimacy) ANOVA was performed. The result showed that the interaction effect existed between service failure type and intimacy level with F = 3.161 and P = .045 (see Table 4). As shown in Fig. 1, a high intimacy group was more satisfied than a low intimacy group in times of outcome failure, providing support for hypothesis 1 as well as substantiating the existence of buffering effect. A t-test on satisfaction level between a high intimacy group and a low intimacy group supported such a result with t(62) = 2.34 and P = .023 for outcome failure. In times of process failure, however, there was

Factor 2

2.808 31.201

.957

.800 .844 .763 2.454 27.271

.910

no significant difference in the scores for high intimacy (M = 3.06, SD = 1.15) and low intimacy (M = 3.19, SD = 1.25) conditions; t(58) = .43, P = .668, contradicting hypothesis 2 as well as invalidating the existence of betrayal effect. Inconsistent with hypothesis 2, it was in accordance with several studies that refuted the relationship between initial negative emotion and consequent negative customer satisfaction (Andreassen, 2000; Menon and Dubé, 2004; Van Dolen et al., 2004; Varela-Neira et al., 2008; Westbrook and Oliver, 1991). Nevertheless, the presence of the directional trend of betrayal effect called for an additional study to clarify the relationship between process failure and service evaluations. In the same manner, ANOVA of re-patronage intention and positive word-of-mouth intention were conducted, and their findings were very similar to those of satisfaction. As shown in Tables 5 and 6, the interaction effect existed for both repatronage intention and positive word-of-mouth intention. In times of outcome failure, a significant mean difference was found between high and low intimacy condition in both dimensions: re-patronage intention with t(51) = 4.82, P = .000 and positive word-of-mouth intention with t(57) = 4.65, P = .000, supporting hypothesis 1b and hypothesis 1c. In times of process failure, however, there was no significant mean difference between high and low intimacy condition in both dimensions: re-patronage intention with t(58) = .53, P = .600 and positive word-of-mouth intention with t(58) = .17, P = .863, which countered hypothesis 2b and hypothesis 2c. To sum up, overall findings supported hypothesis 1, but rejected hypothesis 2. In summary, the study 1 showed that the level of intimacy moderated the impact of service failure on service evaluations. However, it was still questionable why the betrayal effect, despite the

5.828

5.043

5

(1) Satisfaction Level

4.5 High Initmacy

3.911 4

Low Intimacy

3.5

3.189 2.961

3 3.056 2.5 2 No Failure

H1. Outcome Failure

H2. Process Failure

Fig. 1. Satisfaction level depending on service failure type and intimacy level.

Please cite this article in press as: Seongun Jeon, Jonathan S. Kim, Effects of service failure on consumer responses across failure types: A moderating role of intimacy, Australasian Marketing Journal (2015), doi: 10.1016/j.ausmj.2015.12.004

ARTICLE IN PRESS S. Jeon, J.S. Kim/Australasian Marketing Journal ■■ (2015) ■■–■■

6

4

Table 5 ANOVA of re-patronage intention.

Failure type Intimacy Failure type*intimacy level Error Total

df

MS

F

Sig.

2 1 2 180 185

52.836 35.007 7.790 1.561

33.857 22.423 4.990

.000 .000 .008

3.5 3 Satisfaction Level

2.5 2

R 2 = .359 (Adjusted R 2 = .341 ).

1.5 Table 6 ANOVA of positive word-of-mouth intention.

Failure type Intimacy level Failure type*intimacy level Error Total

1

df

MS

F

Sig.

2 1 2 180 185

32.278 18.866 6.997 1.162

27.779 16.236 6.022

.000 .000 .003

R 2 = .320 (Adjusted R 2 = .301).

presence of directional trend, was empirically absent in times of process failure. In order to clearly understand under what settings intimacy interact with process failure, the study 2 was conducted. 5. Study 2 The purpose of study 2 was to identify what led to betrayal effect after all. A few research claimed that customers were more likely to seek revenge after a firm had failed to redress initial service failure (Bechwati and Morrin, 2003) or when double deviation, service failure and failed recovery, was present (Bitner et al., 1990; Grégoire et al., 2010). Adopting this logic, we postulated a situation when process failure occurred and following recovery was a failure, and attempted to confirm whether service evaluations varied depending on the level of intimacy (high vs. low). 5.1. Methods As measures and intimacy scenarios were consistent with those of the study 1, we would focus on the difference peculiar to study 2. We originally obtained data from a sample of 69 respondents but later eliminated six questionnaires which had incomplete or missing responses, or which failed to pass reality assessment (ratings less than four). Accordingly, a total of 63 (high intimacy: 30 vs. low intimacy: 33) responses were used. Participants for this study were undergraduate and graduate students taking a marketing course during a winter semester at a major university in Seoul, South Korea. Most of the respondents were in their 20s (81%) followed by 30s (11.1%), and 47.6% of them were male. The subjects were asked to answer the questions after reading a scenario describing one of the intimacy levels (high vs. low) as well as process failure and following failed recovery in the setting of hair industry. The scenario of process failure and following failed recovery was depicted as a customer experiencing process failure as in the study 1 and then receiving nothing but a token apology although he/she expected a discount coupon or at least authentic apology (see Appendix B). 5.2. Findings We conducted reality assessment and manipulation check on the level of intimacy as in study 1 and the findings were very similar. As depicted in Fig. 2, the respondents differed in their mean level of satisfaction ( Mhigh intimacy = 2.02, SDhigh intimacy = .83, Mlow intimacy = 3.67 ,

High Intimacy

Low Intimacy

Fig. 2. Satisfaction level depending on the level of intimacy in case involving process failure and following failed recovery.

SDlow intimacy = .79 ). The findings of the independent t-test revealed that there was statistically significant mean difference in satisfaction level depending on the level of intimacy; t(61) = 8.07, P = .000, indicating that customers maintaining a high level of intimacy with service providers were unsparing to failed recovery after process failure. Similar results were found for re-patronage intention with t(61) = 6.24, P = .000 and positive word-of-mouth intention with t(61) = 4.76, P = .000. All in all, it was the failed recovery after process failure that drove customers to have significantly low service evaluations. That is, the absence of failed recovery was the very reason why betrayal effect did not exist in study 1. 6. Conclusion 6.1. Summary This study extended the previous research about the impact of intimacy on service evaluations by specifying service failure types into outcome and process failure. We classified service failure in this way to empirically analyze in which circumstances two different consumer responses, buffering and betrayal effect, occur. This paper initially adopted 3 (failure type: no vs. outcome vs. process) × 2 (intimacy level: high vs. low) factorial designs and later compared the effects of intimacy (high vs. low) on service evaluations in the setting of process failure and following failed recovery. Findings of this paper can be summarized as follows. In case involving outcome failure, a high intimacy group had higher service evaluations (satisfaction, re-patronage intention, and positive wordof-mouth intention) than a low intimacy group, demonstrating the existence of buffering effect. On the other hand, in case involving process failure, service evaluations did not differ depending on the level of intimacy, nullifying the existence of betrayal effect. Study 2 revealed that betrayal effect was present when failed recovery had been considered after process failure. 6.2. Implications The findings reported here raise questions of interest to consumer researchers and marketing practitioners alike. On the research front, this research, to our knowledge, is the first to suggest the consolidated classification scheme of service failure types, which promotes future research concerning service failure and recovery. It also addresses opposing consumer responses (buffering and betrayal effect) in a single research, thereby providing a refined understanding of how service failures are perceived by consumers who are more (vs. less) intimate. From a practitioner perspective, service providers are recommended to keep in mind that not all the service failure types are same. On the premise that service provid-

Please cite this article in press as: Seongun Jeon, Jonathan S. Kim, Effects of service failure on consumer responses across failure types: A moderating role of intimacy, Australasian Marketing Journal (2015), doi: 10.1016/j.ausmj.2015.12.004

ARTICLE IN PRESS S. Jeon, J.S. Kim/Australasian Marketing Journal ■■ (2015) ■■–■■

ers have already formed a high level of intimacy with their customers, the customers are more tolerant of outcome failure. It is thus important for service providers to form and maintain a high level of intimacy with their customers in preparation for outcome failure. However, dealing with process failure is a completely different story. A high level of intimacy between service providers and their customers is not at all helpful in mitigating process failure. Such situation would become even worse in the presence of following failed recovery that might bring about anger. In order to react immediately to process failure, service providers should train themselves to recognize cues for process failure, noting that customers express negative emotions through facial, postural, vocal, and verbal exchanges, each of which corresponds to discrete negative emotions Menon and Dubé (2000). All in all, although service providers are inextricably bound up with service failure, mistakes can always occur as those who provide service are human beings. This is why service providers should remember to differentiate service failure types if the failure were to occur. Sometimes, they can benefit from a high level of intimacy with their customers, but in other times, they should come up with effective recovery to prevent betrayal effect. We hope this research can provide service providers with helpful insight on how to utilize intimacy in different service failure situations.

6.3. Limitations and directions for future research Since no empirical research is free of limitations, this paper is no exception. Several limitations exist on a general level. First, this research employed a convenient sample, mostly students who were in their 20s, as the sample was collected at a university in South Korea. Although gender difference would be minimal in that Korean male college students tend to have a lot of interest in beauty care possibly because physical appearance is another qualification in a current job market, age difference is still highly likely. Not many middle-aged men get permed in beauty salons but prefer to get a haircut in barber shops. Since this issue comes under an arena of cultural or age difference, future research should seek a sample of more diversity. Second, a scenario method adopted for this research may have lacked a vivid sense of realism or actuality as it was conveyed through script. Therefore, we could utilize Critical Incident Technique (CIT) or ethnographic research method for further study. Third, we have yet to discover variables that mediate the relationship between the interaction of service failure (outcome vs. process) and intimacy (high vs. low) and the overall service evaluations. Specifically, in high intimacy situation, a sense of stability may serve as a mediator between outcome failure and modest service evaluations; whereas anger may be a mediator between failed recovery after process failure and very poor service evaluations given that customer anger is provoked by severe service failure (Nguyen and McColl-Kennedy, 2003). On a more specific level regarding the failure to uncover betrayal effect in times of process failure, two factors should be further considered in future research. The future research could clearly mention the level of service quality in advance. If we had manipulated service quality as considerably high, consumer expectations might have soared, thereby eliciting greater anger in response to process failure. Next, the further research should specify the identity of a service provider. Considering that many beauty salons are currently branded, consumers would react differently depending on whether the service failure is due to an individual hair dresser or a service brand. Service evaluations might deteriorate far more rapidly on an organization (or brand) than on an individual given that the service relationship often exists with the individual rather than the organization (Sweeney and Chew, 2002). It is likely that

7

consumers might turn away more easily from the organization, nonhuman being void of intimacy.

Appendix A No failure ** I visited H beauty salon to have my hair permed. On arrival, I waited for about 10 minutes as usual and was seated by a hairdresser. The hairdresser gave me a perm service skillfully. Once done with hairstyling, I paid a bill and headed home. Outcome failure ** I visited H beauty salon to have my hair permed. On arrival, I waited for about 10 minutes as usual and was seated by a hairdresser. The hairdresser gave me a perm service skillfully. Once done with hairstyling, I realized that the coating service that used to be provided during a perm was not fulfilled due to the hairdresser’s mistake. Since my hair was already done, I had no other options but to head home as soon as I paid a bill. I arrived home and later found out that the extra cost of $10 had been incurred. It was the hairdresser who mistakenly charged $10 more without considering that my hair was short enough to have $10 discount. Process failure ** I visited H beauty salon to have my hair permed. As the salon was full of customers, I waited for my turn for 40 minutes. In the middle of a long wait reading a magazine, I looked up and found out that another customer who arrived later was already on the seat. Only then did the hairdresser acknowledge my existence and showed me to the seat. While getting a perm, I was so thirsty and asked the hairdresser for a cup of tea. The hairdresser answered but did not bring it caring for other customers. I asked again but the hairdresser did not bring it after all. High intimacy ** I have been regularly visiting H beauty salon for 5 years. I get a hairstyling service from Ms. Kim every time I go there. Ms. Kim is humorous and friendly enough to ask after my hobby and work. I really enjoy having conversation with Ms. Kim. Nowadays, we are even more familiar to each other sharing private affairs. Low intimacy ** I have been recently visiting H beauty salon for 3 months. I think the price there is reasonable and am generally satisfied with its hairstyling service. I do not get a hairstyling service from a certain hair dresser. Hair dressers there are mostly kind and skilled so that I rarely care from whom I receive hairstyling service. This may be why I am not familiar with any individual hairdresser and rarely have a chance to share private affairs.

Appendix B Failed recovery ** Once done with hairstyling, I stood face to face with the hairdresser at the counter. I expected a sort of discount coupon that I could use for my next visit or at least a sincere apology. However, the hairdresser apologized as a matter of form offering nothing to make up for mistakes. I headed home this way.

Please cite this article in press as: Seongun Jeon, Jonathan S. Kim, Effects of service failure on consumer responses across failure types: A moderating role of intimacy, Australasian Marketing Journal (2015), doi: 10.1016/j.ausmj.2015.12.004

ARTICLE IN PRESS S. Jeon, J.S. Kim/Australasian Marketing Journal ■■ (2015) ■■–■■

8

References Aaker, J., Fournier, S., Brasel, S.A., 2004. When good brands do bad. J. Consum. Res. 31 (1), 1–16. Andreassen, T.W., 2000. Antecedents to satisfaction with service recovery. Eur. J. Mark. 34 (1/2), 156–175. Bechwati, N.N., Morrin, M., 2003. Outraged consumers: getting even at the expense of getting a good deal. J. Consum. Psychol. 13, 440–453. Beetles, A.C., Harris, L.C., 2009. The role of intimacy in service relationships: an exploration. J. Serv. Mark. 24 (5), 347–358. Berens, G., Van Riel, C.B.M., Van Bruggen, G.H., 2005. Corporate associations and consumer product responses: the moderating role of corporate brand dominance. J. Mark. 69 (3), 35–48. Bitner, M.J., Booms, B.H., Tetreault, M.S., 1990. The service encounter: diagnosing favorable and unfavorable incidents. J. Mark. 54 (1), 71–84. Blount, J., 2010. You don’t send me flowers: customers are lost most often due to neglect. (accessed 06.26.15). Brady, M.K., Knight, G.A., Cronin, J.J., Tomas, G., Hult, M., Keillor, B.D., 2005. Removing the contextual lens: a multinational, multi-setting comparison of service evaluations models. J. Retail. 81 (3), 215–230. Brock, J.K.U., Zhou, J.Y., 2012. Customer intimacy. J. Bus. Ind. Mark. 27 (5), 370– 383. Brown, T.J., Barry, T.E., Dacin, P.A., Gunst, R.F., 2005. Spreading the word: investigating antecedents of consumers’ positive word-of-mouth intentions and behaviors in a retailing context. J. Acad. Mark. Sci. 33 (2), 123–138. Burns, D.J., Neisner, L., 2006. Customer satisfaction in a retail setting: the contribution of emotion. Int. J. Retail. Distrib. Manage. 34 (1), 49–66. Collier, D.A., 1995. Modeling the relationships between process quality errors and overall service process performance. Int. J. Serv. Ind. Manage. 6 (4), 4–19. Cronin, J.J., Taylor, S.A., 1994. SERVPERF versus SERQUAL: reconciling performancebased and perceptions-minus-expectations measurement of service quality. J. Mark. 58 (1), 125–131. Crosby, L.A., Stephens, N., 1987. Effects of relationship marketing on satisfaction, retention, and prices in the life insurance industry. J. Mark. Res. 24 (4), 404– 411. Forbes, L.P., Kelley, S.W., Hoffman, K.D., 2005. Typologies of e-commerce retail failures and recovery strategies. J. Serv. Mark. 19 (5), 280–292. Fornell, C., Johnson, M.D., Anderson, E.W., Cha, J., Bryant, B.E., 1996. The American customer satisfaction index: nature, purpose, and findings. J. Mark. 60 (4), 7–18. Frederick, F.R., Sasser, W.E., 1990. Zero defections: quality comes to services. Harv. Bus. Rev. 68 (5), 105. Grégoire, Y., Laufer, D., Tripp, T.M., 2010. A comprehensive model of customer direct and indirect revenge: understanding the effects of perceived greed and customer power. J. Acad. Mark. Sci. 38, 738–758. Gremler, D.D., Brown, S.W., 1998. Service loyalty: antecedents, components, and outcomes. Paper presented at 1998 AMA Winter Educators’ Conference: Marketing Theory and Applications, Chicago IL, 165–166. Gustafsson, A., Johnson, M.D., Roos, I., 2005. The effects of customer satisfaction, relationship commitment dimensions, and triggers on customer retention. J. Mark. 69 (4), 210–218. Hansen, H., 2003. Antecedents to consumers’ disclosing intimacy with service employees. J. Serv. Mark. 17 (6), 573–588. Hellier, P., Geursen, G., Carr, R., Rickard, J., 2003. Customer repurchase intention: a general structure equation model. Eur. J. Mark. 37 (11/12), 1762–1800. Hoffman, K.D., Kelley, S.W., Rotalsky, H.M., 1995. Tracking service failures and employee recovery efforts. J. Serv. Mark. 9 (2), 49–61. Hoffman, K.D., Kelley, S.W., Chung, B.C., 2003. A CIT investigation of servicescape failures and associated recovery strategies. J. Serv. Mark. 17 (4), 322–340. Holloway, B.B., Wang, S., Beatty, S.E., 2009. Betrayal? Relationship quality implications in service recovery. J. Serv. Mark. 23 (6), 385–396. Hui, M.K., Zhao, X., Fan, X., Au, K., 2004. When does the service process matter? A test of two competing theories. J. Consum. Res. 31 (2), 465–475. Jeon, S., Kim, G., Lee, S., Park, J., Jeong, M., Kim, S., 2013. A study on influence of intimacy on service evaluations. Serv. Mark J. 6 (2), 5–16. Kelley, S.W., Hoffman, K.D., Davis, M.A., 1993. A typology of retail failures and recoveries. J. Retail. 69 (4), 429–452.

Lopez-Lopez, I., Ruiz-de-Maya, S., Warlop, L., 2014. When sharing consumption emotions with strangers is more satisfying than sharing them with friends. J. Serv. Res. 17 (4), 475–488. Maxham, J.G., III, 2001. Service recovery’s influence on consumer satisfaction, positive word-of-mouth intention, and purchase intentions. J. Bus. Res. 54 (1), 11–24. Maxham, J.G., III, Netemeyer, R.G., 2002a. A longitudinal study of complaining customers’ evaluations of multiple service failures and recovery efforts. J. Mark. 66 (4), 57–71. Maxham, J.G., III, Netemeyer, R.G., 2002b. Modeling customer perceptions of complaint handling over time: the effect of perceived justice on satisfaction and intent. J. Retail. 78 (4), 239–252. Maxham, J.G., III, Netemeyer, R.G., 2003. Firms reap what they sow: the effects of shared values and perceived organizational justice on customers’ evaluations of complaint handling. J. Mark. 67 (1), 46–62. Menon, K., Dubé, L., 2000. Ensuring greater satisfaction by engineering salesperson response to customer emotions. J. Retail. 76 (3), 285–307. Menon, K., Dubé, L., 2004. Service provider responses to anxious and angry customers: different challenges, different payoffs. J. Retail. 80 (3), 229–237. Nguyen, D.T., McColl-Kennedy, J.R., 2003. Diffusing customer anger in service recovery: a conceptual framework. Australas. Mark. J. 11 (2), 46–55. Oliver, R.L., 1997. Customer Satisfaction. A Behavioral Perspective on the Consumer. McGraw-Hill, New York. Parasuraman, A., Zeithaml, V.A., Berry, L.L., 1985. A conceptual model of service quality and its implications for future research. J. Mark. 49 (4), 41–50. Park, S.J., 2007. A study of customer responses to service failure and recovery: the role of service provider’s recovery effort and customer-employee rapport. Asia Mark. J. 9 (3), 75–115. Smith, A.K., Bolton, R.N., 2002. The effect of customers’ process responses to service failures on their recovery effort evaluations and satisfaction judgments. J. Acad. Mark. Sci. 30 (1), 5–23. Smith, A.K., Bolton, R.N., Wagner, J., 1999. A model of customer satisfaction with service encounters involving failure and recovery. J. Mark. Res. 36 (3), 356–372. Spreng, R.A., Harrell, G.D., Mackoy, R.D., 1995. Service recovery: impact on satisfaction and intentions. J. Serv. Mark. 9 (1), 15–23. Sweeney, J.C., Chew, M., 2002. Understanding consumer-service brand relationships: a case study approach. Australas. Mark. J. 10 (2), 26–43. Tax, S.S., Brown, S.W., Chandrashekaran, M., 1998. Customer evaluations of service complaint experiences: implications for relationship marketing. J. Mark. 62 (2), 60–76. Taylor, S., 1994. Waiting for service: the relationship between delays and evaluations of service. J. Mark. 58 (2), 56–69. Van Dolen, W., Lemmink, J., Mattsson, J., Rhoen, I., 2001. Affective consumer responses in service encounters: the process content in narratives of critical incidents. J. Econ. Psychol. 22 (3), 359–376. Van Dolen, W., Ruyter, K.D., Lemmink, J., 2004. An empirical assessment of the influence of customer emotions and contact employee performance on encounter and relationship satisfaction. J. Bus. Res. 57 (4), 437–444. Varela-Neira, C., Vazquez-Casielles, R., Iglesias-Arguelles, V., 2008. The influence of emotions on customer’s cognitive evaluations and satisfaction in a service failure and recovery context. Serv. Ind. J. 28 (4), 497–512. Voss, G.B., Parasuraman, A., Grewal, D., 1998. The roles of price, performance, and expectations in determining satisfaction in service exchanges. J. Mark. 62 (4), 46–61. Westbrook, R.A., 1980a. Intrapersonal affective influences on consumer satisfaction with products. J. Consum. Res. 7 (1), 49–54. Westbrook, R.A., 1980b. A rating scale for measuring product/service satisfaction. J. Mark. 44 (4), 68–72. Westbrook, R.A., Oliver, R.L., 1991. The dimensionality of consumption emotion patterns and consumer satisfaction. J. Consum. Res. 18 (1), 84–91. Wolak, R., Kalafatis, S., Harris, P., 1998. An investigation into four characteristics of services. J. Empir. Gen. Mark. Sci. 3 (2), 22–43. Yi, Y., 1990. A critical review of consumer satisfaction. Rev. Mark. 4 (1), 68–123. Yi, Y., 2013. Service Marketing. Hakhyunsa; Seoul. Zeelenberg, M., Pieters, R., 2004. Beyond valence in customer dissatisfaction: a review and new findings on behavioral responses to regret and disappointment in failed services. J. Bus. Res. 57 (4), 445–455. Zeithaml, V.A., Berry, L.L., Parasuraman, A., 1993. The nature and determinants of customer expectations of service. J. Acad. Mark. Sci. 21 (1), 1–12.

Please cite this article in press as: Seongun Jeon, Jonathan S. Kim, Effects of service failure on consumer responses across failure types: A moderating role of intimacy, Australasian Marketing Journal (2015), doi: 10.1016/j.ausmj.2015.12.004