CUSTOMERS LEAVE - University of Auckland Business Review

WHYCUSTOMERSLEAVEOR DECIDENOT TO LEAVE –40

Much research has focusedon how to improve a service provider’s qualityof service and how to attract and maintaincustomers (particularly with the use of servicerecovery techniques).A major factor that affects customers’perception of a service provider is the numberand extent of the problems they encounter andhow these problems are handled.A problem a customer has with a service canbe called a “service failure”. When customersencounter a service failure, they can either notcomplain, or complain and give the serviceprovider an opportunity to rectify the problem.If they do not complain, they may remain withthe provider, despite their dissatisfaction, or exit.If they do complain, they may also choose tostay or leave, and this is influenced by how thesituation is handled by the service provider (i.e.the service recovery).Service recovery includes all actions takenby a service provider to try to resolve acustomer’s problem.There is a strong focus in the literature onmanagerial ways to improve service recovery.For example, much research looks at whataspects of service recovery encourage acustomer to stay. This plethora of literatureleads us to believe that service recovery is themost important factor in driving a customer toexit or stay after a service failure. This, however,is not the case.The aim of this paper is to generate a model ofhow customers react to service failure by gaininga more detailed understanding of what drives acustomer to remain or exit after a service failure.It is important that we compare customers inthe same service context (i.e. same industry andsame segment of customers) so that we obtainmore comparative results. The extant literaturedoes not investigate this area fully, but insteadtends to focus on one dimension of this researchfield (e.g. service recovery or exit).This leads us to the research question: Why dosome customers exit their service provider aftera service failure, while others stay?▼THEIR BANKBY MARK COLGATEAND MELISSA NORRISUNIVERSITY OF AUCKLANDBusiness ReviewVolume 2 Number 2 200041

Spurious loyalty … has beenneglected by managersand academics alikeThis paper is broken down as follows:• An examination of the literature on all aspectsof service failure.• Propositions for the study, based on theliterature review.• The results of interviews with customers whohave stayed after a service failure and withcustomers who have exited.• Development of a model, based on theresults of the interviews, that aims to presenta more complete picture of the outcomes ofservice failure.• Implications of this research, both academicand managerial.SERVICE FAILUREThe differing reactions customers may haveto service failure have been discussedthroughout the literature. These include loyalty,voice (e.g. complaining to the service provideror another party) and exit.LOYALTYA service provider develops loyalty overtime by consistently meeting, and sometimesexceeding, customer expectations.Loyal customers are good for business. They“often cost less to service, spend more as theirtime with the firm lengthens and provide agood source for new business” (Levesque andMcDougall, 1993: pg 40). Customer loyaltyusually leads to lower marketing costs, moreefficient operations and higher profits.Loyalty comes in several differing forms,including pure loyalty, latent loyalty andspurious loyalty (Dick and Basu, 1994).Pure loyalty is patronising a company often aswell as thinking highly of it. This is the kind ofloyalty most people associate with the term.Latent loyalty is when a customer has lowrepeat patronage, but thinks highly of theorganisation.Finally, spurious loyalty is the opposite – acustomer has high repeat patronage, but arelatively low opinion of the company.Hirschman (1970) describes this as “remainingwith a company despite dissatisfaction”. It isthis notion of spurious loyalty that is of themost interest as it has been neglected bymanagers and academics alike.Initially, the choice by a spuriously loyalcustomer to remain with a provider does notthreaten profits, but with continueddissatisfaction they may spread negative wordof-mouthor exit (Hirschman, 1970).Companies need to understand why acustomer still patronises them, as it may notnecessarily be a positive, conscious choice.Customers may stay with an organisation evenif they are dissatisfied because they perceivethey have no choice. Hence, a “loyal” customermay not necessarily be a satisfied customer.The findings from a study conducted byLevesque and McDougall (1993: pg 52)suggested that “even when a problem is notsolved, approximately half of the respondentswould remain with the firm”. This may be due toswitching costs, lack of perceived differentiationof alternatives, locational constraints on choice,time or money constraints, habit or inertia.Hirschman (1970) also proposes thatdissatisfied customers may choose to remainwith a service provider because they think thatthe likelihood of an improvement outweighs thecost of searching for another supplier.The literature suggests, then, that loyalty isone of the responses a customer makes to aservice failure. But we should not believe thatloyalty is one-dimensional. Customers may staywith a service provider because they generallyhave a high opinion of the provider and see theservice failure as an aberration. Or they may bespuriously loyal – they feel trapped, areapathetic or feel that there are no alternatives.42

W HY CUSTOMERS LEAVE – OR DECIDE NOT TO LEAVE – THEIR BANKVOICEThe terms “voice” and “exit” wereintroduced by Hirschman (1970) when hesuggested that dissatisfaction can provoketwo active negative responses: voice and exit.Day and Landon (1977) extended thenotion of “voice” by explaining that it can becomplaining to the service provider, complainingto acquaintances (negative word of mouth), orcomplaining to third parties to seek redress.Any problem a customer has with theirservice provider will arouse some emotion,whether they complain about it to the provideror not. If they do make such a complaint, theygive the service provider a chance to rectify theproblem. A customer must have sufficientlystrong emotions about a situation before theymake an effort to complain, however.High transaction costs are another reasonconsumers choose not to complain. “These[costs] include the time, cost and effortinvolved, uncertainty about how to complain,and beliefs that redress will not be provided”(Day and Landon, 1977).A complaining customergives the companyan opportunityService providers must provide open lines ofcommunication if they want customers tocomplain, so they can try to solve their problems.If a customer “voices” by any other means, suchas complaining to a business colleague, or exits,it is difficult for the service provider to find outwhy that customer was unhappy.The literature identifies customercomplaints (or lack of) as an integral part ofservice failure. A customer who does notcomplain may either exit and give theAny problem a customerhas with their serviceprovider will arousesome emotionorganisation no chance to “save” thecustomer, or stay and “fester”. Neither isdesirable. But a complaining customer givesthe company an opportunity, which should beseen as such.STAY OR EXIT?A customer’s final decision after a servicefailure is to stay or exit. To exit in response toa service failure, the customer must bemotivated and must expend effort (Singh,1990; Stewart, 1998).Customer exit can cost the seller a great dealthrough decreased income from the customer’sfuture spending, higher costs in attracting newcustomers, loss of free advertising through wordof mouth, and decreased employee retention.The benefits of customer retention have alsobeen emphasised. For example, Reichheld andSasser (1990) found that by retaining five percent more customers, a service provider canincrease profits by almost 100 per cent.Many factors influence a customer’s reactionwhen they face a problem with their serviceprovider. This paper uncovers what thesefactors are and how they interact, leading to thedecision by the customer to leave or stay.SERVICE RECOVERYIf the above reactions to service failure, voice,when it involves complaining to the serviceprovider, creates the potential for the customerand the organisation to resolve the problem.Gronroos (1990) defined service▼UNIVERSITY OF AUCKLANDBusiness ReviewVolume 2 Number 2 200043

If service recovery is not regardedas important or handledwell, customer satisfactioncan be erodedrecovery processes as “those activities in whicha company engages to address a customercomplaint regarding a perceived servicefailure”. It encompasses all the actions taken toget a disappointed customer back to a state ofsatisfaction.Several reasons are posited throughout theliterature for the growing tendency forcompanies to invest in service recovery efforts. Ifservice recovery is not regarded as importantor handled well, customer satisfaction can beeroded. Ultimately, the service provider’sprofitability may diminish.Results from Boshoff (1997) support this. Heshowed that poor service delivery cannot becompensated for by great service recovery,although service recovery can help significantly inlimiting the harmful impact of poor service delivery.Stewart’s model, although most insightful,focuses solely on a customer’s decision to exit.It needs expansion when applied to the servicefailure process as a whole.Boshoff (1997) introduced a model of servicerecovery, not just the customer exit process. Heattempted to bring together all the aspects ofservice recovery identified by earlier authors.The model also depicts, to some extent,customers’ different types of reactions todissatisfaction. Customers are broken down into:• Those who have seriously considered leavingthe service organisation and do not complain,but value loyalty, so they stay.• Those who do not complain, but just exit.• Those who do complain and, depending onthe outcome of their complaint, either staybecause of satisfaction or loyalty, or leave.Despite the importance of these models otherresearch suggests they are incomplete. Ourresearch attempts to tie together all thedisparate areas of research. Only then can theoutcomes of service failure be fully understood.MODELS OF THE SERVICERECOVERY PROCESSTwo models directly relate to this paper:Stewart’s (1998) customer exit process modeland Boshoff’s (1997) service recovery model.Stewart developed a comprehensive model ofthe customer exit process, designed to helpmanagement understand the process. She notesthat customers usually make complaints beforeexit and go through a series of emotions.The availability of alternatives and therelationship the customer has with the serviceprovider are also considered, as well as thecomplaint handling, before exit is invoked.She suggests that by the time a customer hasdecided to exit an organisation, it is too late torecover them. Often they are leaving because ofthe organisation’s lack of responsiveness. If theorganisation does respond when they havedecided to leave, this is viewed as ironic.SERVICE FAILURE OUTCOMESAcustomer may stay after experiencing aproblem because they have complainedand they are satisfied with the service recoveryprocess. This is the ideal situation for serviceproviders who plan for good service recovery.Many times, however, customers stay eventhough they are dissatisfied with the servicerecovery they have received. This spuriousloyalty may result from a customer’s perceptionthat the barriers to exit are high and/or that thequality of alternative providers is no better.The spuriously loyal are “prospective switchers”.Given a change in the rules (e.g. more alternativescome along), they may leave immediately.The other outcome, when a customer experiencesa service failure, is to leave. They may havecomplained, but the service provider’s recoveryefforts may have been poor, causing the customerto be even more dissatisfied and leave.44

W HY CUSTOMERS LEAVE – OR DECIDE NOT TO LEAVE – THEIR BANKAlternatively, the customer could besatisfied with the recovery they havereceived, but still exit. This could be due toa perceived lack of barriers to exit, a senseof apathy about which provider they use,variety seeking, or the feeling that therecovery could not make up for theproblem they have experienced.Hence, service recovery may not be the mainreason why a customer remains or exits.Furthermore, service providers do not alwaysget the opportunity to conduct service recovery,as they are not always aware that a problem hasoccurred. A customer may experience a problem,not complain, and stay. This could be becauseof loyalty towards their service provider, orspurious loyalty. Or a customer may leavewithout complaining.In light of the literature review, six propositionsare presented:P 1 = Customers exit without complaining, thusoffering the service provider no chance toconduct service recovery.P 2 = Customers exit after they have complained,because they have received poor service recovery.P 3 = Customers exit after they have complained,even though they have received good servicerecovery.P 4 = Customers stay with a service providereven though they encountered a service failureand did not complain.P 5 = Customers stay with a service providerService providers are notalways aware that aproblem has occurredwhen they complained and received goodservice recovery.P 6 = Customers stay with a service providereven though they complained and receivedpoor service recovery.These can been seen in Figure 1.METHODOLOGYQualitative research is the most appropriatefor these propositions. Research into theservice failure process is inconclusive andadditional outcomes of service failure mayexist, given the absence of research in this area.Also, quantitative research does not allowcustomers to express their true feelings about asituation, so it may not be possible tounderstand what has really happened withoutthe use of qualitative research.We looked at small-business customers of thebanking industry. Banking for this group has astrong relationship/loyalty aspect to it, makingthe customer’s decision to exit a harder, morethought-involving process.▼FIGURE 1ProblemExitRetentionP 1 NoComplaintP 2 - PoorServiceRecoveryP 3 - GoodServiceRecoveryP 4 - NoComplaintP 5 - GoodServiceRecoveryP 6 - PoorServiceRecoveryUNIVERSITY OF AUCKLANDBusiness ReviewVolume 2 Number 2 200045

There was a general feelingof dissatisfaction withthe banks’ servicerecovery effortsAs well, small-business customers are arelatively unresearched group, making thefindings from this research more important.Narrowing down our focus made it easier tocollect data.Respondents fall into two different groups:those who left their bank after a service failureand those who remained with their bank afterseriously considering leaving (a service failurehad to have occurred).We adopted Stewart’s (1998) definition (witha slight revision), so that to leave meant thatthe customer had shifted their main account toanother provider. A main account is defined as“the account where most of your transactionstake place”.Those who remained but seriouslyconsidered leaving still kept their main currentaccount and did most of their transactions withthe service provider. The problem did notadversely affect the amount of business theyhad with that provider.Most of the studies conducted to date aboutservice failure tend to ask behaviouralintentions questions. They ask what customersexpect will happen in a service failure situation.It has been proven that this style of researchdoes not accurately reflect what customersactually do when faced with the problem.Therefore, this study investigates those customerswho have actually remained or exited.Twenty respondents were selected. Ten hadleft their service provider recently afterexperiencing a service failure and 10 hadremained after recently experiencing a servicefailure and seriously considering leaving.We conducted face-to-face in-depth interviewswhich had been arranged over the telephone. Atotal of 435 small businesses were telephonedand if there was no reply or the contact was busy,up to three call-backs were made. A stratifiedrandom sample (by Auckland region) was usedand the companies were selected from a smallbusinessdatabase purchased from a third party.Approximately one in 16 companies contactedhad encountered at least one service failure largeenough to make them seriously consider leaving oractually leave their bank. Out of those companies,one in four refused to participate in the research.The results from these interviews weretranscribed and then the content analysed usinga number of methods, including checklistmatrices, “eyeballing” and searches in NUDIST(a software program).RESULTSTHE “REMAIN” GROUPThe majority of the respondents in the“remain” group were emotional about theservice failure they had experienced and this ledeight out of 10 to complain to the bank. Therewas a general feeling of dissatisfaction with thebanks’ service recovery efforts.Of the eight respondents who gave their bankan opportunity to conduct service recovery,only three were satisfied with the recovery. Eventhen, they were still not entirely satisfied, butacknowledged that the bank had made aconcerted effort to rectify the situation.More than half of the respondents in the“remain” group felt their relationship with theirbank was not close. The four respondents whohad relatively close relationships tended to havebeen with that bank for more than 10 years andthey expressed moderate to strong levels of loyalty.The lack of a close relationship with the bankwill have contributed to the low sense ofloyalty half of the respondents in the “remain”group felt towards their bank. This suggeststhat these respondents remained for reasonsother than loyalty.Seven out of 10 respondents in this groupperceived high barriers to exit and an46

W HY CUSTOMERS LEAVE – OR DECIDE NOT TO LEAVE – THEIR BANKadditional two perceived moderatebarriers. The barriers mentioned included:the amount of time and hassle it wouldtake to change; having to get new chequebooks and transfer funds; and having tolearn a new cash management system.The amount of paperwork involved, issueswith existing loans and guarantees and a goodhistory and rapport with their current bankseemed also to be factors. This was accentuatedby the overall feeling by most of theserespondents that the alternative banks were ofthe same (or a similar) quality as their own.THE “EXIT” GROUPThe majority of the respondents in the “exit”group complained to their bank about thefailure. Only three out of the eight whocomplained were satisfied with the recoveryefforts of their bank.There appeared to be correlation between thelength of time the relationship had existed and thecloseness of the relationship. The length ofrelationships between these respondents and theirbanks varied from four to 52 years.Seven of the respondents in the “exit” groupperceived the barriers to exit to be low, witheight of the respondents indicating thatalternative banks were of a marginally orclearly better quality.Finally, the general trend experienced bythis group was to encounter a progression ofproblems over time rather than a singlecritical incident.DISCUSSIONThe literature implies that a direct relationshipexists between service recovery and howcustomers react. If a service recovery is satisfactory,a customer will remain, but if it is unsatisfactory, acustomer will exit. This is not always the case,however, as our results show.Our interviews revealed that three majorfactors influence a customer’s decision toremain or exit after a service failure. These are:Three major factors influencea customer’s decision toremain or exit after aservice failuresatisfaction with recovery (if the customercomplained), loyalty towards the provider, andperceptions of barriers to exit. Each of these isdiscussed below.SATISFACTION WITH RECOVERYA number of customers in the “exit” group leftdespite satisfaction with service recovery and anumber of customers in the “remain” groupstayed despite dissatisfaction. This indicatesthat service recovery does not always make adifference.Our results show that if a customer isdissatisfied with recovery they will exit, exceptwhen other factors such as loyalty and barriersto exit are high, enticing the dissatisfiedcustomer to stay.Great service recovery can limit the harmfulimpact of service failure, as Boshoff (1997)suggests. Still, factors other than satisfactoryrecovery seem to be more important in drivinga customer’s decision to remain or to exit.Similarly, Stewart (1998) suggests that bythe time a customer has decided to exit anorganisation it is too late to recover them.Even if recovery efforts are satisfactory, thecustomer may exit because they have alreadymade their decision.Our results support the concept that acustomer may receive satisfactory recovery butstill exit. Three of the respondents in the “exit”group exited despite satisfactory recovery,implying that they had already decided to exitbefore the recovery was provided.▼UNIVERSITY OF AUCKLANDBusiness ReviewVolume 2 Number 2 200047

Some stay because of loyaltyand some because ofperceived barriers to exitLOYALTYWe found that customers who felt a strongsense of loyalty to their bank do not alwaysremain after service failure – although themajority do. When service recovery fails, thedegree of loyalty seemed to moderate thecustomer’s decision to exit or stay.This loyalty may result from the customer’sreliance on the relationship they have formedwith the service provider (Gwinner, Gremlerand Bitner, 1998); it may reflect how long therelationship has existed; or it may be that thecustomer feels their provider is giving themthe “best deal”.The results also showed, however, that somepreviously loyal customers still left. Theservice failure is often strong enough to forcethese customers to leave, despite the strengthof the relationship.Alternatively, the loyal customers who leftsometimes felt that the bank was notreciprocating their sense of loyalty. They leftbecause they had a feeling of almost betrayal,especially after many years of relationship withthat bank.PERCEIVED BARRIERS TO EXITThe results from this study support the theoryin the literature that if a dissatisfied customer isnot loyal and has a perception that the barriersto exit are low, they will exit rather thanremain. All customers who remained eitherperceived that there were at least some barriersto exit or felt at least a moderate sense ofloyalty towards their bank.A further dimension of barriers to exit is acustomer’s perception of alternative providers.We found that a far greater number ofcustomers in the “exit” group than in the“remain” group perceived alternative banks tobe of a better quality than their current bank.Customers who remained tended to perceive allbanks as the same or other banks as being onlyslightly better.In terms of the propositions, evidencesuggests that all are supported, albeit weaklygiven the nature of the data. Interestinglythough, new categories seem to exist that werenot fully accounted for in the propositions.Some customers stay after poor servicerecovery because of perceived barriers to exitwhile some stay because of loyalty. Similarly, ofthe customers who do not complain after aservice failure, some stay because of loyalty andsome because of perceived barriers to exit.Finally, some customers may exit, even aftergood service recovery, because of low levels ofloyalty (i.e. variety-seeking behaviour) orbecause of a perception of low barriers to exit.All of these categories can be built into amodel of customer reactions to service failure,which appears in Figure 2.Most of the research into customer retentionand customer exit investigates the processesseparately, not linking the two processes. Bybuilding on the models of Stewart (1998) andBoshoff (1997) and by incorporating therelevant literature and the summary of theresults above, this model shows the differentways customers react to service failure withtheir service provider.EXPLANATION OF THE MODELThere are four concepts in the model thatcover exiting (Boxes 1-4). In the first ofthese, a customer exits without complainingbecause they are dissatisfied. This stems fromthe research that says customers can have twoactive negative reactions to dissatisfaction:voice and exit. In this instance, customers havechosen not to voice and exited instead.The next three exit concepts encompass theidea that if a customer is dissatisfied with aprovider they usually complain in the hope thattheir problem will be resolved (i.e. they hope to48

W HY CUSTOMERS LEAVE – ORDECIDE NOT TO LEAVE – THEIR BANKreceive satisfactory service recovery efforts).Even if a customer is satisfied with recoveryefforts, they may still exit. Sometimes, this isbecause they perceive the barriers to exit to below. A change to an alternative provider is notperceived to be difficult.Other times, the customer may feel littleloyalty to their provider. “A satisfiedcustomer may be willing (or even eager) topatronise alternative suppliers, hoping toreceive even more satisfying results” (Rustand Zahorik, 1993). This implies that thefactors that make a customer loyal aredifferent from those that make them satisfied.The final exit concept refers to the scenariowhen a customer leaves after being dissatisfiedwith the service recovery from their provider. Thismakes sense given that it is the second time theservice provider has failed to meet the customer’sexpectations.The other reaction a customer may have to aservice failure, after seriously consideringleaving, is to remain. The next five concepts(Boxes 5-9) capture the different iterations ofremaining.If a customer complains, they may remainbecause they are satisfied with the recoverythey have received. The loyalty the customerperceives in the ninth concept is contributingto their decision to remain.Concepts five through to eight are“A satisfied customermay be willing (or eveneager) to patronisealternative suppliers”characterised by a customer remaining, despitedissatisfaction. This could be dissatisfactionwith the service recovery efforts of theirprovider after a complaint was made (Boxes 5and 7), or dissatisfaction with the provider thathas not been remedied because the customer hasnot complained (Boxes 6 and 8).Concepts five and six have spurious loyaltyaspects to them because of the perception that thebarriers to exit are high (Dick and Basu, 1994).On the other hand, concepts seven and eight havegenuine loyalty aspects to them. Customersremain, despite dissatisfaction, because they areloyal and the incident they encountered was notharmful enough to make them exit.IMPLICATIONSThe aim of this paper was to investigate servicefailure more fully than previously, and toprovide a more complete picture of its outcomes– which is represented in Figure 2.▼FIGURE 2ProblemExitRetention1–Nocomplaint2– Complaint– Goodrecovery– Low inertia3– Complaint– Goodrecovery– Low loyalty4– Complaint– Poorrecovery5– Complaint– Poorrecovery– High inertia6–NoComplaint– High inertia7– Complaint– Poorrecovery– Loyalty8–NoComplaint– Loyalty9– Complaint– Goodrecovery– RelativelyincreasedsatisfactionUNIVERSITY OF AUCKLANDBusiness ReviewVolume 2 Number 2 200049

The discovery that moreresearch is required withinthis area is an importantimplicationThe purpose is to draw the attention of managersand academics alike to the importance ofunderstanding fully what drives certain servicefailure outcomes.A customer’s perceptions of barriers to exit,perceived value of the relationship, perception ofservice recovery, and feelings of loyalty to theprovider will all affect the decision to exit or stay.Managers, therefore, need to understand therole barriers to exit play in their industry, theextent to which loyalty can be built to isolatethe effects of service failures and the power ofservice recovery.This suggests that even before customersencounter a service failure, providers shoulddevelop (and endeavour to sustain) closerelationships with their customers. Relationshipstrength and loyalty are driving factorswhen customers are deciding whether toremain or exit.This is not to suggest that service recoveryis not important. Far from it. Given thatMark ColgateSENIOR LECTURERUniversity of Auckland Business SchoolEmail: m.colgate@auckland.ac.nzMelissa NorrisPOSTGRADUATE STUDENTUniversity of Auckland Business Schoolmarketers need to increasingly justify theirinvestments in service quality improvements,however, it is essential for managers to realiseexactly how powerful service recovery can beand that there are other ways of insulating acompany from service failures.For academics, the results indicate thatinvestigating the “mix” of factors that drivethe outcomes of service failure is important.This research suggests that perceived barriersto exit, loyalty and service recovery all play arole in service failure outcome decisions. Thediscovery that more research is required withinthis area (particularly on spurious loyalty) isan important implication of this paper.It is important to note the limitations of thisstudy. One is the size of the interview sample.With only 10 respondents in each of the twogroups, the findings from their interviews arenot very generalisable. The aim of theresearch was not to generate generalisableresults, however, but rather to provideevidence that service failure is a multi-facetedconstruct that is inherently complex.A second limitation is that the results werecollected from only one industry – thebanking industry. This may be particularlyimportant given that the banking industry ischaracterised by high levels of customerinertia. The results, therefore, may be uniqueto this industry.CONCLUSIONResearch into services is growing rapidly andstill-unexplained phenomena need to beinvestigated. A review of previous research,both theoretical and empirical, demonstrated aneed for further knowledge of how customersreact to service failures.We identified three major factors that affect acustomer’s decision to remain or exit. They are:50

W HY CUSTOMERS LEAVE – ORDECIDE NOT TO LEAVE – THEIR BANKthe customer’s satisfaction with recoveryefforts; their sense of loyalty to theprovider; and their perception of thebarriers to exit. Service recovery may not be asimportant in driving exit and retention as wasinitially thought.By using these factors, we have proposed amodel that encapsulates all the major reasonsa customer may stay or leave a serviceorganisation after a service failure. The aimof the model is to develop a more in-depthexplanation of how and why customersbehave as they do, after experiencing aservice failure.With this knowledge, service providersshould ultimately be able to deliver a moreeffective service.FURTHER READINGFor further reading on service recovery, Hart et al(1990) is a worthwhile overview for managers, whileTax et al (1998) provides a more academic, but veryinsightful, analysis of service recovery. Keaveney’s(1995) article is excellent and concerns the reasonswhy customers switch from service organisations.Service recovery may not beas important in driving exitand retention as wasinitially thoughtREFERENCESBoshoff, C. (1997). An experimental study of service recoveryoptions, International Journal of Service Industry Management,8(2), 110-130.Day, R. L., and Landon, E.L. Jr (1977). Toward a theory ofconsumer complaining behaviour, in Woodside, A.G., Sheth, J.N.,and Bennett, P.D. (Eds) Consumer and Industrial BuyingBehaviour, North-Holland, New York, pp 425-37.Dick, A. S., and Basu, K. (1994). Customer loyalty: toward anintegrated conceptual framework, Journal of the Academy ofMarketing Science, 22(2), 99-113.Gronroos, C. (1990). Relationship marketing approach to themarketing function in service contexts: the marketing andorganisational behaviour influence, Journal of Business Research,20(1), 3-12.Gwinner, K.P., Gremler, D.D., and Bitner, M.J. (1998). Relationalbenefits in services industries: the customer’s perspective, Journalof the Academy of Marketing Science, Spring, pp 101-114.Hart, C., Heskett, J., and Sasser, E., (1990). The profitable art ofservice recovery, in Christopher et al. Turning Sow’s Ears into SilkPurses, Dryden, New YorkHirschman, A.O. (1970). Exit, voice and loyalty: responses todecline in firms, organisations, and states, Harvard UniversityPress, Cambridge, USA.Keaveney, S.M. (1995). Customer switching behaviour in serviceindustries: an exploratory study, Journal of Marketing, 59(2),71-82.Levesque, T.J., McDougall, G.H.G. (1993). Managing customersatisfaction: the nature of service problems and customer exit,voice and loyalty, Asia Pacific Journal of Quality Management,2(2), 40-58.Reichheld, F.F., and Sasser, W.E. Jr (1990). Zero defections:quality comes to services, Harvard Business Review, September-October, pp 105-111.Rust, R. T., Zahorik, A. J. (1993). Customer satisfaction,customer retention, and market share, Journal of Retailing,69(2), 193-215.Singh, J. (1990). Voice, exit, and negative word-of-mouthbehaviours: an investigation across three service categories,Journal of the Academy of Marketing Sciences, 18, 1-15.Stewart, K. (1998). An exploration of customer exit in retailbanking, International Journal of Bank Marketing, 16(1), 6-14.Tax, S. S., Brown, S. W., and Chandrashekaran, M. (1998).Customer evaluations of service complaint experiences:implications for relationship marketing, Journal of Marketing,62 (April), 60-76.UNIVERSITY OF AUCKLANDBusiness ReviewVolume 2 Number 2 200051