The Cost of Service Recovery in the Financial Service Sector

ABSTRACT - This research focuses on an investigation into the cost of service recovery in the financial services sector. The research empirically tested whether the strength of the buyer-seller relationship before a negative service experience is made serves to affect the cost of compensation offered in an attempt to return the customer to a satisfied state.

Citation:

Mark Michael and Tony Ward (2001) ,"The Cost of Service Recovery in the Financial Service Sector", in AP - Asia Pacific Advances in Consumer Research Volume 4, eds. Paula M. Tidwell and Thomas E. Muller, Provo, UT : Association for Consumer Research, Pages: 82-90.

Asia Pacific Advances in Consumer Research Volume 4, 2001      Pages 82-90

THE COST OF SERVICE RECOVERY IN THE FINANCIAL SERVICE SECTOR

Mark Michael, Central Queensland University, Australia

Tony Ward, Central Queensland University, Australia

ABSTRACT -

This research focuses on an investigation into the cost of service recovery in the financial services sector. The research empirically tested whether the strength of the buyer-seller relationship before a negative service experience is made serves to affect the cost of compensation offered in an attempt to return the customer to a satisfied state.

A sample of 71 respondents were surveyed using a personal interview. The purpose of the survey was o determine the respondent’s perception of seriousness of the five hypothesised negative service experiences (errors by their bank) and the compensation they felt matched the severity of the negative experiences.

The result of the data analysis showed that customers, in general, do expect that compensation should be related to the seriousness of a banking error, and that a positive relationship perceived by the customer with the bank does positively affect customer recovery. It was found that female customers exhibited a stronger correlation between perceived seriousness of banking error and appropriate compensation than males. There was significant variability between respondent perceptions of the less serious errors, but less variability demonstrated for the more serious errors.

The research addressed a significant gap in the service recovery literature by making contributions to both theory and practice, and identifies a number of areas for future research.

BACKGROUND

The importance of service recovery has long been recognised, but few empirical studies have focused on service recovery and customer retention (Kelly & Davis 1994). No literature could be found which tested for a possible correlation between the seriousness of a negative service experience made by a financial institution, and the dollar value of effort required to restore the relationship to the same state as before the negative experience was caused by the financial institution. In addition, the strength of the buyer-seller relationship before a negative experience is caused, is used as a moderating variable in order to determine what effect, if any, relationship strength has on the dollar value of effort required to restore or re-establish the original relationship. The financial services sector (banking) was chosen for the purpose of this research as it is recognised as a service industry where the level of service provided by each institution is the primary means of product differentiation (Berry & Parasuraman 1991).

As today’s customers are becoming more demanding, better informed, and more assertive than they once were (Hoffman, Kelley & Rotalsky 1995), there is a greater need for research which focuses on customer recovery and retention programs to remedy the effects of a negative service experience. Study in this area is also required as Hart, Heskett and Sasser (1990, p. 148) stated "over one half of business responses to customer complaints actually strengthen customer’s negative feelings toward the business and its representatives", which suggests that customer’s negative service experiences are being dealt with inadequately. "When a defect or failure does occur, it is imperative for retailers to rectify mistakes through the utilisation of effective recovery strategies" (Hoffman, Kelly & Davis 1993, p. 429).

This research examines and empirically tests the correlation between the customer’s perceived seriousness of a negative service experience (an error) caused by a bank and the dollar value of effort required by the bank to re-establish the original relationship.

A measure was also gained of the personal variance between respondents, with respect to how serious they rated different negative service experiences that a bank can cause with regards to service quality. In addition, customer’s relative rating of the seriousness of the banking error was evaluated to determine the style of error which customers found most serious.

Definitions

Bank: is any Bank or non Bank entity which is committed to providing a range of personal and commercial financial products and services.

Relationship: The relationship perceived by the customer towards the bank (adapted from Preece & ard 1999)

Relationship Strength: Identifies how strong the relationship between bank and the customer, as perceived by the customer (adapted from Ward & Smith 1998)

Recovery: There are currently a number of varying perceptions held by researchers as to how service recovery is defined. Bell and Ridge (1992, p. 58) suggested that "service recovery includes all the actions the people in your training unit take to move a customer from a state of disappointment to a state of satisfaction". Using their analogy, "service recovery is like a hospital or doctor that gets a sick patient back to normal" (Bell & Ridge 1992, p. 58). Thus, service recovery attempts to get a customer back to having a sound relationship with the service provider as that before the problem occurred. A similar view regarding service recovery is saying that it involves resolving problems to the customer’s satisfaction (Spreng, Harrell & Mackoy 1995). Parasuraman and Berry (1991, p. 34) stated that service recovery involved "doing the service very right the second time". Zemke (1994, p. 17) suggested that service recovery was about "solving customer’s problems" and further commented by saying "do it right the first time. If you don’t, be sure you do it right the second time".

For this research, recovery is defined as that part of quality management designed to alter the negative perceptions of dissatisfied consumers following a negative experience caused by the financial institution and to ultimately maintain a business relationship with these consumers (adapted from Schweikhart, Strasser & Kennedy 1993, p. 3).

Recovery Effort: is the degree of special attention and/or compensation exerted by the bank to correct the situation, preferably before service delivery is completed (adapted from Lovelock 1996).

RESEARCH PROBLEM

As more and more service providers are adopting the doctrines of relationship marketing theory, service recovery methods have been put under the microscope in order to make recovery techniques more refined and efficient. In particular, banks go to great lengths to build strong relationships with their customers, however, inevitably negative service experiences do occur, for example, cheque books are lost, statements are sent to the wrong address and customer instructions are misunderstood. Therefore, it is essential to have a service recovery process ready in order to return a dissatisfied customer to a satisfied state, and subsequently retain their custom. This research ascertained the dollar amount of effort required by the bank to recover the relationship with the customer relative to the seriousness of the negative service experience. Further, the strength of the customer’s relationship with the financial institution before the negative service experience is caused, is considered in this study.

Justification of the research

The importance of the successful development and implementation of service recovery programs has been highlighted by a number of service marketing researchers, including: Hoffman, Kelley and Rotalsky (1995); Spreng, Harrell and Mackoy (1995); Zemke (1994); Schweikhart, Strasser and Kennedy (1993); and, Bell and Ridge (1992). However, a wide ranging review of the literature revealed no extensive empirical studies or surveys which have explored this area of services marketing. This view is confirmed by Spreng, Harrell and Mackoy (1995, p. 15) who stated "given the acknowledged importance of service recovery, it is surprising that so few large-scale field studies have focused on this topic."

Due to the lack of empirical research in the areas of service recovery and customer retention, this study addresses one aspect of this identified gap in the current marketing literature, thereby making a cntribution to the field of marketing theory and practice. The financial services sector was selected because of it’s importance, and due to the changing nature of banking following widespread deregulation in the last two decades.

Boundaries of the research

This research focused on service recovery and customer retention in the banking sector. Although it is anticipated that some generalisability to other financial service sectors could be made, the changing relationship between banks and their customers with the introduction of more remote banking practices (such as, Automatic Telling Machines, the Internet and telephone banking) may affect the ability of these results to be applied to other financial sectors.

Contribution to knowledge

As discussed above, there is a lack of empirical research regarding the areas of service recovery and customer retention (Kelly & Davis 1994). As there has been little empirical research in the areas of service recovery, especially regarding the variables which affect the recovery effort required, a gap was identified in the current services marketing literature. The purpose of this study was to address this gap and make a valuable contribution to the field of marketing theory.

SERVICE RECOVERY

The ultimate goal of service recovery is to motivate customers to continue to engage in business transactions with the firm which caused the initial problem (Schweikhart, Strasser & Kennedy 1993). However, Hart, Heskett & Sasser (1990, p. 148) noted that "studies show that more than half of all efforts to respond to customer complaints actually reinforce negative reactions to a service". To ensure that recovery efforts are successful, Hart, Heskett & Sasser (1990, p. 148) suggested:

The surest way to recover from service mishaps is for workers on the front line to identify and solve the customer’s problem, but doing so requires decision making and rule breakingBexactly what employees have been conditioned against. A customer’s problem is an opportunity for the company to prove its commitment to service, even if the customer is not to blame. Training can give employees the perspective that service recovery requires, but the company must empower them to act; the company must think of the value of pleasing a customerBnot the cost.

Service recovery involves a series of steps that are taken by a service provider to restore a customer to a state of satisfaction after a negative incident has occurred. Bell and Ridge (1992, p. 61) suggested that "just as customers have expectations for service, customers also have notions about what should happen after a service breakdown has left them disappointed". However, researchers opinions differ as to what steps are taken and when they are taken (for example, Reichheld & Kenny 1990). Zemke (1994, p. 17) identified five things customers expected after being dissatisfied, which include:

1. To receive an apology

2. To be offered a "fair fix" for the problem

3. To be treated in a way that suggests the company cares about the problem, about fixing the problem, and about the customer’s inconvenience

4. To be offered some value-added atonement for their inconvenience

5. Keep your promises.

Taking a different perspective, Schweikhart, Strasser and Kennedy (1993, p. 5) stated that recovery initiatives can be categorised along two dimensions, that is, psychological and tangible:

Apologising for poor service and explaining why the error occurred are examples of psychological recovery responses. Tangible methods focus on compensation for damages a customer perceives were incurred, e.g., reducing a bill to compensate for inconvenience.

The benefit of successful recovery efforts is two fold. "Good recovery can create more good will than if things had gone right in the first place" (Hart, Heskett & Sasser 1991, p. 72), and also the customer is retained. This is a most important perspective in customer service and for service recovery and retention, as it demonstrates that customers acknowledge that mistakes will occur, what is important is correcting that mistake promptly.

No literature was found which directly linked the value of compensation with dollars, although a number of authors have alluded to the value of compensation (for example: Hart, Heskett & Sasser 1991; Zemke 1994; and Bell & Ridge 1992).

In summary, service recovery should be an integral part of any service provider’s marketing strategy. It is only through an effective service recovery program that organisations can capitalise on the benefits of customer loyalty.

THEORETICAL FRAMEWORK

The theoretical model for this research is presented in figure 1. As shown in figure 1, the primary focus of the research is to determine the cost of service recovery and to seek a correlation between the seriousness of an error and that cost. There are a number of other foci:

- to determine whether there is a moderating effect of the strength of the relationship between the service provider and the customer (as perceived by the customer) before a negative service experience is encountered, on the customer’s perceived seriousness of the negative service experience and the subsequent cost to the service provider to re-establish the original relationship;

- the effect of gender;

- to investigate the variability of errors made by the service provider as perceived by customers;

- to identify the degree of variability between customers; and,

- to look for any pattern in the type(s) of error which are perceived as most serious.

Three hypotheses and two research questions were developed to address the research problem.

FIGURE 1

THEORETICAL FRAMEWORK

Hypothesis 1 (H1). This study addressed the primary gap in the literature by firstly determining the cost of a recovery effort relative to the seriousness of a negative experience made by a bank. Zemke (1994, p. 17) identified a degree of compensation as part of a 'fair fix’ for an error (among other things), though the literature offers many disparate solutions, some involving compensation and some not (for example, Heskett, Sasser & Earl 1990; Reichheld & Kenny 1990; and Heskett, Sasser & Earl 1991). For this research it was assumed that compensation would be offered as this is common practice in modern banking, so the aim of H1 was to determine whether the stronger the customer perceives the seriousness of a negative experience, the greater the dollar value of effort that is required to recover the relationship. Thus,

H1: The dollar value of effort required to re-establish the buyer-seller relationship to its original state is positively correlated with the customer’s perceived seriousness of a negative experience caused by a financial institution.

Hypothesis 2 (H2). The strength of the buyer-seller relationship before a negative experience is made is then employed as a moderating variable to test how, if at all, it affects the cost of a recovery effort. A basic premise of relationship marketing is that a strong relationship perceived by the customer will tend to prevent the customer from switching to another supplier. In addition, a strong relationship may have the effect of the customer perceiving a smaller compensation as adequate in comparison to the situation where a weak relationship exists, or where the customer perceives no relationship to exist. Thus, H2 aims to determine whether the strength of the buyer-seller relationship before a negative experience is caused affects the dollar value of effort that is required to recover the relationship. This approach also investigated the effect of the customer having, or not having, a perceived relationship with the bank, irrespective of the strength of that relationship.

H2: A strong buyer-seller relationship reduces the positive effect of the customer’s perceived seriousness of a negative experience on the dollar value of effort required to re-establish the buyer-seller relationship to its original state.

Hypothesis 3 (H3). The effect of gender was tested to see if this had any comparative effect. No literature was found which indicated whether males or females should be expected to seek greater compensation.

H3: The dollar value of effort required to re-establish the buyer-seller relationship to its original state is greater for male than female customers.

Research Question 1 (RQ1). Humans vary considerably in their perceptions of events, and in particular adverse events. Thus, a measure of customers personal variations of the perceived seriousness of each negative experience on the part of the financial institution would assist in identifying the potential range of customer reactions to errors.

RQ1: What is the extent of the individual variation between customers with respect to their perception of the seriousness of typical negative service experiences?

Research Question 2 (RQ2). Developing the theme of the different perceptions of customers to errors, customer’s relative rating of the seriousness of the banking error was evaluated to determine the type of error which customers found most serious. Further, the relationship between the relative seriousness of the error and the potential dollar loss to the customer was investigated.

RQ2: What type(s) of error are perceived by customers to be the most serious?

METHOD

For the purposes of this study, the research design selected is primarily exploratory. Little relevant literature was found to guide the selection of 'typical’ banking errors or the dollar values which are 'typically’ used to compensate customers for errors of differing seriousness. Thus, these parameters were determined following discussions with bank employees, a process assisted by one of he researchers being employed by a bank. The research design involved the use of a test instrument administered through face-to-face interviews to obtain the primary data.

The test instrument consisted of two parts:

- Section 1 asked initial questions about the respondent, such as gender, type of banking primarily used, whether or not the respondent perceived they had a relationship with the bank or not, and if so, how strong they perceived this relationship with the bank to be;

- Section 2 introduced a series of five scenarios which the respondents were asked to evaluate. Respondents rated the seriousness of typical negative service experiences if they had been caused by their bank, and then which compensation should apply from the selection. These five scenarios are:

Error 1Bplacing an order for a deposit book, only to find that the order was not input;

Error 2Bdepositing $200 which was recorded by the bank as being a deposit of $20;

Error 3Bopening a new account and being told that no fees were associated with the account, only to later find out that fees are associated with the account;

Error 4Bafter inquiring about interest rates on a term deposit, the customer decided to open an account, to find out later that the rate was lower than they were told;

Error 5Breceiving rude service, for example, foul language by one of the employees.

For each scenario the respondents were asked to rate their perception of the seriousness of the error on a bi-polar semantic scale of 1 to 7. They were then asked to identify what level of compensation they considered appropriate to restore their relationship with the bank. The compensations used were taken from those typically used by banks in Australia, as follows (with their dollar equivalents which were not given to respondents as this information would not be given to a customer in a real banking situation):

On the spot apology     $2

A phone call from the manager     $5

A follow up letter     $10

Waived fees for a month     $15

Movie tickets     $21

A pen     $30

Two free meals at a local restaurant     $80

Weekend at a resort     $200

The cost for the first three options was equated on the basis of a $60 per hour labour charge. Thus, each of the first three options allowed 2, 5 and 10 minutes, respectively, to be completed. The cost for the waived fees was based on an average monthly cost to a customer. The pen offered as compensation is the equivalent of a dress pen. The cost of the meal and weekend at a resort was determined on the basis of inquiries made.

The respondent sample selected consisted of members of the public in covered shopping malls. The sample population consisted of 71 randomly surveyed customers. The names of respondents surveyed were not requested so all respondents were anonymous.

RESULTS AND ANALYSIS

A total of 95 people were asked to respond to the survey to obtain the 71 completed instruments, giving a response rate of 75%. Twenty six of the respondents were male (37%), while 62 (87%) answered as personal rather than business users of the bank.

Respondents were asked how long they had been with their current bank, 24 respondents (34%) had been with their bank for seven or more years, while thirty one respondents (44%) did not perceive that they had a relationship with their bank. The strength of relationship which the remaining 40 respondents (56%) who did perceive they had a relationship with the bank is shown in figure 2.

As shown in figure 2, of the forty respondents (56%) who did perceive they had a relationship with their bank, 37 perceived they had a strength of relationship of 4 or above, on a scale of 1 to 7. The perceived strength of relationship did not correlate with years with the current bank.

H1BRe-establishing the buyer-seller relationship

For testing H1 the seriousness of the i~pothesised error made by the bank was the independent variable, and the compensation offered by the bank was the dependent variable. This compensation was established by using the dollar value of the compensation which each respondent considered appropriate for each error. The results of the ANOVA test for the five errors is shown in table 1.

As shown in table 1, in each of the five cases the seriousness of the error and the corresponding compensation which respondents considered appropriate were highly significant. The R² value showed that the relationship accounted for between 20 and 23% of the variability between the seriousness of the error and the compensation considered appropriate.

Thus the results support H1, and show that the seriousness of an error made by the bank and the corresponding compensation which respondents considered appropriate were hm{hly correlated, and overall this relationship accounted for over 20% of variability.

H2BEffect of strength of buyer-seller relationship

For testing H2, the strength of the relationship between customer and bank (as perceived by the customer) as shown in figure 2 above, was introduced into the analysis of H1 as a moderating variable. The results of the ANCOVA test are shown in table 2.

As shown in table 2, three of the five error cases (cases 2, 4 and 5) were significant with R² values ranging from 0.21 to 0.45 for the forty respondents who perceived they had a relationship with their current bank. The test was then performed with the existence of a relationship or not used as the moderating variable, rather than the strength of relationship perceived, as shown in table 3.

As shown in table 3, the existence or otherwise of a relationship was more highly significant than the strength of a relationship where it existed, indicating that any relationship as perceived by the customer would reduce the amount of compensation required to recover the situation. The R² values where a relationship was perceived to exist ranged between 0.23 and 0.34.

FIGURE 2

STRENGTH OF RELATIONSHIP WITH CURRENT BANK

TABLE 1

ERROR VS COST-ALL RESPONDENTS

TABLE 2

EFFECT OF STRENGTH OF BUYER-SELLER RELATIONSHIP

TABLE 3

EFFECT OF EXISTENCE OF BUYER-SELLER RELATIONSHIP

TABLE 4

ERROR VS COST-MALE VS FEMALE RESPONDENTS

H3BEffect of gender

A significant difference was found between male and female respondents, as shown in table 4.

As shown in table 4, the relationship between the seriousness of the error and the corresponding compensation which respondents considered appropriate was much stronger for female than for male respondents. All of the female cases were highly significant with R² ranging from 0.15 to 0.35, while two of the male cases (1 and 5) were not significant, and the R² values for the three significant cases were lower overall.

Thus, females showed a greater correlation than males, with the female R² values averaging 0.27.

RQ1BEffect of individual variations

RQ1 sought to identify the extent and nature of individual variations in the perceptions of bank customers to the five hypothetical errors. Table 5 shows the variability of these perceptions to the five cases by tabulating the number of respondents who perceived each degree of seriousness for each case, as measured on a seven point scale.

As shown in table 5, cases (errors) 2 and 5 were perceived as most serious having means of 6.51 and 6.17 respectively. Error 2 was the deposit of $200 being recorded as only $20, and error 5 was receiving rude service. The variability as shown by the standard deviations (0.77 and 0.83 respectively) was lower than for the 3 cases which were not perceived as so serious (1.31, 0.98 and 0.84).

These results indicate that in answer to RQ1, the more serious the error as perceived by customers the less the variability between customers, thus customers perceived the more serious errors as more uniformly serious. It is also evident that for the less seriously perceived errors there is more variability in the degree of seriousness perceived by the customers compared with the two more seriously perceived errors.

RQ2BType of error and dollar loss

RQ2 sought to identify the respondents perceptions of the nature of errors, and whether these perceptions were related to the potential dollar loss to the customer. Table 6, takes the seriousness means from table 5 for each error and places them with the error.

As shown in table 6, rude service and a deposit incorrectly recorded were perceived as the two most serious errors by respondents, the degree of seriousness being 6.17 and 6.51 (out of a maximum of 7) respectively. The potential dollar loss due to a deposit being incorrectly recorded was $180, while there is no potential dollar loss due to rude service. Error 3 (told no fees associated with new account which had fees) was judged marginally more serious than error 4 (interest rate was lower than told). Both of these two errors encompass a potential loss, but this loss was not quantified. The lack of input for a replacement cheque book is more of an inconvenience than a potential dollar loss, and scored the lowest mean of 3.85.

Thus, for RQ2, these results show the importance of errors where there is a potential and quantifiable dollar loss, and the serious effect of rude service on customers.

TABLE 5

INDIVIDUAL VARIATIONS

TABLE 6

INDIVIDUAL VARIATIONS

DISCUSSION

The results of this research clearly show that there is a relationship between the perceived seriousness of an error made by a bank and the dollar level of compensation that customers expect in order to recover the situation. It is also evident that where the customer perceives that they have a relationship with the bank the level of compensation expected is generally lower than when no relationship is perceived, although the strength of that relationship did not appear to have an effect. These factors may reflect a different perspective of the function of the bank by customers who perceive, and those who do not perceive, a relationship with the bank, but this aspect was not investigated in this research. In all significant cases the correlations were of moderate strength, varying quite widely between 0.15 and 0.45.

Female respondents gave more significant results than males, and the correlations for females were generally stronger, indicating that females had a higher level of correlation between perceived seriousness of error and value expectation for compensation than males.

There was also clear evidence of the variability of perceptions by respondents regarding the degree of seriousness of the different hypothesised errors, and there was no relationship evident that customers directly link seriousness of error with potential dollar loss. This factor was evidenced by the seriusness of rude service perceived by respondents, in comparison with, say, the actual lower interest on an account which could represent a potential high dollar loss to the customer.

Implications for theory

The results indicate that this is an area of service recovery research which requires much greater attention. The difference between male and female respondents is of great potential importance, in particular if it was found for a range of different products. The role of the relationship which customers perceive between themselves and their bank is also potentially very important, in particular with respect to the current emphasis that banks are placing on non-personal banking, such as using automatic teller machines, telephone and Internet banking. These non-personal banking activities may be more cost-effective but will reduce the number of times that a customer has to visit their bank each year, with a consequent reduction in the opportunities for direct human interaction and the consequent building and nurturing of relationships (for those customers who want and perceive such a relationship). In contrast, for those customers who do not perceive that they have a relationship with their bank, which other research indicates that they do not want such a relationship (Ward & Smith 1998), the move towards non-personal banking will not be of so much consequence. A revised theoretical model can be developed from the results of this research. This revised model is based on the theoretical framework developed at figure 1, with the strength of the correlations added and the research question variables deleted, as shown in figure 2.

As shown in figure 2, there is a highly significant direct relationship between the Seriousness of an error and the Dollar effort required as compensation (0.04<p>0.002), and the Seriousness of an error accounts for between 20% and 23% of the variance of the Dollar effort required as compensation. When the strength of a relationship as perceived by the customer is introduced as a moderating variable, the relationship was significant for three of the five errors (0.04<P>0.002), and the Seriousness of an error accounts for between 21% and 45% of the variance of the Dollar effort required as compensation. When the existence of a relationship with the bank (as perceived by the customer) was used as a moderating variable, the relationship remained highly significant (0.002<P>0.000) and the Seriousness of an error accounts for between 23% and 34% of the variance of the Dollar effort required as compensation. Thus, the existence of a relationship, as perceived by the customer, has a significant and positive effect as a moderating variable upon the primary relationship between the Seriousness of an error and the Dollar effort required as compensation. Finally, the model indicates that in all cases females demonstrated a stronger relationship between the Seriousness of an error and the Dollar effort required as compensation than males.

This research has thus made a contribution to marketing theory by taking the initial step in addressing the identified gap in the service marketing literature.

FIGURE 3

REVISED THEORETICAL MODEL

Implications for industry

The results of this research have several implications for the banking industry and potentially for other service providers. This research has identified a number of practical implications for financial institutions:

$Financial institutions should not assume that customers with whom they perceive they hold a strong relationship will be tolerant to negative experiences caused by that institution, but rather that the dollar value to compensate some customers will be reduced if a relationship is perceived by the customer to exist

$There is a need to develop a recovery program, which should consist of various forms of prescribed compensation based on the customer’s perceived severity of a negative experience;

$Employees need to be empowered to an extent which allows them to be able to offer the customer an appropriate form of compensation. Existing research, for example by Zemke (1994), has indicated the value of promptly dealing with complaints;

$Customers are particularly sensitive to negative experiences relating to inaccurately processed deposits and rude service.

The main difficulty for banks in the current environment is to find a balance between their apparent desire to force all customers towards a more impersonal banking style with reduced inter personal contact when relationships can be developed and nurtured, and the indications from this research which show the value of a customer perceived relationship, by only 56% of the respondents. This apparent conflict will require a new and more convincing approach by the banks than the methods which have served them well in the past.

FURTHER RESEARCH

In the course of developing this research a number of prospective areas for future research were identified:

$the scope of the research was limited to respondents in Rockhampton. Future studies involving a greater number of respondents in a broader range of geographic and demographic areas would improve the external validity of the results. Comparisons among these groups could also be performed in order to determine if differences exist between the varying groups;

$this study focused only on banking within the financial service sector, however, replication studies could also be performed in a variety of service industries and then compared to identify if any variances exist between industries;

$for the purpose of this research a personal interview technique was adopted as it was the most appropriate method of data collection in this instance. Future studies could benefit from the application of other techniques or a combination of techniques to improve the richness of the results;

$the test instrument employed for the purpose of this research was developed without the aid of any other surveys as a base, due to the absence of previous studies performed in this area. Future studies involving this body of services marketing literature would benefit from this research in developing a future test instrument, for example: the inclusion of an age scale; the use of focus groups to better understand customer perceptions of such experiences and the whole issue of compensation; making allowances for customers who could not be recovered no matter what form of compensation was offered; and inclusion of specific financial institutions in order to compare one institution against another;

$the analysis of the personal interviews revealed that 44% of respondents perceived that no relationship existed between themselves and their financial institution. Exploring the attitudes of respondents who do not perceive they have a relationship with their financial institution would be useful as a part of a focus on the body of literature relating to relationship marketing. This has the potential to benefit service providers both within and outside the auspices of the financial service sector.

In summary, this research has made the initial steps in the expansion of service recovery and customer retention theory. The suggestions for subsequent research detailed in this section provide a path for the further development of these areas of service markting theory.

CONCLUSION

This research has made a contribution to marketing theory by addressing the identified gap in the service marketing literature relating to service recovery. A positive correlation between the severity of a negative service experience and the cost to recover the customer to a satisfied state was identified. These findings highlight the need for financial service providers to formulate service recovery strategies in order to improve retention levels.

Analysis further indicated that the relationship strength before a negative service experience occurs has an effect on reducing the cost required to restore the customer to a satisfied state. As a consequence, further research is necessary to determine the role relationship strength plays in the recovery and retention process. Through a greater understanding of the role of relationship strength plays in this process, financial service providers can better develop and implement recovery and retention strategies.

REFERENCES

Bell, C.R. & Ridge, K 1992, 'Service Recovery for Trainers’, Training and Development, May, pp. 58-63.

Berry, L.L. & Parasuraman, A. 1991, Marketing Services: Competing Through Quality, Maxwell Macmillan, Canada Incorporated.

Hart, C.W.L., Heskett, J.L. & Sasser, W.E. 1990, 'The profitable art of service recovery’, Harvard Business Review, vol. 68, JulyBAugust, pp. 148-56.

Hart, C.W.L., Heskett, J.L. & Sasser, W.E. 1991, 'Surviving a Customer’s Rage’ Successful Meetings, April, pp. 68-79.

Hoffman, K.D., Kelley, S.W. & Davis M.A. 1993, 'A typology of retail failures and recoveries’, Journal of Retailing, Winter, vol. 69, no. 4, p. 429.

Hoffman, K.D., Kelley, S.W. & Rotalsky, H.M. 1995, 'Tracking service failures and employee recovery efforts’, Journal of Services Marketing, vol. 9, no. 2. pp. 49-61.

Kelly, S.W. & Davis, M.A. 1994, 'Antecedents to customer expectations for service recovery’, Journal of the Academy of Marketing Science, vol. 22, Winter, pp. 52-61.

Lovelock, C. H. 1996, Services Marketing, Prentice-Hall, New Jersey.

Reichheld, F.F. & Kenny, D.W. 1990 'The Hidden Advantages Of Customer Retention’, Journal of Retail Banking, vol. XII, no. 4, Winter, pp. 19-23.

Schweikhart S.B., Strasser, S. & Kennedy, M.R. 1993, 'Service Recovery in Health Services Organization’ Hospital & Health Services Administration, 38:1, Spring, pp. 3-21.

Spreng, R.A., Harrell, G.D. & Mackoy, R.D. 1995, 'Service Recovery: impact on satisfaction and intentions’, Journal of Services Marketing, vol. 9, no. 1, pp. 15-23.

Ward, A. & Smith, T. 1998, 'Relationship marketing: Strength of relationship time versus duration’, European Marketing Academy Conference Proceedings, Vol 1, Stockholm, Sweden, pp. 569-88.

Preece, M. & Ward, A. 1999, 'The effect on service quality and customer satisfaction of personality types of service providers in the hotel industry’, European Marketing Academy Conference Proceedings, Berlin, Germany.

Zemke, R. 1994 'Service Recovery’, Executive Excellence, Sept, pp. 17-18.

----------------------------------------