This action might not be possible to undo. Are you sure you want to continue?
Mohr Source: The Journal of Marketing, Vol. 58, No. 4 (Oct., 1994), pp. 95-106 Published by: American Marketing Association Stable URL: http://www.jstor.org/stable/1251919 Accessed: 23/10/2008 02:36
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=ama. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact firstname.lastname@example.org.
American Marketing Association is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Marketing.
Service quality researchers have suggested that "the proof of service [quality] is in its flawless performance" 1991. just as it has become essential in manufacturing (Heskettet al. Drawingon insights from role. the most immediate evidence of service occurs in the service encounteror the "moment of truth"when the customer interacts with the firm. authors gratefully acknowledge support of the First Interstate Center Services for and of Marketing theCollege Arizona University. this The fulcomments three of JM are anonymous reviewers alsoappreciated. 1994. script. and Hart 1990). Program. is Professor Marketing.in the context of the same threeindustries.e. Because the service encounterinvolves at least two people. firms are betterable to design processes and educate both employees and customers to achieve quality in service encounters. Although more firms are realizing the importanceof service quality and customer satisfaction. University Lois A. In service encounterssuch disagreements. he worldwidequality movementthathas swept the manufacturing sector over the last decade is beginning to take shape in the service sector (Business Week1991. the shift to a quality focus is essential to the competitive survival of service businesses. and Tetreault 1990. hotel. Previous research has identifiedthe sources of satisfactionand dissatisfaction in service encounters fromthe customer's point of view. the Armed with such understanding. Sasser. of Washington-Tacoma.it should hold for differentrespondentgroups. it is important to understand encounterfrom multipleperspectives. rather 100% satisfying but performancefrom the customer'spoint of view. underscorethe importance of understandingthe types of events and behaviors that cause customers to be satisfied or dissatisfied. is Business of versity. negative word of mouth. what kinds of events lead to dissatisfyingservice encountersfor the customer?What causes these events to be rememberedwith distaste? *Do customers and employees report the same kinds of events and behaviorsleading to satisfactionand dissatisfaction in service encounters? Critical Service Encounters 95 / . employees) and the customer. Journal of Marketing Vol. From the customer'spoint of view.the kinds of events and behaviorsthat employees believe underlie customer satisfaction. Mohr anAssistant State The the Georgia University. restaurant. Situations arise in which quality is low and the problem is recognized by both the firm Jo is Professor Marketing. Crosby 1991). and decreasedemployee morale. Booms. 15).Here. flawless performanceis not meant to imply rigid standardization.A second purpose of the study is to evaluatethe usefulness of the classification scheme developed by BBT (1990). Schlesinger and Heskett 1991). The cost of not achieving flawless performanceis the "cost of quality. employee empowerment and training. The findings have implicationsfor business practice in managing service encounters. 95-106 (i.. but there may be disagreementon the causes of the problemand the appropriate solutions. The primarypurposeof this study is to examine the contact employee's perspective of critical service encountersand to understand. & LoisA. Mohr MaryJo Bitner. customer satisfactionis often influencedby the qualityof the interpersonalinteraction between the customer and the contact employee.it is not always clear how to achieve these goals. and managing customers. Results generallysupportthe theoreticalpredictionsand also identifyan additionalsource of customer dissatisfaction-the customer's own misbehavior.BernardH. State inconducting research. Critical Service Encounters: The Employee's Viewpoint In service settings. H.sure to diminish customer satisfaction.and airlineindustriesare analyzed and compared withpreviousresearch.whatkindsof eventslead to satisfying serviceencounters the cusfor tomer? Whatcausestheseeventsto be remembered favorably? *Fromthe contactemployee's point of view. 774 criticalservice encounters reportedby employees of the hotel. p." which includes the costs associated with redoing the service or compensating for poor service. this study explores these sources in service encounters fromthe contact employee's pointof view. Previous researchin the context of the restaurant. Thus. If the scheme is conceptually robust.The employee perspective is then compared with BBT to gain insight into any disparitiesin perspectives. Booms. 58 (October 1994). of Arizona UniState Mary Bitner anAssociate Bernard Booms a Professor. a concept akin to the (Berry and Parasuraman notion of "zero defects" in manufacturing. one central goal in the pursuitof "zero defects" in service is to work toward 100% flawless performancein service encounters. The researchis guided by the following questions: *From contact the employee's pointof view. helpBusiness. hereafterBBT). According to some. lost customers. and attribution theories.Others have noted that "breakthrough" service managerspursuethe goal of 100% defect-free service (Heskett. and airline industriesidentifiedcategories of events and behaviors that underlie critical service encounters from the customer's point of view (Bitner.
Berry. to blame failure on external causes. We would thereforeexpect.63 and r = .67. contact personnel can be expected to look frequently for cues that tell them how their service is received by customers. Given these biases we would expect employees to blame the system or the customer for service failures. Researchershave theorizedand found some evidence thatopen communicationbetween frontline personnel and managers is important for achieving service quality (Parasuraman. 1985). There are two basic ways that customer knowledge obtained by contact employees is used to improve service: (1) Such knowledge is used by the contact employees themselves to facilitate their interactions with customers and (2) It is used by the firm for making decisions.or interferencesrequire the enactmentof complex or less routine subscripts.the roles are well defined and both the customer and employee know what to expect from each other. Similarly.. and role expectations are the standardsfor role behavior(Biddle 1986)."two types of interferences that may occur in otherwisepredictablescripts. Role. to find similarities in employee and customer views of the service encounter. In many routine service encounters. a complaintcontext (Resnik and Harmon 1983). Research shows that there are many biases in the attribution process (Fiske and Taylor 1984). and attribution theories provide conceptualbases for these expectations.but we would expect significant differences as well. and Buxton (1980) found high correlations(r = .however. October1994 studies have found differences when comparing customer and employee evaluationsof business situations using scenarios and role playing in product failure contexts (Folkes and Kotsos 1986). Anotherstudy of 1300 customersand 900 customerservice professionals conducted by Development Dimensions Internationalfound differences in perceptions between the two groups (Services Marketing Newsletter 1989). a self-protectingbias).are repeated frequently throughouta person's life. It seems reasonableto conclude that accurateemployee of understanding customersenables both the employee and the firm to adjust appropriately customer needs. 1985). The differences they found were ratherlarge and inversely related to overall patient satisfaction. many types of service encounters. respectively) between employee and customerattitudes about overall service quality in a bank setting. structuresthat describe appropriatesequences of role behaviors)(SchankandAbelson 1977).particularly experiencedemployees and for customers. In addition. especially decisions regarding new service developmentand service modifications. To the extent that this is true. Other 96 / Journalof Marketing. Howevto er. when their attributions differ. Zeithaml. Theoretical Explanations Role and Script Theories Similarities in how customers and employees view service encountersare most likely when the two partiesshare common role expectationsand the service script is well defined (Mohr and Bitner 1991. As a result. resultingin strong. previous research correlating customer and employee views of service is sparse and offers mixed conclusions. When service encountershave strongscripts. First. Schneider(1980) argues that people who choose to work in service occupations generally have a strong desire to give good service.the employee and customer are likely to share expectations about the events that will occur and the orderof occurrence. Second. such as seating customers in a restaurant. a self-enhancingbias) and deny responsibility for failure (i. a participantis unfamiliarwith expected behaviors. the more likely theirbehavioraladjustmentsare to improvecustomer satisfaction.Langeardand colleagues (1981) found thatfield managersat two banks tended to overestimate(comparedwith customerratings)the importance of six broad service delivery dimensions. and Parasuraman 1988).. employees often modify their behavior from moment to moment on the basis of feedback they receive while serving customers. Berry. suggests that customers and employees are likely to share a common perspectiveon service experiences. Attribution Theory Dissimilarities in viewpoint may arise when service encounter partnershave conflicting views of the underlying causes behind the events.e. script.They are less likely to shareideas aboutsubscripts. they serve a boundary-spanning role in the firm. The more accuratetheir perceptionsare. Customer and Contact Employee Viewpoints Frontline personnel are a critical source of information about customers. we discuss relevant researchand theory.e. Their results are contradicted. and Zeithaml 1990. because contactpersonnelhave frequentcontact with customers. Most clearly relevant for the perceptionsof service providersand customers is the self-serving attribution bias. Solomon et al. Schneider and Bowen (1985) and Schneider. in a study by Brown and Swartz (1989).whereasthe customerwould be more likely . Role and scripttheory..e. A role is the behavior associated with a socially defined position (Solomon et al.which are prescriptions for handling what Schank and Abelson describe as "obstacles and errors. to give internalattripeople butions for their successes. It is also clear that differencesin perspective may arise when roles are less defined. Parkington. they often have better understandingof customer needs and problems than others in the firm. standardized.combined with the routinenature of many service encounters. Schneider and Bowen (1984) argue that firms should use informationgatheredfrom contact personnel in making strategic decisions.Before presenting the empirical study. and well-rehearsedscripts (i. thatis. Customer service professionals in that study consistently rated the importanceof particular service skills and competencies and theiractualperformancehigherthancustomersratedthe same skills and competencies. on the basis of these studies. and the context of retailer responses to customer problems(Domoff and Dwyer 1981). This is the tendency for to take credit for success (i. These researchersgathered data on patient experiences with their physicians and compared them with the physicians' perceptions of their patients' experiences.
a systematic procedurefor recordingevents and behaviors that are observed to lead to success or failure on a specific task (Ronan and Latham 1974). the self-serving attribution bias leads to the expectationthatthe perspectives of the employee and customer will differ more in service failure than in service success situations. the same classification system should be appropriate.It is less clear thatthis bias would operatein the case of a service encounter success. Because respondentsare asked about specific events ratherthan generalities. Whatresulted thatmadeyou feel the interaction was fromthe customer's satisfying (dissatisfying) pointof view? 5. This researcher a second then workedtoand getheron categorizingthis groupof 86 incidents(1 1%of the total). leaving 774 incidents (397 satisfactory and 377 dissatisfactory). They were then asked the following questions: 1. the data provide insight into the nature of these disparities. then. Exactlywhatdid you or yourfellow employeesay or do? 4. valid with respect to the content identified. White and Locke 1981). meaningful categories. satisfying the customer. the critical incident method is reliable in terms of stability of the categories identified across judges.and 4 airlines. the fact that the customer is paying the firm for a service would probablyprecludethe bias on the customer's side. Whendidtheincident happen? 2. and (4) have sufficientdetail to be visualized by the interviewer.(2) be very satisfying or dissatisfying from the customer's point of view.tryto see yourfirmthrough yourcustomers' eyes. Both empiricalresearchand theory suggest that similarities as well as differences in perspectiveare likely to occur between service encounterparticipants.Role and scripttheories suggest that in relatively routine situationssuch as the ones studied. Thinkof a recenttimewhena customer yourfirmhad of a particularly interaction with satisfying(dissatisfying) or Describe situation the and yourself a fellowemployee. Whatspecificcircumstances up to thissituation? led 3.5 years of workingexperience in their respective industries. restaurant.interpretation. (For more detailed discussions of the method. and the results are content analyzed. Because all the interviewerswere employed in the hospitality sector. asking each employee to describe one incident that was satisfactoryand one thatwas dissatisfactory from the customer'spoint of view. The instructions to the employees being interviewedwere as follows: Put yourselfin the shoes of customers your firm. Whatshouldyou or yourfellowemployee havesaidor done?(fordissatisfying incident only) To be used in the analysis. However. in this case. and resorted until a consistent coding scheme was developed that combined similar incidents into distinct.And. Researchers have concluded that when used appropriately(Flanagan 1954. When the new categories were labeled and the two researchersachievedconsensus on assignmentof the incidents. Overall. combined. . Hotel. and airline employees were interviewed and asked to recall critical service encounters that caused satisfaction or dissatisfactionfor customers of their firms. The result would be different views of the causes of service dissatisfaction. the employees providingthe incidentshad 5. Classification of Incidents The incident classification system developed by BBT was used as a starting point for sorting the data with the assumptionthat. We explore to what extent the perspectivesof contact personnel and those of customers are different. These incidents were read and sorted. Wilson-Pessano 1988). to the degree that customers and employees remember satisfying and dissatisfying encounters in the same way. 152 restaurants.Seven incidentsfailed to meet these criteria. Although the desire for self-enhancement might lead both the employee and customer to give themselves credit for the success.They were instructednot to interviewfellow students. Thirty-seven trained student interviewers collected the data-781 total incidents. Respondentsare asked to reportspecific events from the recent past (within 6 to 12 months). an incident was requiredto (1) involve employee-customerinteraction. Any that did not fit into the scheme were put aside. see BBT. Wilson-Pessano 1988). reliable information about cognitive processes. These accounts provide rich details of firsthand experiences in which customers have been satisfied or dissatisfied.In of otherwords. They included general instructions for Critical Service Encounters 97 / Method and Analysis Data Collection Data were collected using the critical incident technique (CIT). Each one recruited a minimum of ten employees from among the same three industries studied in BBT. Flanagan 1954.Incidentsthat could not be classified within the original scheme would then provide evidence for differences in perspective. The incident sample represented58 hotels.to blame the system or the employee. exactlywhathappened. the new categories (one majorgroup with four subcategories) were added to the original classification system.they recruitedfellow employees and employees of establishmentswith which they were familiar. to the degree that they are different. or conclusions. (3) be a discreteepisode. A set of complete coding instructionswas then written (see Appendix A).The refusal rate was negligible. openended questions. Using the CIT. there will be strong similaritiesin perspective. attributionbiases suggest that there will also be significantdifferences in viewpoint.and relevant in that the behaviorsilluminatedhave provento be important to the success or failure of the task in question (Ronan and Latham 1974. One researcher trained in the classification scheme coded the incidents. The employees ranged in age from 16 to 65 (mean age 27) and were 55% female and 45% male. data are collected throughstructured. On average. this proceduremeets criteriaestablishedby Ericsson and Simon (1980) for providing valuable.
668 were classified into one of these threegroupsand the 12 categories within them.coders.(See BBT for detailed descriptions of the groups and categories and sample incidents. rized into one major group labeled "problemcustomer beand havior.911 and . our discussion focuses on the four major groups.e.816 for the satisfying and . 3. and rarely were they able to deal with them in such a way as to bring about customer satisfaction. Drunkenness-The the to perceives customer employee be clearly intoxicated creating and such problems as harassingothercustomers nearby. The coding instructions were used to train a third researcherwho in had not participated the categorizationdecisions. the lower the percentage agreement is likely to be) (Perreaultand Leigh 1989).5 Group 1-Response to failures 3 action 22. (2) employee response to customerneeds and requests. To facilitateunderstanding. Of the 774 employee incidents. notjust on the agreement between judges.4 Group2-Response to requests 2 27. This researcherthen coded the 774 employee incidents.however. or disruptingthe atmosphereof the establishment. Then the results are compared with customer perceptions using the BBT data. Results and Discussion The categories of events and behaviors that employees believe underlie their customers' satisfaction and dissatisfaction in service encountersare identified and discussed first. givingthe employeea hard time. The data are shown in full in Table 2.Nor can greaterimportancebe inferredby greaterfrequencies in a particularcategory (Wilson-Pessano 1988). For this reason.and (3) unprompted unsoand licited employee actions.4 Group3-Unprompted 4 . especially consideringthatthe classification system in this study contains 16 categories. the root cause was the customer. Fromthe employee's the is perspective. Verbaland physical abuse-The customerverbally and/or abuseseitherthe employeeor other physically customers. and decision rules for assigning incidents to categories.thatis. Such customerswere basically uncooperative. only 3 of these incidents were satisfactory.823 for the dissatisfyingincidents.The percentageagreement statistic probably underestimates interjudge reliability in this case because this statisticis influencedby the numberof coding categories (i. To correct for this they designed an alternativeindex of reliability. unwilling to cooperatewith the service provider. These situationscreated problems for the employees.0 Group4-Problem customers 3 16.7 Group 1-Response to failures 2 22. employee-reported the incidents are summarizedand rankedaccording to the percentageof incidents in the four majorincident groups: Distributionof DissatisfactoryIncidents Rank Order Group# Percentage 1 51. These incidents were catego98 / Journalof Marketing. Perreaultand Leigh (1989) argue.8 Group3-Unprompted action Distributionof SatisfactoryIncidents Rank Order Group# Percentage 1 49. the coders could not attribute satisfactionand dissatisfactionto an action or attitudeof the employee-instead. The incidents were very similar in detail to those providedby customers. appropriate marketfor ing data.) Eighty-six encounters(11% of the total) did not fit any of the predeterminedgroups.Ir is based on a model of the level of agreementthat might be expected given a true the (population)level of reliability. These figures are respectably high. was found to be .Ir. The actual frequency of occurrence of the type of event representedby a particulargroup or category cannot be inferredfrom the data. 2. It shouldbe noted thatthe frequenciesand proportionsshown in the table reflect numbersof reported events." they were addedto the categorizationscheme as the "Group4. which corrects for the likelihood of chance agreementbetween judges.that K is an overly conservativemeasure of reliability because it assumes an a prioriknowledge of the likely distributionof responses across categories. Uncooperative customers-The customer is generally rude and uncooperative unreasonably or demanding.914 for the satisfying and dissatisfying incidents.4 Group2-Response to requests 4 9. providing an interjudgereliability check on the classification system. October1994 Classification The Employee's View of Satisfactory Versus Dissatisfactory Encounters Here we examine the frequencies and proportionsof employee accounts in the four groups and 16 categories as shown in Table2.industry regulations. two other measuresof interjudgereliability were calculated. index focuses on the reliabilityof the whole coding process. Ratherthan contrastinginterjudgeagreementwith an estimate of chance agreement.and/or laws.other customers.andthe employee attempts enforce to and compliance. These are proceduresrecommendedby Perreaultand Leigh (1989) for improving the reliability of judgment-baseddata. matter her. Discrepancies between the first and third researchers'assignments were resolved by the second researcher.8 Group4-Problem customers . four categories emerged (Table 1 provides examples of incidents from the four new categories): 1. Ir was found to be .Furthermore. Cohen's K. however. Withinthe problemcustomerbehaviorgroup.. 4. The interjudgeagreementbetween the first and thirdresearcherswas 84% for the satisfying incidents and 85% for the dissatisfying incidents. respectively." In these cases. the more categories. of Employee-Reported Incidents The criticalincidentclassificationsystem based on incidents gathered from customers (BBT) consists of three major groups of employee behaviors that account for all satisfactory and dissatisfactoryincidents:(1) employee response to service delivery system failures. operationaldefinitions of each category. Breaking recompany policiesor laws-The customer fusesto complywithpoliciesor laws. customer unno whatis doneforhimor willingto be satisfied.
Because the employees and customersin these two studies all describeddifferentincidents. situationalfactors than to attributethe failure to their own shortcomings. The largestproportionof satisfactoryincidents. a customer need. Uncooperative Customer When a man was shown to his table in the nonview dining area of the restaurant.annoyingthe other passengers. He accepted the coffee the passenger refused. The smallest percentage of dissatisfactoryincidents were classified in Group 3. rudeness. Almost half of particularly satisfying customer encounters reportedby employees resulted from their ability to adjust the system to accommodate customer needs and requests.A relatively modest (when compared with the customerview) numberof satisfactoryincidents were categorized as unpromptedand unsolicited employee actions (Group3).Again. asked the familyto leave. hotel security staff finallybroke in. in a friendlyand sympathic way. people are more likely to blame external." they instead rememberthem in but association with a specific external cause (e. The flightattenthen hit another passenger. knockinghim into his seat. C. the and became quieterand friendlier. This makes sense. occurredin response to customer needs and requests (Group 2). Drunkenness An intoxicatedman began pinchingthe female flightattenA person who became intoxicatedon a flightstarted speakdants. lack of attention).again placing the blame on an external source. Success is attributedin None Satisfactory these cases to the employee's own ability and willingness to adjust.TABLE 1 Group Four Sample Incidents: Problem Customers Incident Dissatisfactory A.Although we rely on role and attribution Critical Service Encounters 99 / . Perhapsemployees do not view theirown behaviors as "spontaneous. The copilot then "decked" man. The restaurantwas very busy.The next largest proportionof satisfactoryincidents were categorizedin Group 1. This is an interestingset of incidents. They found the guests using drugs and called the police. because each one began as a failure but ended as a success because of the ability of the employee to recover. but he complainedall the way throughthe dinnerand left withouttipping. The copilot was called and dant asked the passenger if he would be drivingwhen the asked the man to sit down and leave the others alone.g. because it is difficult to imagine a very problematiccustomerleaving the encounter feeling satisfiedexcept underhighly unusualcircumstances. the explanationsare somewhat theospeculative.g.. When employees were asked to reportincidents resulting in customer dissatisfaction. In many of these cases. but the hostess told him he could get a window seat in a half hour. He refused to wait and took his previouslyreserved table. but he continuedand ing loudly. When things go wrong. The father knocked all the plates and glasses off the table before leaving. Employees clearly remembertheir ability to recover in failure situations as a significant cause for ultimate customer satisfaction.These results are not unexpectedgiven what attribution theorysuggests. B. but plane landedand offeredhimcoffee. there were virtuallyno satisfactory incidents categorized in the problem customer group (Group4).from the employee's point of view. One attendanttold him to stop. Breaking Company Policies or Laws Five guests were in a hotel room two hours past checkout None time. a service failure). D.conclusions from employee-customercomand parisonsare exploratory.By far the largestnumber inappropriate of dissatisfactoryincidents were categorizedin Group 1 (response to delivery system failures). with the next largest proportionfalling into Group4 (problemcustomers).the father None began hittinghis child. which reflects spontaneous negative employee behaviors (e. they tended to describe problemswith externalcauses such as the delivery system or customerbehaviors. Comparing Customer and Employee Views Table 3 combines data from the currentstudy with the original BBT data for purposesof comparison. Because they would not answer the phone calls or let the staff into the room. Verbal and Physical Abuse Whilea familyof three was waitingto orderdinner.he became extremelyangry and demanded a windowtable.Anothercustomer complainedabout this to the manager who then.A modest numberof dissatisfactory incidents were found in Group2. Finally. the employees implied that they were unable to satisfy customerneeds due to constraintsplaced on them by laws or theirown organization'srules and procedures. this is consistent with the bias towardnot blaming oneself for failures..
% 7.0 .8 16.it is possible that these differencescould be due to samplingvariationsor differences in the incident pool from which the two groups drew. Because of the significantthree-wayin100 / Journalof Marketing.0 0.This is clearly reflected in the observationthat customersreportno dissatisfactoryincidentscaused by their own problembehaviors(Group4).5 20. 2.Group4 ColumnTotal ries to explain the differenceswe observed. or spontaneous behaviors.1 22. the additionof a fourthgroup and the significantdifferences in frequencies and proportionsof incidents found in the groups suggest that there are dissimilaritiesin what they report as well. Gestalt evaluation E.8 9.3 39.0001).7 RowTotal No. will be likely to blame the employee ratherthan anythingthey themselves might have contributed. Employeebehaviorsin the context of culturalnorms D. thatboth studies were conductedin the same city using the same threeindustries. Truly out-of-theordinary employee behavior C. which predicts larger differences in perceptions in failure than in success situations.4 33.1 5.To potentially disruptive 196 Subtotal. type of outcome (satisfactoryor dissatisfactory).0 0.17. or 4). 37 48 110 195 14 43 0 5 62 6 28 3 0 0 37 16 9 16 42 83 377 % 9.3 0. Again.0 3. To admittedcustomer error 6 others D. which contains incidents reportedby employees only.2 51.4 1.TABLE 2 Group and Category Classification by Type of Incident Outcome (Employees Only) Type of Incident Outcome Satisfactory Group and Category No.1 18.8 0.Within the satisfactoryincidents.8 4.Group2 Group 3. Employee Response to Customer Needs and Requests 80 A.8 6.2 21.4 10. October1994 43 25 7 0 14 89 3 0 0 0 3 397 Dissasatisfactory No. X2change = 8.3 2. Problematic Customer Behavior A. To customer preferences 11 C.0 1. Breakingcompany policies or laws D.and thatmany of the same firms were the source of incidentsin both studies.8 51. Employees are highly unlikely to describecustomerdissatisfactionas being caused by their own predispositions.8 6.04).R.3 Group 1.8 0.8 0. Hierarchicallog-linear analysis of Table 3 shows a significantthree-wayinteractionbetween group (1. customers and employees have relatively similarproportionsin Groups 1 and 2. albeit less extreme. The significant interaction is caused by Group 3.9 2. this is consistent with attributiontheory. To unavailable 23 B.2 2.31.Group1 Group 2.4 .3 6. X2 change = 263.3 12.3 1.9 27. which is dominated by customer incidents. There is also a significanttwo-way interaction between group and incident source (L. Verbaland PhysicalAbuse C.8 1. Performanceunderadverse circumstances Subtotal. A large majorityof the employee incidentsfrom the current study could be categorizedin the original three groups and 12 categories.5 1. we have confidence in our theoreticalexplanations of the results.0 9. To "specialneeds" customers 99 B. Employee Response to Service Delivery System Failures service 31 A.7 3. .0 0. These results are very consistent with expectations based on attribution biases.2 2.8 1.3 1. on the other hand. Uncooperative customer Subtotal. 68 71 165 304 94 142 11 11 258 49 53 10 0 14 126 19 9 16 42 86 774 % 8.8 12.4 11. and incidentsource (employee or customer)(L. The significant interactionis the result of Group2 being dominatedby employee incidents and Group 3 being dominatedby customerincidents. Drunkenness B.3 6. 3. Groups 1 and 4 are equally representedfor both customers and employees.7 11.5 22.4 1.0 48.3 16.1 100% teraction.6 7.4 0.Recall that these are relativelyroutine service encountersand in both studies the respondentswere experiencedservice participants. To other core service failures 109 Subtotal.4 .2 24. The differences in how customersand employees report satisfactoryencounters are provocative as well.4 4.0 13. and Group 4.R. attitudes. p < .However.given the care takenin collecting the datato avoid systematicbiases. suggesting strong similaritiesin the way employees and customers reportthe sources of satisfaction and dissatisfactionin service encounters.7 29.2 11. Even so.5 49. To unreasonablyslow service 55 C.0 1. Customers. Within the dissatisfactoryincident classifications.the results are discussed separatelyfor satisfactory and dissatisfactoryincidents. p = . Unprompted and Unsolicited Employee Actions A. Attentionpaid to customer B.Group3 Group 4.
3 Employee Data CustomerData 347 49.3 42. such as the following: What types of problems do customers cause? What are the most frequentproblems?What types of customers tend to be problem customers? Under what circumstances do customerscreate either more or fewer problems? And. then the scheme can be viewed as more robust and of greatertheoretical as well as practical value.0 100% 100% for Implications Researchers Generalizability of the Service Encounter Classification Scheme The importance and usefulness of robust classification schemes for theory development and practical application have been discussed by social scientists (e. Hunt 1991.g.."The addition of this new group provides a more complete classification system that can be furtherexamined in other contexts.primarily because the classification schemes that have been proposedhave rarelybeen subjectedto empiricalvalidation across times and contexts. suggests the term "jaycustomers" label custo tomers who "misconsume"in a mannersimilar to jaywalkers who cross streets in unauthorizedplaces.0 0. and Davis 1993) and a study of 16 consumer services (Gremler and Bitner 1992).4 15.Booms. from a management viewpoint. Schrage 1992.3 33.. Othershave suggested the existence of problem customers (e. This study representsone contributionin a programof researchdesigned to test the validity and generalizabilityof a scheme for categorizing sources of service encountersatisfaction and dissatisfaction(BBT).6 11.9 16. differentindustrycontexts.6 9. customers versus providers. Problematic Customer Behavior 3 .2 33..2 16. This areais ripe with importantresearchquestions.7 42. Zemke andAnderson 1990). McKelvey 1982) and marketing scholars (e.0 48. Although no one really believes customers are always right.8 41. % Group 1. Other studies have reportedthat the three major groups of behaviorsidentified by BBT are also found in a retail context (Kelley.8 Employee Data CustomerData 0 0.3 2. or in internalas well as externalencounters)and across differentrespondents (e.0 Column Total 397 51.g. 304 232 258 169 126 298 86 0 774 699 % 39.6 aCustomerresponse data fromBitner. Hoffman. Our research provides empirical evidence that these difficult customer types do exist and in fact can be the source of their own dissatisfaction. for example. 195 151 62 55 37 146 83 0 377 352 % 51.g. Lovelock.4 Employee Data CustomerData 114 32. and Tetreault (1990) Disasatisfactory No.7 50. Lovelock 1994. Employee Response to Customer Needs and Requests Group 196 49. Needless to say...g.3 24. such avoidance leads to stresses and strains for managers and frontline personnel alike and potentially bigger problems for firms. customers in differentcultures). Through replication.g. (See Hochschild 1983 for a discussion of personal and organizationalimpacts of nonauthenticways of dealing with customers.4 Row Total No. firms have policies that pretend this is so. and managers urge and demand that customer contact employees treatcustomers as if they are always right.the frameworkbecomes more valuable in identifying generalizable"service behaviors. what can be done to identify problem customers. and how can and should employees deal with them? This initial researchrepresentsa startat addressingsome of these questions and the beginnings of a typology of probCritical Service Encounters/101 .9 Group 3.5 22.TABLE 3 Comparison of Employee and Customer Responses: Incident Classification by Type of Incident Outcomea Type of Incident Outcome Satisfactory Groups No. If the scheme holds in different settings (e. Problem Customers A primarycontributionof this researcheffort is the empirically based finding that unsatisfactory service encounters customer behaviors-the nomay be due to inappropriate tion that sometimes customers are wrong.5 Employee Data CustomerData 81 23. Employee Response to Service Delivery System Failures 109 27.)With a better understanding of customerscan come bettermethodsfor eliminating problem or dealing with the underlyingcauses of the problems.8 Group 4.1 0. Yet we have few such frameworksin marketing.4 Employee Data CustomerData 152 43. Lovelock 1983). Unprompted and Unsolicited Employee Actions 89 22." The resultsof ourresearchindicatethatall the categories found in the original customer-perspective study were also found when employees were asked to reportexcept "problem customers.
Theory Implications Role and scripttheories suggest thatcustomersand employees in routine. 102 / Journalof Marketing. These disparitiesshow up in the distributionof incidents across the major groups. In these cases. nor will he or she always behave in acceptableways. the companies began with the existing groups in the classification scheme and fleshed out the categories with useful specifics that could be employed in service training or service redesign. Sharedviews of the encounter should result in common notions of the sources of customer satisfactionand dissatisfaction. customers frequentlyhave transaction-basedencounterswith the service personnelratherthan encounters. as mentioned previously. some Madison Avenue ad agencies say that "some accounts are so difficult to work with that they simply cannot-or will notservice them"(Bird 1993). althoughwe believe that the majorproblem customergroup will surface whenever employees are asked to relate instances of dissatisfactory encounters. In each service the customers are in close physical proximity for extended periods of time. unexpected peaks in demand. Thus.lem customer behaviors. the framework is now more complete. they may not be good relationship customers for the organizationbecause they do not meet the targetmarketprofile.g. but frequentlythey are being told that the "customeris king" and are not given the appropriate trainingand tools to deal with problemcustomers. furtherresearch is needed to identify other subcategories within the group and relate problem types to service industryconditions. Employees can also be taughtto recognize characteristicsof situations(e.. some are beginning to acknowledge these facts more explicitly and are attemptingto quantifythe impactof problem or "wrong"customers on profitabilityand organizational stress. The types of service encounters studied here and in the original study do representfrequently encounteredand routineservices. October1994 The self-serving attributionbias suggests explanations for why some of these differenceswere observed.airline.legal services). Personal social interactionsare carried out in front of other customers who are most often strangers.It is assumed that long-term relationship-based these circumstances influenced the nature of the subcategories of problemsidentifiedin Group4. the types of encountersstudiedhere are all relativelyroutineand commonly experienced. For examhas ple. some car rentalcompanies have attemptedto refuse customers with bad driving histories by checking recordsin advance and rejectingbadrisk drivers(Dahl 1992). Several implications are suggested by the problem customer group. Through the addition of the problem customer grouping. and the differences were most dramatic for the dissatisfactory service encounters. managers must acknowledge that the customer is not always right.The fact that 89% of the employee incidents could be classified in the original classification scheme suggests that customers and employees do indeed reportincidentswith most of the same sources of satisfactionand dissatisfaction. even when customers do not misbehave. inordinatedelays) and anticipate the moods of theircustomersso thatsome potentialproblem situations can be avoided completely or alleviated before they accelerate."For example.theremay be other types of "wrongcustomers. Zeithamland Bitner 1995). and hotel customersare many times in tight public spaces that put them cheek to jowl with other customers. they are not profitablein the long term.. It is beyond the scope of this article to discuss the full conceptualizationof in wrong customers. The categories of behaviors discovered are not surprisinggiven the natureof the industries studied.thereare also significantdifferences. The Customer Is Not Always Right In the industriesstudied here. .A basic problemcustomerstrategymight be conceptualizedas ranging along a continuumfrom "refuseto serve them"to "satisfy them at all costs. the framework been used for proprietary purposesin medical and travel agent contexts. Each service involves the possible serving of food and drink-including alcoholic beverages. Restaurant. Finally. Although we have identified problem customersby exploring the sources of customerdissatisfaction.g. and the scheme itself can provide a startingpoint for a company or industry to begin identifying with greater specificity the events and behaviorspeculiarto its own setting. In a differentcontext.circumstances.And. Implications Managerial Using the Classification Scheme One purposeof this study was to evaluatethe soundness of the classification scheme developed by BBT in a distinctive context. An interestingissue for furtherresearchis whether the overall strong similarity of views between customers and employees would result if the industriesstudied were ones in which the scripts were less routineand well practiced. health care. well-understood service transactions will shareparallelviews of theirroles and the expected sequence of events and behaviors.or in some cases they may not be compatiblewith the service provider in terms of personality or work style (Lovelock 1994. problem customers were the source of 22% of the dissatisfactoryincidents."For example.but it may be fruitfulfor researchers the future to incorporatethe misbehaving customers we have identified into this more extensive conceptualscheme. education. First. Contactemployees who have been on the job any period of time know this. Results of the study indicate that though employees and customers do reportmany of the same sources of customer satisfactionand dissatisfaction. To provide employees with the appropriate trainingand skills for workingwith problemcustomers. Although organizationshave intuitivelyrecognized that not all customersegments are right for the firm and thateach individualcustomeris not rightall the time.and customer segments. Employees need appropriate coping and problemsolving skills to handle customersas well as their own personal feelings in these situations.the organization must clarify its position regardingsuch customers. This group may be even largerin industriesin which the customerhas greaterinput into the service delivery process (e.
reThus. typesof were interactions questionnaires used. respondents employees of restaurants. Shostack 1984.An incident must: (A) include employee-customerinterac- Critical Service Encounters /103 .and Berry 1988. This pride comes throughin the large percentage of satisfactoryincidents found in Group2. We also learned from employees that customers can be the source of theirown dissatisfactionthroughinappropriate behavioror being unreasonablydemanding. Zeithaml. The best way to ensure satisfaction. When employees have the skills and tools to deliver high-qualityservice. In addition.However.real-timechallenge for service employees. Our study confirmsthese findings insofaras employees of hotels. You will be askedto categorizeeach incidentinto one of 16 categories. It is suggested that you read througheach entire service encounterbefore you attemptto categorizeit. put it aside. of First.Youwill be provided witha set of written criticalservice encounterevents. It has been suggested that by treatingcustomers as "partialemployees" they can learn to contributeto the service in ways thatwill enhancetheirown satisfaction(Bowen 1986). In other more complex and less familiar service situations (e."appreciateinformationon what to wear and otherexpected behaviors while at the resort. theywere asked takethecustomer's to pointof viewin responding to the questions.This result. based on the key factor that triggeredthe dis/satisfactoryincident. although they report the same basic categories. Parkingtonand Buxton 1980). or lack of knowledge. 1990).oneforsatisfying andone fordissatisfying interactions. Employees as Sources of Customer Data Previous researchhas suggested that contact employees are good sources of informationon customerattitudes(Schneider and Bowen 1985. 3. Conclusion The research suggests that many frontline employees do have a true customer orientationand do identify with and understandcustomer needs in service encounter situations. 1987).g. may considerably deepen understandingof and ability to cultivatecustomerrelationships. Eachserviceencounter reflects events the questionnaire andbehaviors associated anencounter is memwith that orable because is either it or particularly satisfying parThe were ticularly dissatisfying. Second.Beyond the need to develop employee skills. Oftentimesthe inabilityto do so is governed by inadequateor poorly designed systems.and airlines report all the same categories of customer satisfaction and dissatisfaction reported by customers in the same industries. acknowledgmentthat wrong customers exist.or the lack of authorityto do anything. 2. customers may truly appreciateknowing more about their role in the service process and the behaviorsand informationthat are needed from them to make the service succeed (Bloom 1984). Appendix A Instructions for Coders Overview 1.Sortingrules and definitionsof categories are detailed below. there is the need for "training" customersso thatthey will know what to and appropriate behaviorsin given situations. the proportionsof incidents found in the categories are significantly different from those reportedby customers.inability to provide a logical explanation to the customer.do not attemptto categorize incidents that do not meet the basic criteria. restaurants. If an incident does not appearto fit within any of the 16 categories. Each "story"or "event"is recorded a standardized on Two questionnaire. 4. contact employees may not be as accuratein their assessment of customer expectations and satisfaction (see Brown and Swartz 1989). is not to have a failure in the first place.however. poorly designed systems or procedures. we would caution against relying too much on contact employee interpretations customer satisfaction for two reasons. who may not be accustomedto the "rulesof behavior.We suspect that this new group of dissatisfactoryincidents caused by problem customers would surface in any service industry and that its existence representsa strategicchallenge for the organizationas well as an operational.the datareflectemployees' membrancesof times when customers had particularly dis/satisfying encounterswith their firms.For exexpect some upscale resorts that offer highly discounted ample. Systems can then be redesigned and processes implementedto ensure higher reliability from the customer'spoint of view. poor or nonexistent recovery strategies. professional services). abilities. Reliability Is Critical The data show thata majorityof the dissatisfactoryincidents reportedby employees resulted from inadequateresponses to service delivery system failures. In a time when "customeris king" is the stated philosophy of most forward-thinking organizations. However. Balancing out this sense of pride are a large number of frustratingincidents in which employees believe they cannot for some reason recoverfrom a service failureor adjust the system to accommodatea customerneed. they are proudof their ability to do so. in which employees' own skills.. Schneider. Employee Desire for Knowledge and Control It is apparent readingthe incidentsthatcontactemployees in want to provide good service and are very proud of their abilities to do so. im- plies a need for service process and system analysis to determine the root causes of system failures (KingmanBrundage1989. cumbersome bureaucraticprocedures.togetherwith otherresearchreportingservice reliabilityas the single most importantdimension used by consumers to judge service quality (Parasuraman. in some industriesin which service encountersare less routine. They have respect for customersand a desire to deliver excellent service. rates in nonpeak seasons find that their discount customers. airlinesandhotels. These reasons usually stem from lack of basic knowledge of the system and its constraints. coupled with creative thinking about customer roles and managementof customer expectations. and willingness to accommodate customer needs were the sources of customer satisfaction.
(C) employee behaviorsin the context of culturalnorms. Then. larly extraordinary or profanity. is the customer asking (either explicitly or implicitly) that the system be somehow adjustedto accommodatehim/her? Is it the employee's response that causes the event to be rememberedas highly satisfactoryor dissatisfactory? If the answer is yes. (C) other core service failures.inappropriate touching. (D) gestalt evaluation. Then.. Group 2. dietary. Response to unreasonablyslow service (services or employee performances are perceived as inordinately slow). place the incident in Group 1. Employee behaviors in the context of cultural norms (norms such as equality. it obviously implies that there is no service failure and no explicit/implicitrequest. Response to "specialneeds"customers(customerswith medical.elderly customers). 1041 Journalof Marketing. not triggered by a service failure.(B) verbal/physicalabuse. Gestaltevaluation(no single featurestandsout. 1. A. (C) breaking/resisting company policies or laws. fairness. If the answer is no. and show no evidence of the customer having a special need or making a special request). hotel room not clean.(B) truly out-of-theordinaryaction. theft. incorrectorder. Response to customerpreferences(when the customer makes "special"requests due to personal preferences. honesty. damagedbaggage). Unprompted and unsolicited employee actions (events and behaviors that are truly unexpected from the customer's point of view.. place the incident in Group 3. Then ask.(E) exemplaryperformanceunder adverse circumstances. Is there an explicit or implicit request or need for accommodation or extra services(s)? That is. place the incident in Group 2. go to question 4. intoxication. instead "everythingwent right"or "everythingwent wrong"). e. Response to potentially disruptiveothers (when other customers exhibit behaviors that potentially strain the encounter. use the triggeringevent. place the incident in Group4. A. the flight. e.. D. e. system failures). (C) be a discrete episode. does a spontaneousaction or attitudeof the employee cause the dis/satisfaction? (Since this follows rules 1 and 2. what type of failure?(A) unavailableservice.rudeness). Response to unavailable service (services that should be availableare lacking or absent. Employee response to customer needs and requests (when the customer requires the employee to adapt the service delivery system to suit his/her unique needs. go on to question 3. is the root cause actually the customer? If the answer is yes. Truly out-of-the-ordinary employee behavior (particuactions or expressions of courtesy. Group 3.g. ask what type of unpromptedand unsolicited action took place: (A) attentionpaid to customer. Attention paid to customer (e. (B) unreasonablyslow service.deviance). (B) be very satisfying or dissatisfying from the customer'spoint of view. Definitions of the categories are attached. regulations.violations of basic etiquette. CIT Classification System-Definitions Group 1.g. is there an initial failureof the core service thatcauses the employee to respondin some way? Is it the employee's response that causes the event to be remembered as highly satisfactoryor dissatisfactory? If the answer is yes.g..tion. this includes rudeness. making the customer feel special or pampered. lost tickets.rudeness. or refraining from the above when such behaviorwas expected).e. Employee response to service delivery system failure (failure in the core service. or the service provider. ask what type of behavior is causing the problem: (A) drunkenness. and (D) have sufficient detail to be visualized by the interviewer.overbooked airplane. A. (B) customer preferences. 4. this includes times when the customer requests a level of service customizationclearly beyond the scope of or in violation of policies or norms).) C.ignoring or being impatient with the customer). go on to question 2.g. put the incident aside. D. C.).children.g. the hotel room. B. lost hotel room reservation. Then ask what type of need/requestis triggering the incident: (A) 'special needs' customer. abusiveness. If the answer is no. 2. B. (D) uncooperative customer. (C) admitted customer error. Exemplary performanceunder adverse circumstances (when the customer is particularlyimpressed or displeased with the way an employee handles a stressful situation). you should begin asking the following questions in orderto determinethe appropriate category. October1994 . Is there a service delivery system failure? That is.missed reservations. contains either an explicit or inferred request for customized [from the customer's point of view] service). Response to othercore service failures(e. or a general unwillingness to indicate satisfaction with the service regardless of the employees' efforts).psychological. restaurant meal cold or improperlycooked. C. Problematic customer behavior (customer is unwilling to cooperate with laws. Is there an unpromptedand unsolicited action on the part of the employee that causes the dis/satisfaction? That is. the restaurant meal service. Response to admittedcustomererror(Triggeringevent is a customer error that strains the service encounter. 3. E.g. Once you have read the incident. language. or sociological difficulties.unavailablereserved window table). discrimination. lying. (D) potentially disruptive other customers. Does the dis/satisfaction stem from the actions/attitudes/behaviors of a "problem customer"? That is. Coding rules Each incident should be categorized within one category only. If the answer is no. If the answer is no.) If the answer is yes. (Note: When service is both slow and unavailable.. ratherthan the dis/satisfaction being attributable an to action or attitudeof the employee.. Group 4. B.
Parasuraman New York:The Free Press."Annual Review of Sociology. James L. Drunkenness (in the employee's perception. Robert Dwyer. Bird. W. Chicago:American MarketingAssociation. Scott W. "The Reliability andValidityof the CriticalIncidentTechnique: Closer Look. and Leonard A. and Mary StanfieldTetreault (1990). and Pierre Eiglier (1981). "MutualUnderstanding Between Customersand Employees in Service Encounters. REFERENCES (1991). Jr. 6 (Spring).g.OH: South-Western PublishingCompany."WallStreetJournal (October23). Jr. customermay appearunwilling to be satisfied. 71-84. Evolution.47 (Summer)."Psychological Review. 2 (2). 371-83. (1986).A. Holman and Michael R. . "ClassifyingService EncounterSatisfactionAcross Industries.Benjamin (1980). the Crosby. Inc. 67-92. Organizational Systematics: Taxonomy. 82-101. Taylor (1984). Bateson. and BarbaraKotsos (1986). 12. B. Ericsson. StephenW.Berkeley.LeonardL. Christopher (1983)."International Journal of Service Industry Management. "The ABC's of Service System Kingman-Brundage. 146-57." Harvard Business Review (March/April). Provo. and MarkA. (1986). Uncooperative customer (customer is generally rude and uncooperativeor extremelydemanding. "Managing Customers as Human Resources in Service Organizations. Goals and Understanding. Abelson (1977). McKelvey.New York: John Wiley and Sons. 429-52."Journal of Retailing. Cambridge.and the employee has to handle the situation). Rebecca H. D. Services Marketing:New Insightsfrom Consumers and Managers. Laura (1993). Bloom. Heskett (1991). and Gary P. George and ClaudiaMarshall. "Putting the Service-Profit Chain to Work. Development and Implementationand the Employee. Folkes." Journal of Retailing. William D. "ConsumerComJourplaints and ManagerialResponse:A Holistic Approach.52-65. C. Service Breakthroughs. Latham (1974). Scripts. Schneider. 25 (3). Berry (1988). Classification.G. Langeard. MA: Marketing Science Institute.MA: MarketingScience Institute. Dahl." Harvard Business Review (September/October). Paul N. New York:The Free Press..47 (Winter). Jones.A. MA: Addison-Wesley. 87 (May).611-17.86-97."OrganizationalDynamics (Autumn). William R. "The Service Encounter: Diagnosing Favorable and Unfavorable Incidents. Cambridge.Cincinnati. Gary W.any efforts to compensatefor a perceivedservice failureare rejected. and James L." nal of Marketing. Arlie Russell (1983). John E." Designing a WinningService Strategy. Simon (1980). Anders and HerbertA." in Advances in ConsumerResearch." Critical Service Encounters /105 . Bill (1982). Bowen (1984). Social Cognition. Biddle. Earl Sasser. 54 (January). Leigh (1989). "EffectiveMarketingfor ProfessionalServices.LawrenceA." WallStreetJournal (November 2). LeonardA.forthcoming. William W. 215-50. Jr. Fiske. Hochschild. 69 (4). "Rental Counters Reject Drivers Without Good Records. Lovelock." Human Resource Management. Lovelock. Hunt. Crosby. BernardH. 92-98.. Harmon(1983). "A Typology of Retail Failuresand Recoveries. 12-40. 15 (Summer). Booms. CA: University of California Press. 5-19."in Developing New Services. and the employee has to enforce compliance). Jane (1989). B1. Doroff. Plans. Loveman.J. B 1. 53 (April).Lois A.50 (April)." of MarketingResearch. Breaking/resistingcompany policies or laws (the customer refuses to comply with policies [e." MarketingTheory in and Applications. Valarie Zeithaml. "The Clients That Exasperate Madison Avenue. Business Week(1991). The Managed Heart. "Employee and Customer PerJourceptions of Service in Banks:Replicationand Extension. "Verbal Reports as Data. and Valarie A. (1991). David E. ChristopherH. Bowen. Chicago:AmericanMarketingAssociation. Berkeley." Harvard Business Review 71-81.John C. "The Service Organization:Climate Is Crucial. Ronald J. Valerie S. 51 (July). andA. Product Plus. (1994). Roger C. Journal ity of Nominal Data Based on QualitativeJudgments. ChrisT. K. Kelley. Resnik." The Journal of ConsumerAffairs.. . . 74-80. and Leonard L. and (1985). the customer is clearly intoxicated and creatingproblems. and Robert P. Moder MarketingTheory.and LaurenceE. Ronan. "Buyers' and Sellers' Explanationsfor ProductFailure:Who Done It?"Journal of Marketing. eds. 64 (Spring). Bitner." Journal of Marketing.Dwayne and MaryJo Bitner (1992). "Classifying Services to Gain StrategicMarketingInsights. and F. MarketingServices. Schlesinger (1994). 9-20. Chicago: American MarketingAssociation.Eric. eds. eds. and the employee has to handle the situation). Davis (1993). Mary Jo. eds. and (1990). Heskett. Special Issue on Quality. in Blueprinting. 26 (May). Alan J. Shelby (1991). (1954)."Journal of Retailing. (1984).L." Report No. Schank.Mary Jo Bitner and Lawrence A. A "SERVQUAL: Multiple-ItemScale for MeasuringConsumer Perceptions of Service Quality. and ChristopherW. Reading. "Expanding Role of CSM in Total Quality. Allen et al. Douglas Hoffman. Gremler. 102-10. Thomas O. and Shelley E. 67 (4). and Robert R. 111-18. "The CriticalIncidentTechnique. use of illegal drugsin the hotel room]. 18.164-72." Psychological Bulletin. W. (1989) "A Gap Analysis of Professional Service Quality. "Refinement and Reassessment of the SERVQUAL Scale. 135-48. and David E.Verbal and physical abuse (the customer verbally and/or physically abuses either the employee or other customers. Zeithaml (1991). 30-33. "Recent Developments in Role Theory. 327-58. "ReliabilPerreault. CA: University of CaliforniaPress." A Studies in PersonnelPsychology. Schlesinger.and the employee has to handle the situation). "AnEmpiricalExamination of Relationships in an Extended Service Quality Model. and TeresaA. Berry... Flanagan. showing airplaneticket to the flight attendantbefore boarding] or laws [e. Swartz. Parasuraman. and MaryJo Bitner (1991).g. 90-122. 53-64... (1981) "PerceptualDifferences in MarketTransactions Revisited:A WaningSource of Consumer Frustration. K. "New Services Design. Earl Sasser. B. "The Service-Driven Service Company. 420-50."Journal of Marketing. Susan T..Vol. Brown. Solomon. Hart (1990). UT:Association for ConsumerResearch. Berry."Journal of Marketing. New York: McGraw-Hill. Leonard L. (September/October). Jonathan(1992).. Mohr.
18 (4).invited addressto the Annual Meeting of the American Educational Research Association.52 (April). Locke (1981). "PerceivedDeterminants of High and Low Productivity in Three Occupational Groups: A Critical Incident Study. lyn G. 1.. "Employee and Customer Perceptions of Service in Banks."Journal of Marketing.99-111. and EveSolomon. "A Role Theory Perspectiveon Dyadic Interactions: The Service Encounter. G. Parkington. (1987). Lynn (1984).34-43." Journal of Marketing. October1994 . 423-33. 375-87. A8." HarvardBusiness Review (January/February).forthcoming. Ron and KristinAnderson(1990)." AmericanInstitutesfor Research. Carol Surprenant. . and EdwinA. John J.nal of Applied Psychology. 70 (3). White. Wilson-Pessano. "Recent Study Shows Gap Between Customersand Service Employees on CustomerService Perceptions." Journal of Marketing. and Mary Jo Bitner (1995). Czepiel. 106 / Journalof Marketing. Zeithaml. Services MarketingNewsletter (1989). (1988)." (Summer). Leonard L. Buxton (1980). Berry.25-33. "Communication ControlProcesses in the Delivery of Service Quality. and A. FrankM. "Customers Training(February). "Service Positioning Through Structural Change." AdministrativeScience Quarterly.. New Orleans. and Virginia M. Schrage. JohnA.49 (Winter)."WallStreetJournal (March 16).New York:McGraw-Hill. Gutman(1985). SandraR. "Designing Services That Deliver. Michael R. Services Marketing. Parasuraman and (1988). Michael (1992). 252-67. "FireYourCustomers.51 (January). 133-39. "Defining Professional Competence: The Critical Incident Technique 40 Years Later.25 (June). Valarie A." Zemke. 35-48."Journal of Management Studies. 5 Shostack. From Hell.
This action might not be possible to undo. Are you sure you want to continue?