HOW DO SERVICE CLUES INFLUENCE CUSTOMER SATISFACTION?

(1)

How do service clues influence customer satisfaction? 1

HOW DO SERVICE CLUES INFLUENCE CUSTOMER

SATISFACTION?

(2)

HOW DO SERVICE CLUES INFLUENCE CUSTOMER

SATISFACTION?

Author: Pim Busker Student number: 1516981

Department of Marketing Faculty of Economics and Business University of Groningen Master thesis Supervisor: Prof. Dr. J.C. Hoekstra

(3)

MANAGEMENT SUMMARY

Customer satisfaction is a topic that has interested marketers over the years. Increasingly, the importance of measuring and controlling the drivers (or: antecedents) of customer satisfaction has been established. In general, according to the marketing literature, the degree of similarity between consumer expectations and perceived performance determines customer satisfaction. However, there may be a division created in consumer expectations and their perceptions. For instance: there are calculative perceptions of quality and there are emotional perceptions of quality.

In this thesis, the antecedents of customer satisfaction are the main theme. It is important for companies to know which buttons they have to press to obtain high customer satisfaction. This study embraces the ‘service clue’ concept of Berry, Wall and Carbone (2006) who argue that there are three categories of drivers which may be assumed to influence customer satisfaction: functional, mechanic and humanic service clues. In this study, we explicitly link these three categories of service clues to customer satisfaction. It is important to identify categories of antecedents of customer satisfaction, because this enables companies to manage them by clue management to obtain a better customer satisfaction. This is desired because of the positive outcomes of higher customer satisfaction: greater customer loyalty, positive word-of-mouth, lower costs of future transactions, decreased price elasticity, and minimized likelihood of customers defecting in case of faltering quality.

The central question that is answered is: How do service clues influence customer satisfaction? To answer this question, first of all a study of existing marketing literature is conducted; secondly, qualitative research is used to find industry-specific service clues. These are then used to find out which service clues have a significant effect on customer satisfaction.

The results indicate that two of the three categories of service clues, namely the functional and the mechanic service clues, have a positive effect on customer satisfaction. Therefore, when the score in one of these categories rises, the customer satisfaction also rises.

(4)

PREFACE

Analyzing the drivers of customer satisfaction is part of my daily work at Store Support. Not only do we research what these drivers are for the customers of our clients, we also measure how our clients score on these drivers using mystery shopping.

However, when I started this study I was not working at Store Support. At the time, I was employed by GlassConnect, a windshield-repair franchise in the Netherlands. As a young company in the market, the managers wanted to know what the specific drivers of customer satisfaction were in the car-repair industry, because they wanted to put the customer central in their plans. The general idea was that 'if we know what is important to our customers, we also know what we have to invest in'. The results of the study provided very useful insights for the company. They were also used to show insurance companies what their customers would find important, and therefore, what would be taken into account when insurance companies choose a service provider. I would like to thank former owner of GlassConnect, Johan van 't Land, for the opportunity to conduct this research and the support I received during it. I would also like to thank Wouter Dam who built the online survey. Moreover, I would like to thank Janny Hoekstra and Hans Berger for being my supervisors. It must have been very hard to work with a student like me, who always has a lot of "other things" on his mind. I also would like to thank Jelmer Weijschedé and my colleague Marnix van Loenen ( famous for his word jokes) for helping me with my SPSS analyses.

The process of writing this thesis was not always easy, because many things came on to my path. I started my own company (Marketing Noord); founded a water polo club (de Groningsche Polo Club); had an eye injury for almost half a year; switched jobs two times; bought a house; but most importantly, my girlfriend and I became the parents of the love of our lives: our daughter Fien! Of course I would like to thank my parents for always supporting me and helping me in any way they can. But I am most grateful to my girlfriend Simone for supporting and motivating me during stressful periods. It is nice to know that there is someone to rely on.

I hope you will find the thesis useful in providing you with valuable insights on what causes customers to be satisfied.

Pim Busker

(5)

How do service clues influence customer satisfaction? 5 TABLE OF CONTENTS MANAGEMENT SUMMARY ... 3 PREFACE ... ... 4 CHAPTER 1: INTRODUCTION ... ... 6 1.1 Initial motivation ... ... 6 1.2 Problem specification ... ... 7 1.3 Research method ... ... 8

1.4 Academic and managerial relevance ... 8

1.5 Structure of the thesis ... 8

CHAPTER 2: RESEARCH FRAMEWORK ... ... 9

2.1 Customer satisfaction ... ... 9

2.1.1 Definition of customer satisfaction ... ... 9

2.1.2 Antecedents of customer satisfaction... ... 10

2.1.3 Service clues ... ... 11

2.2 Conceptual framework ... ... 13

CHAPTER 3: RESEARCH DESIGN ... ... 14

3.1 Research method ... ... 14

3.2 Qualitative research - expert interviews ... ... 14

3.3 Population and sample size ... ... 15

3.4 Questionnaire and data collection ... 15

3.5 Measurements of variables ... ... 16

3.5.1 Contact quality ... ... 17

3.5.2 Waiting area ... ... 17

3.5.3 Appearance repair location... ... 17

3.5.4 Materials quality ... ... 18

3.5.5 Warranty ... ... 18

3.5.6 Speed ... ... 18

3.5.7 Appointment convenience ... ... 18

3.5.8 Reimbursement by insurance company ... ... 18

3.5.9 Repair quality ... 18

3.5.10 Price ... 19

3.6 Reliability and Factor analysis ... 19

3.7 Plan of analysis ... 23

CHAPTER 4: RESULTS ... ... 24

4.1 Response and reliability ... ... 24

4.2 Mean of the constructs... ... 24

4.3 Regression ... ... 26

CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS ... 28

5.1 Conclusions ... ... 28

5.2 Management recommendations ... ... 29

5.3 Limitations and suggestions for future research... ... 29

REFERENCES ... ... 30

(6)

CHAPTER 1: INTRODUCTION

Customer satisfaction is a topic that has interested marketeers for years. Increasingly, the importance of measuring and controlling of customer satisfaction is acknowledged. The main theme for this thesis is customer satisfaction as well. This chapter starts with the initial motivation for this study, then the problem and research method are specified. In the following section the academic and managerial relevance is discussed, and finally the structure of the thesis is outlined.

1.1 Initial motivation

Many companies see customer satisfaction as a requirement to perform well, because it brings them 'good things'. There is much scientific literature describing the positive relationship between higher customer satisfaction and greater customer loyalty (Anderson & Sullivan, 1993; Bearden & Teel, 1983; Bolton & Drew, 1991; Fornell, 1992; Kumar, 2002; LaBarbera & Mazursky, 1983; Mittal & Kamakura, 2001; Oliver, 1980; Oliver & Swan, 1989; Reichheld, 2003; Sharma et al., 1993; Y1, 1990; Zeithaml, 2003). It is believed that customer satisfaction is a great indicator for the company's future revenues (Kotler, 1991, p.19; Fornell, 1992; Rust, Zahorik, & Keiningham, 1994; Rust & Keiningham, 1995), reducing the costs of future transactions (Reichheld & Sasser, 1990); decreasing price elasticity (Anderson, 1996); and minimizing the likelihood of customers defecting in case of faltering quality (Anderson & Sullivan, 1993). Logically, this is what drives these companies. Much has also been written about the determinants (or antecedents) of customer satisfaction (Berry, Parasuraman & Zeithaml, 1994; Parasuraman, Zeithaml & Berry, 1985, 1988; Ravald & Grönroos, 1996; Selnes, 1996; Anderson, Fornell & Lehmann, 1994; Bearden & Teel, 1983; Berry, Seiders & Grewal, 2002; Bolton, Lemon & Verhoef, 2004; Fornell, Johnson, Anderson, Cha, & Bryant, 1996; Naumann & Giel, 1995; Szymanski & Henard, 2001.

Relatively few studies in marketing literature investigate the categories of antecedents of satisfaction. Rather, these studies tend to investigate post-purchase evaluation behaviour, such as the consequences of customer satisfaction. The drivers of customer satisfaction are not often categorized in these studies. Therefore, this study investigates the categories of drivers.

For companies, it is important to know what needs to be done to obtain high levels of customer satisfaction. Furthermore, the antecedents of customer satisfaction, as investigated in marketing science, need to be adapted to the industry they are used in.

(7)

How do service clues influence customer satisfaction? 7 clues. In this study, we explicitly link these three service clues to customer satisfaction. It is important to identify categories of antecedents of customer satisfaction, because this enables companies to manage them by clue management (clearly and consistently designing and orchestrating the total customer experience to guide the mood of the consumer) in order to obtain a better customer service experience and therefore a better customer satisfaction. Business strategies centred on the holistic design and delivery of total customer experience consistently create superior customer value. And they result in stronger, more sustainable customer preference than do independently managed communication, process, and service-centric strategies (Haeckel et al., 2003). For instance:

 According to a case study in the banking industry, tactical use of service clues boosted customer satisfaction, customer loyalty and employee engagement between 10 and 30 % (Deluxe Financial Services, 2005).

 The potential of clue management and clue integration is also shared by the Westin hotel chain case where Westin uses clue management and integration to gain positive brand differentiation, strengthened customer loyalty and a therefore a new revenue stream (Berry et al., 2006)

 Functional clue management is the starting point for any company to build and preserve the customer's confidence in its competence while managing mechanic clues means directly influencing customer service expectations. By managing the humanic clues it is possible to exceed customers’ expectations.

 Clues most be managed by category rather than by the individual drivers of customer satisfaction, because the consumer will evaluate the service as a whole.

The question this thesis provides an answer to is, how do the service clues influence customer satisfaction and are these clues equally important? The conceptual framework of clues and their relationship with customer satisfaction will be tested in this study to find which (categories of) antecedents of customer satisfaction are significantly influencing customer satisfaction. Hypotheses are drawn up on how the three categories of clues affect customer satisfaction.

1.2 Problem specification

The objective of this research is to link the service clues (functional, mechanic and humanic) to customer satisfaction and to find differences in the strengths of these relationships.

(8)

How do service clues influence customer satisfaction?

As part of this central question the following research questions are also posed:

 Which antecedents of customer satisfaction can be identified in marketing literature?  How do these antecedents influence customer satisfaction?

 How can antecedents of customer satisfaction be clustered into service clues?  Which service clues significantly influence customer satisfaction?

1.3 Research method

The research methods used in this thesis were desk research, qualitative research and survey research. A random sample of 500 was drawn from a population of 2000 consumers in the car-repair industry. The response rate was 14.8 %.

1.4 Academic and managerial relevance

This study develops the constructs of functional, mechanic and humanic service clues as introduced by Berry et al. (2006) and empirically tests their influence on customer satisfaction. As stated in paragraph 1.1 much has been written in marketing literature about the consequences of customer satisfaction, but little about the categories of antecedents of customer satisfaction. The importance of the use of categories lies in its potential to manage them and create a total consumer experience. This study increases our knowledge in the field of service clues, studies the dimensionality of the constructs, and studies the constructs in a specific services setting.

The managerial contribution of this research lies in its potential to give insight into the direct relationship between these category clues and customer satisfaction. The model presented in this study will be of benefit to managers, since it provides service companies with an insight into the factors determining customer satisfaction and therefore into their future benefits.

1.5 Structure of the thesis

(9)

CHAPTER 2: RESEARCH FRAMEWORK

This chapter describes the theory regarding customer satisfaction. This literature review answers the following research questions:

 How can antecedents of customer satisfaction be clustered into service clues?

Section 2.1 addresses the definition and antecedents of customer satisfaction and the service clues. In section 2.2, a conceptual model is presented which contains the hypotheses tested in this study.

2.1 Customer satisfaction

In this section, the definition of customer satisfaction is discussed first. Secondly, the antecedents as found in marketing literature are described. Thirdly, the service clues are introduced and in the final section the conceptual framework is presented.

2.1.1 Definition of customer satisfaction

There are various definitions of customer satisfaction in marketing literature. A common difference is in the role of expectations. Kotler (2006, p. 13) defines customer satisfaction as the product's perceived performance relative to a buyer's expectations. According to Kotler, the customer is satisfied when the perceived performance of the product meets the expectations of the customer. When these expectations are exceeded by the perceived performance of the customer, the customer will be very satisfied or even enthusiastic about the product. The true performance of the product is irrelevant because it is the perceived performance which is compared with the expectations of the product.

(10)

How do service clues influence customer satisfaction? 10 Another definition of customer satisfaction is given by Oliver (1980) as a function of expectation and expectancy disconfirmation. Customer satisfaction is determined by the evaluation of the surprise inherent in a product acquisition and/or consumption experience, coupled with the consumer’s prior feelings about the consumption experience. In 2009 Oliver adds the following to his definition: Satisfaction is the consumer's fulfilment response. It is a judgement that the service feature, or the product or service itself, provided (or is providing) a pleasurable level of consumption-related fulfilment, including levels of under- or over-fulfilment. The degree of pleasure determines the satisfaction according to Oliver.

Fornell (1992) defines customer satisfaction as the result of post-purchase perceived product performance compared with pre-purchase expectations, while Anderson, Fornell and Lehmann (1994) define customer satisfaction as the total evaluation of the products and services by the consumer. This means that not only the product of service itself determines customer satisfaction, but also the extra service around the products and service.

Given the variety of definitions of customer satisfaction, it is hard to find a generally agreed-upon definition. There can be satisfaction with events that occur during consumption, with final outcomes, and with the level of satisfaction received. Therefore, satisfaction is determined by the type of consumption (quickly or with a lengthy consumption period). The definition this study uses is the one the AMA proposes, i.e. the degree to which there is a match between the customer's expectations of the product or service and the actual performance of the product.

2.1.2 Antecedents of customer satisfaction

As pointed out in chapter 1, there is much marketing literature about the antecedents of customer satisfaction. The common denominator is that most articles define customer satisfaction as the degree of similarity between perceived service quality and consumer expectations. Table 2.1 gives a summary of antecedents of customer satisfaction as found in empirical marketing literature.

Table 2.1: antecedents of customer satisfaction found in empirical marketing research

Author, journal Researched antecedents Results

Anderson, Fornell & Lehmann (1994)

Expectations of the service Quality

Price

Customer satisfaction is determined by how the perceived performance of a service matches the expectations, by service quality and by price. Bearden & Teel (1983) Expectations

Disconfirmation

Expectations and disconfirmation are plausible antecedents of customer satisfaction

Berry, Parasuranan & Zeithaml (1994) Reliability Responsiveness Assurance Empathy Tangibles

(11)

Berry, Wall & Carbone (2006)

Functional, mechanic and humanic service clues

There are functional, mechanic and humanic antecedents of customer satisfaction which should be managed.

Berry, Seiders & Grewal (2002)

Service convenience Service convenience is an antecedent of customer satisfaction

Bolton, Lemon & Verhoef, 2004

Price, Service quality, DM-instruments, CRM, advertising, distribution channel

Among others, price influences customer satisfaction.

Fornell, Johnson, Anderson, Cha & Bryant (1996)

perceived quality perceived value customer expectations

Perceived quality, perceived value, and customer expectations are determinants of overall customer satisfaction

Naumann & Giel, 1995 Perceived performance determines customer satisfaction

Product- and service quality, price and image determine the perception of customer value Oliver, 1980 Expectations

(dis)confirmation

Expectations, (dis)confirmation and the perceived quality determine customer satisfaction

Parasuraman, Zeithaml & Berry (1985)

Access, Communication,

Competence, Courtesy, Credibility, Reliability, Responsiveness, Securities, Tangibles, Understanding/knowing the customer

Researched antecedents are determinants of service quality.

Parasuraman, Zeithaml & Berry (1988) Service quality: - Reliability - Assurance - Tangibles - Empathy - Responsiveness

The authors find the given items (reliability, assurance, tangibles, empathy and responsiveness) to be antecedents of customer satisfaction.

Selnes, 1996 Competence, communication, Commitment and Conflict handling

Communication, commitment and conflict handling are antecedents of customer satisfaction

Szymanski & Henard, 2001 Meta-analysis of customer satisfaction

Five antecedents of customer satisfaction: - Expectations

- Disconfirmation - Performance - Affection - Equity

As can be seen in table 2.1, customer satisfaction is, in general, determined by the degree of similarity between perceived service quality and the expectations about the service quality by the consumer. Therefore, it is important to understand what determines perceived service quality and how can expectations be influenced. In the next section, service clues are introduced which make it possible to meet, influence and exceed customer experiences.

2.1.3 Service clues

There are three categories of service (or: experience) clues. Each category plays a different role in influencing customer satisfaction. Firstly, there are the

functional clues which are about meeting basic customer expectations. Secondly, there are the mechanic clues which influence first impressions, expectations and value creation. Thirdly there are the humanic clues which are about exceeding customer expectations. In figure 2.1, the concepts

(12)

How do service clues influence customer satisfaction? 12 described by Berry et al., is presented, with the different roles of the clues. The distinction among functional, mechanic, and humanic clues can be subtle. For example, a retail salesperson who answers a customer’s question about when an out-of-stock item will be available is producing both functional and humanic clues. The accuracy of the information is a functional clue. The salesperson’s choice of words and body language are humanic clues. The following paragraphs contain a description of all three categories.

Functional clues support the core of any service, because they address the problem which brings the customer to the market. Some examples of functional clues are a correct diagnosis and treatment for back pain, a clean hotel room, a correct bill from the accountant, a proper repair of a car or an accurate bank statement. This category of clues is the least emotional of the three and is about meeting basic expectations. The two most important dimensions for functional service clues are reliability (the ability to perform the promised service dependably and accurately) and problem resolving (Berry et al., 1994). In this study, the categories of clues are tested for their effect on customer satisfaction. Functional antecedents (or clues) are about meeting basic expectations and by doing this they affect customer satisfaction. However, the importance of the categories of clues varies per industry. For example, humanic clues are more important in industries that are labour-intensive and interactive. This study measures what clues are important in the car-repair industry, which has fewer interactions with the customer than for instance the hotel industry.

Functional clues comprise the evaluation of the core service, which is fundamental to meeting customers' service expectation. As customer satisfaction is determined by the perceived quality of the service and the expectations about the (core) service, the functional clues play an important role in determining customer satisfaction. Therefore, the first hypothesis is:

H1: Functional antecedents significantly affect customer satisfaction

(13)

How do service clues influence customer satisfaction? 13 Coulter, 1995). By doing this mechanic clues influence expectations. These expectations are matched by the customer with the perceived performance and the degree by which this is done successful determines the customer satisfaction (Anderson et al 1994). The second hypothesis is therefore:

H2: Mechanic antecedents significantly affect customer satisfaction

Humanic clues created by employees are most salient for labour-intensive, interactive services. The more important, personal, and enduring the customer-provider interaction, the more pronounced and emotional the humanic effects are likely to be. Examples of humanic clues are the degree to which hospital staff address emotional needs, how well nurses keep patients informed, the customer-friendliness and helpfulness of call-centre employees or greeting every customer that comes in a store. Humanic clues are synonym with employee effort. Customer perception of employee effort in delivering a service has an especially strong impact on service satisfaction and loyalty (Keaveney, 1995; Mohr and Bitner, 1995). Therefore the third hypothesis tested in this study is:

H3: Humanic antecedents significantly affect customer satisfaction

2.2 Conceptual framework

As explained in chapter 1 this study builds on the work of Berry et al. and combines their concepts (see figure 2.1) with marketing literature to come to the following conceptual framework. In the framework, categories of clues are linked with customer satisfaction, because the literature review has proven the link between customer perceptions of service quality and customer satisfaction.

Model 2.1: Conceptual framework

(14)

CHAPTER 3: RESEARCH DESIGN

In this third chapter, an overview of the research method used in this study is given. Next, the data collection, including the sample, is discussed. In section 3.4, the measurement of the variables in this research will be explained, and finally, the data analysis plan can be found in section 3.5.

3.1 Research method

This empirical research consists of two parts. Firstly, qualitative research was conducted to determine the industry-specific service clues. Although the categories of clues are always the same for every industry (functional, mechanic and humanic), they need to be filled with specific variables. In this case, this was done by means of expert interviews. Secondly, the variables were measured with an online survey.

3.2 Qualitative research - expert interviews

The quantitative data for this research was collected through an online survey, after the service clues (or, in this study, antecedents of customer satisfaction) had been realized by using the qualitative data obtained from expert interviews. Expert interviews are semi-structured interviews in which the person interviewed is less important than the function and/or expertise held by that person (Meuser & Nagel, 2002). It is very important to make an interview guide in advance, because the expert sometimes turns out to be not an expert at all, or would rather discuss day-to-day business with the interviewer instead of the subject of the interview. Furthermore, experts often tend to switch from being interviewed to lecturing on their expertise. The interview guide ensures, firstly, that the interviewer appears capable and well-prepared and, secondly, that the expert talks specifically about the topic selected by the interviewer (Flick, 2006, p. 165). For this study, expert interviews were conducted with a CEO, a financial and legal adviser, a manager front-office and three subsidiary managers.

(15)

Table 3.1: service clues according to industry experts

FUNCTIONAL MECHANIC HUMANIC

Repair quality Waiting area Contact quality Speed Appearance of repair location

Convenience Materials quality Price

Warranty

Reimbursement by insurance

3.3 Population and sample size

This study has used the online survey because of the convenience for respondents (who can choose their own time to complete the survey) and the ease of processing (an Excel-output of the results was supplied by the survey software). Because the e-mail addresses of respondents were not always available, respondents were invited to participate in the survey by letter, which contained a simple URL with instructions. The population from which these respondents were randomly chosen consisted of 2000 customers of a car-repair company in the Netherlands from the first quarter of the year 2009.

Because the population of this research was relatively homogeneous, accuracy was set to 0.05 (i.e. a maximum error of 5% compared to the population’s mean). The chosen confidence level was 95%, which is usual in similar research. The formula for calculating the sample size was according to Malhotra (1999): n = [π (1 – π) * z²] / D². The z-value associated with a reliability of 0.95 is 1.96. The following sample size was calculated with this formula and with the chosen accuracy and reliability: N = (0.5 (1-0.5) * 1.962) / 0.052 = 384.16 (= rounded up to 385).

The given sample size of 385 was necessary to be representative of the population. However, to obtain a response of 385 completed surveys, more invitations would need to be sent. According to Anseel et al. (2010), the average response rate for mailed surveys, when 25% of the population is invited, is between 23 and 29 %. It was expected that the industry on which this study focuses would have a lower response rate (circa 25%) because it is of little interest to the consumer. This meant at least 1540 invitations would need to be sent to obtain 385 completed surveys. By personalizing the invitations and using incentives (a raffle of three mobile phones) respondents were encouraged to complete the questionnaire. However, for financial reasons the sample size was limited to 500 customers (see section 5.3 Limitations).

3.4 Questionnaire and data collection

(16)

How do service clues influence customer satisfaction? 16 smaller stages the customer goes through. The stages in the car-repair industry are: making an appointment, bringing the car in, waiting in the waiting room while the car is repaired, reimbursement by the insurance company. The questionnaire ended with a general figure for customer satisfaction. By using this routing for the survey the respondents were able to recognize the service process and relate to it in their memory.

The questions in this questionnaire were based on findings from the literature review and the qualitative research. The experts were not only asked what they considered to be service clues in this industry, but also how these clues are determined. For sub-items of the service clues there were two questions, one which measured the importance of the items and one which measured the perceived realization. Both items were measured so that their multiplication was possible, therefore enhancing the ability to predict the attitudes, in this case customer satisfaction, (Azjen & Fishbein, 2008).

Most questions in the survey have an five-point answer-range (1= fully agree ... 5= fully disagree), because a desirable Likert-type response scale-length ranges from five to eight response options (Lietz, 2010).The demographic questions were put at the end of the questionnaire to avoid negative feelings about the provision of personal information impacting on answering behaviour or participation (Converse & Presser, 1986; Oppenheim, 1992). See appendix 2 for a full overview of the questions used in the survey.

The process of data collection was as follows: firstly, the members of the sample were requested in writing to complete the online survey. The response-rate was enhanced through personalization (an image of the license plate of the invited person was included in the writing) and incentives (three mobile prepaid telephones could be won). Secondly, two weeks after the initial request reminders were sent in writing to the non-respondents. Thirdly, the respondents were thanked for their participation by email. Finally, the winners of the incentives were drawn from the respondents and the phones were sent to the winners.

3.5 Measurements of variables and reliability

(17)

number was measured by rating two statements. The first was: the operator was friendly and helpful

and the second: It important to me that the operator is friendly and helpful. For the analysis of the results, the items were multiplied because, as Azjen and Fishbein (2008) found, multiplying belief' strength and evaluation adds substantially to the prediction of attitudes. What follows next is a description of the questions and measurement method per construct.

3.5.1 Contact quality

According to Selnes (1996), communication, commitment and conflict handling are drivers of satisfaction. Berry, Parasuraman and Zeithaml (1994) find responsiveness (the willingness to help customers and provide prompt service), assurance (the knowledge and courtesy of employees and their ability to convey trust en confidence) and empathy (the caring, individualized attention proved to customers) criteria for customers to judge service quality. The construct contact quality, which incorporates humanic and functional clues, was measured by asking the respondents to rate two items: contact by telephone with the central phone number and contact (by telephone and face-to-face) with the repair location.

3.5.2 Waiting area

The appearance of the physical facility, equipment, personnel and communication materials (or: tangibles) is a criterion for customers to judge service quality (Berry et al., 1994). The mechanic clues of facility design, equipment, furnishings, displays, signs, colours, textures, sounds, and lighting, among other sensory clues, paint a visual picture of the service, are important non-verbal communicators for customers (Berry et al., 2006). The waiting area is therefore a typical mechanic clue. To measure the mechanic clue waiting area, respondents were asked to rate five items indicating how important they found the waiting room for their satisfaction: general pleasantness of the waiting area; quality of the beverages; child-friendliness; the presence of reading material; and a view of their car in the workshop.

3.5.3 Appearance repair location

(18)

3.5.4 Materials quality

Another typical mechanic service clue according to Berry et al. (1994 & 2006) is materials quality which is measured by one question rating the importance of the quality of the parts used for the repair.

3.5.5 Warranty

The functional clue warranty was measured by asking the respondents how important the issued lifelong warranty would be to them. Fornell et al (1996) and Parasuraman et al (1988) note that reliability is a driver of perceived quality which leads to customer satisfaction. Reliability is defined as the degree to which the firm's offering is reliable, standardized, and free from deficiencies. The warranty is an assurance of this reliability for the customer.

3.5.6 Speed

The variable speed is a functional service clue which was measured by respondents rating the speed of planning a repair date and the speed of the repair itself. Ravald and Grönroos (1996) speak of episode value (increasing the benefits / reducing the sacrifice for the customer) which affects the relationship value by enhancing customer satisfaction. By reducing the repair time and the time to appointment the sacrifices for the customer are reduced.

3.5.7 Appointment convenience

Appointment convenience was measured by asking the respondents to rate the convenience and

speed of making an appointment for repair. The episode value discussed in section 3.5.6 is also important here. Also Berry, Seiders and Grewal (2002) discuss various types of service convenience. The type of convenience that is measured here is called access convenience. Access convenience involves consumers' perceived time and effort expenditures to initiate service delivery.

3.5.8 Reimbursement by insurance company

The functional antecedent of customer satisfaction reimbursement by insurance company was measured by asking the respondents to rate the importance of reimbursement and of an excess waiver by the insurance company.

3.5.9 Repair quality

The functional antecedent repair quality was measured by asking the respondents to rate the importance of the quality of the repair. Parasuraman, Zeithaml and Berry (1985) find that being

(19)

How do service clues influence customer satisfaction? 19 respect and an explanation of the suggested repairs), it could be defined as the core competence of the car-repair industry. Fornell et al (1996) also find that the perceived performance is one of the key drivers for customer satisfaction.

3.5.10 Price

The final functional clue is the price which was measured by two items: the rating of the price itself and the rating of price/quality ratio. According to Fornell et al. (1996), the perceived value is a determinant of overall customer satisfaction. The perceived value is the perceived level of product quality relative to the price paid. Bolton, Lemon and Verhoef (2004) consider price to be one of the determinants of customer behaviour, particularly relationship, usage, cross-buying, word-of-mouth communication which are al consequences of customer satisfaction (see Section 1.1)

3.6 Reliability and Factor analysis

Factor analysis was performed to analyze the uni-dimensionality of the constructs. According to Malhotra (2008), factor analysis is used for data reduction and testing the dimensionality of the constructs. This study is using factor analysis to determine to which category of service clues the different constructs belong. This means factor analysis is used for testing the dimensionality of the constructs to find three factors (or categories of service-clues).

Factor analysis and correlation analysis were used for data reduction to find factors which consist of the ten constructs. This is done because the constructs are measured in the questionnaire by a variable number of questions. The following tests were conducted:

1. Factor analysis on all twenty-two items. 2. Cronbach's alpha on the ten constructs. 3. Factor analyses on ten constructs.

4. Factor analysis per category of constructs. 5. Cronbach's alpha on the categories.

(20)

Table 3.2 Results factor analysis on all items

FACTOR INITIAL EIGENVALUES % OF VARIANCE CUMMULATIVE

VARIANCE % Factor 1 8.483 38.560 38.560 Factor 2 2.554 11.611 50.170 Factor 3 1.746 7.937 58.107 Factor 4 1.322 6.010 64.118 Factor 5 1.075 4.885 69.002

The constructs were measured by a different number of items. For instance, the construct waiting

area was measured by five items, but repair quality by just one. The internal reliability of the

constructs was measured with a Cronbach's alpha analysis. Using Cronbach's alpha for each construct determines if the items could be plotted on to a single scale for comparison. Table 3.3 gives an overview of the reliability of the constructs measured by more than one question in the final column.

Table 3.3: Tested constructs

CONSTRUCT VARIABLES SOURCE ITEMS USED RELIABILITY

Humanic service clues

Contact quality  Selnes (1996)

 Berry, Parasuraman & Zeithaml (1994)

 Expert interviews

 Operator central phone number

 Contact with workers on location

.658*

Mechanic service clues

Waiting area  Berry, Parasuraman & Zeithaml (1994)

 Berry, Wall & Carbone (2006)

 Expert interviews  Pleasantness  Beverages  Child-friendliness  Reading material  View on car .874 Appearance repair location

 Clean and dry workshop

 Recognizable repair location

 Distance from house

 Directions to location

.803

Materials quality  Anderson, Fornell & Lehmann (1994)

 Fornell, Johnson, Anderson, Cha & Bryant (1996)

 Naumann & Giel (1995)

 Materials used for repair -

Functional service clues

Repair Quality  Parasuraman, Zeithaml & Berry (1985)

 Quality of the repair -

Speed  Ravald & Grönroos (1996)

 Speed of repair Appointment

convenience

 Berry, Seiders & Grewal (2002)

 Expert interviews  Convenience of making an appointment  Speed of making an appointment .673*

Price  Anderson, Fornell & Lehmann (1994)

 Bolton, Lemon & Verhoef (2004)

 Fornell, Johnson, Anderson,

 Fair price

 Price - Quality

(21)

Cha & Bryant (1996)

 Expert interviews Warranty  Parasuraman, Zeithaml &

Berry (1988)

 Warranty -

Reimbursement by insurance company

 Expert interviews  Reimbursement of full bill

 Excess waiver

-**

Customer satisfaction

 General grade for service.

-

* Measured with a Pearson correlation analysis because of the two-item scale and significant at the 0.01 level (two-tailed) (Cohen et al., 2003).

** Because of the low Pearson correlation score (.301) one item was removed from the results. The item excess waiver was removed.

Table 3.4 shows the means and standard deviations of the new variables that measure the constructs. For the constructs measured by more than one item the items were computed to one scale.

Table 3.4 Construct scores

CONSTRUCT MEAN STANDARD

DEVIATION

Functional service clues

Repair quality 4.68 .506

Speed* 4.55 .628

Appointment convenience 4.56 .536

Price* 4.26 .746

Warranty 4.41 .709

Reimbursement by insurance company 4.47 .904

Mechanic service clues

Materials quality 4.35 .753 Appearance repair location* 3.66 .736 Waiting area* 4.22 .781

Humanic service clues

Contact quality* 4.54 .543 * = New variable, consists of average of all questions measuring the construct (see Table 4.1)

The results as presented in table 3.4 show that repair quality, speed and appointment convenience were considered to be the most important by the respondents. These three constructs all reduce the sacrifice for the customer. According to Ravald and Grönroos (1996), this means that episode value (which consists of benefits and sacrifices) is created which drives customer satisfaction.

(22)

Table 3.5 Variance explained for found factors

VARIANCE %

Factor 1 4.721 47.206 47.206 Factor 2 1.142 11.421 58.627

Table 3.5 shows that there are two factors with an Eigenvalue exceeding 1. These factors account for 58.6% of the total variance. Table 3.6 shows which items correlate most with the various factors.

Table 3.6: results factor analysis on constructs

COMPONENT

CONSTRUCT 1 2

Contact quality .469 .680 Waiting area .838 .009 Appearance repair location .749 .300 Materials quality .451 .544 Repair quality .284 .744 Speed .532 .381 Appointment convenience .629 .438 Price .657 .178 Warranty .384 .686 Reimbursement by insurance .083 .769

The constructs contact quality, appearance repair location, materials quality, speed, appointment

convenience, and warranty have too high a cross-loading. This means they have a high loading (>0.3)

on more than one factor. For this reason, these constructs would need to be removed from further analysis. Unfortunately again, SPSS did not find three factors which clearly correspond to the three categories of service clues (functional, mechanic and humanic). Only two factors were found, while the distribution of the constructs on the factors strongly differs from the aforementioned distribution (see table 3.1). It is assumed that this is caused by the divisibility of the constructs. It is possible to further divide the items that are used to measure the constructs. For example, the item friendliness

and helpfulness of the operator at the central telephone number can consist of two sub-items was the

information given correct (functional), but also was the tone-of-voice used appropriate to the conversation (humanic). From theoretical considerations we continue with the original distribution on the bases of the literature and qualitative research.

To test if factor analysis would find just one factor per category of service-clues, two separate factor analyses were undertaken on the constructs for the mechanic and the functional service-clues. The results in table 3.7 show that this is indeed the case.

Table 3.7: results of two separate factor analyses on items for the mechanic and the functional service clues

VARIANCE %

Mechanic service clues: appearance repair location, waiting area & materials quality

2.059 68.630 68.630

Functional service clues: repair quality, speed, appointment convenience, price, warranty, reimbursement

(23)

How do service clues influence customer satisfaction? 23 Next, three new variables were formed on the basis of the described literature (see table 3.1 for the distribution of the constructs across the categories) because the factor analysis described in chapter 3 did not result in the three constructs of functional, mechanic and humanic service-clues. The reliability of the three categories of service-clues was tested by Cronbach's alpha analysis. The results of this test are presented in Table 3.8.

Table 3.8 Internal reliability of the constructs.

CATEGORIES OF SERVICE CLUES CRONBACH'S ALPHA N OF CONSTRUCTS

Functional service clues .753 6 Mechanic service clues .767 3 Humanic service clues 1

The categories of service clues are reliable according to the Cronbach's alpha analysis, because their Cronbach's alpha is higher than the required 0.6. The category of functional service clues consists only of the item contact quality.

3.7 Plan of analysis

(24)

CHAPTER 4 RESULTS

In this chapter the results of the online survey are discussed. In the first section, the response and the representativeness of the sample are analyzed. Then, in the next three sections the results are described; next the clue results are described; and, finally, an overview of the influence of service clues on customer satisfaction is presented in the research framework.

4.1 Response and reliability

In all, 78 surveys were completed out of a possible 500. Three respondents only partially completed the survey. One respondent provided the same answer throughout. The usable response rate was therefore 74 respondents, which corresponds to a response rate of 14.8 %. This is lower than the expected 25 %.

The response rate does not meet the reliability requirements for the sample to be representative, because the required number of 385 respondents was not met. The confidence interval that corresponds to this response rate is 61 %.

Calculation of the confidence interval: n = (0.5 (1-0.5) * z2) / 0.052 = 74 z2 = 0.74

z= 0.86

The z-value of 0.86 corresponds to a confidence interval of 0.61. Therefore, the outcomes of the analyses have to be interpreted with care.

4.2 Mean of the constructs

The mean and standard deviation per question are presented in table 4.1. For all constructs the importance and the realization are measured. Standard deviations are given in order to assess the distribution of the answers.

Table 4.1 Means and standard deviations per question (measured on a five-point scale)

CONSTRUCT/ITEM QUESTION IMPORTANCE REALIZATION

Humanic service clues

Contact quality The operator of the central telephone number was friendly and helpful.

4.57 S=.664

4.28 S=.899 Contact quality The workers at the repair location were friendly and

helpful (on the telephone).

4.73 S=.531

4.95 S=.739

Mechanic service clues

Waiting area I waited in a pleasant waiting area. 4.12 S=.859

3.82 S=1.151 Waiting area I liked the beverages in the waiting area. 4.05

S=1.032

(25)

Waiting area The waiting area was suitable for children. 3.36 S=1.154

3.26 S=1.034 Waiting area The reading material in the waiting area was nice to read. 3.45

S=1.049

3.41 S=1.097 Waiting area From the waiting area keep I could view the repair of my

car. 3.61 S= 1.203 3.73 S=1.114 Appearance repair location

The repair was executed in a clean and dry workshop. 4.43 S=1.021

4.35 S=1.013 Appearance repair

location

The repair location was easy to recognize from the outside. 3.96 S=1.164 3.73 S=1.208 Appearance repair location

The distance between my house and the repair location was not too long.

4.42 S=.860 4.12 S=1.110 Appearance repair location

It was easy to find the repair location. 4.55 S=.779

4.23 S=1.105 Materials quality The parts that were used for the repair of my car are of

the highest (Original Equipment) quality.

4.61 S=.737

4.09 S=1.009

Functional service clues

Repair quality The repair of the car is of high quality. 4.77 S=.511

4.58 S=.597 Speed The repair was the repair was carried out as quickly as

possible so that the waiting time was limited.

4.62 S=.676 4.49 S=.815 Appointment convenience

It was possible for me to make an appointment for repair at short notice. 4.58 S=.662 4.47 S=.815 Appointment convenience

The process of making an appointment was convenient. 4.68 S=.576

4.51 S=.798 Price I paid a fair price for the repair. 4.27

S=.880

4.14 S=.881 Price The price in comparison to the quality is fair. 4.42

S=.759

4.22 S=.781 Warranty The service provider gives me a lifelong guarantee on the

repair. 4.51 S=.707 4.31 S=.875 Reimbursement by insurance company

The insurance company reimbursed the bill. 4.65 S=.766

4.28 S=1.266 Reimbursement by

insurance company

The insurance company waives my excess when I choose this service provider.

4.27 S=1.102

3.65 S=1.512

Customer satisfaction How would you rate the provided service? 8.51 S=1.024

As table 4.1 shows, customers find the quality of the repair of the car most important. It is the core service in this case. The mean of this item is 4.77 on the five-point Likert scale (which was used for all items). Contact quality (4.73), convenience (4.68) and reimbursement (4.65) were the other items the consumers rated as important. The items which measured the construct waiting area were rated least important. In only two cases the realization had a higher score than the importance. The realization of the item that measured the friendliness and helpfulness of workers at the repair location is remarkably high at 4.95. Another remarkable fact is that on average the importance score is lower than the realization.

(26)

How do service clues influence customer satisfaction? 26 4.3 Regression

To test the hypotheses three separate regression analyses were performed to discover the impact of each independent variable separately on the dependent variable, customer satisfaction. The hypotheses H1 and H2 can be confirmed according to the results shown in table 4.6. All three categories have a positive influence on customer satisfaction, but for humanic service-clues this effect is not significant. The positive beta values mean that when the value of these factors rises, customer satisfaction will rise as well. The most influential category is the category of functional clues with a beta value of .790.

Table 4.6 Results separate regression analyses.

DEPENDENT: CUSTOMER SATISFACTION R2 BETA STD. ERROR SIGNIFICANCE

Functional service clues .123 .790 .248 .002 Mechanic service clues .117 .526 .170 .003 Humanic service clues .040 .375 .199 .089

To find out if the clues have an effect above and beyond the effect of the other clues a multivariate regression analysis was conducted. Table 4.7 shows the results of this analysis.

Table 4.7 Results multivariate regression analysis

DEPENDENT: CUSTOMER SATISFACTION BETA STD. ERROR SIGNIFICANCE

Constant 5.126 1.121 .000 Functional service clues .307 .212 .152 Mechanic service clues .752 .425 .081 Humanic service clues .261 .314 .408 F Value = 4.388

R =.398 R2 =.158

(27)

Model 4.1: Results in conceptual framework

(28)

CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS

The objective of this study was to test whether the service clues (functional, mechanic and humanic) described by Berry et al (2006) affect customer satisfaction and to find differences in the strengths of the relationships. The central question of this study was: How do service clues influence customer

satisfaction?

To answer this, the following research questions were formulated for this study:

 How can antecedents of customer satisfaction be clustered into service clues?  Which service clues significantly influence customer satisfaction?

Afterwards a literature review and expert interviews were conducted to pinpoint the antecedents (or: service clues) which influence customer satisfaction. The resulting antecedents were tested for significance. In this chapter, first of all the most important results of these tests are reported in the conclusions. Secondly, management recommendations are given and the limitations of this study are discussed. Finally, recommendations for further research are given.

5.1 Conclusions

Multiple regression analysis was executed in order to test the hypotheses for their significance. Table 5.1 shows an overview of the hypotheses, their significance level, whether the hypothesis is confirmed or not, and what their relative importance is in describing customer satisfaction.

Table 5.1 Hypotheses results

HYPOTHESES CONFIRMED? INFLUENCE

H1 Functional antecedents significantly affect customer satisfaction.

YES + H2 Mechanic antecedents significantly affect customer

satisfaction.

YES + H3 Humanic antecedents significantly affect customer satisfaction. NO +

(29)

How do service clues influence customer satisfaction? 29 5.2 Management recommendations

When managers in the car-repair industry want to increase their customer satisfaction, they should focus on improving the feelings of the customers regarding the functional service-clues. In this case, that would mean the quality of the repair, the speed of repair, the convenience and speed of making an appointment, the price (in respect to the quality), the warranty and the reimbursement by the insurance company. For insurance companies who send their customers to service providers, the functional clues should be the criteria for choosing their suppliers, because these are the drivers of customer satisfaction.

5.3 Limitations and suggestions for future research

The sample size used in this study was limited to 500 for budgetary reasons. When the average reported response rate set by Anseel et al. (2010) as between 23 and 29 % is taken into account, the sample size should have been at least 1540. The response rate of the online survey used in this study is 14.8 %. Given this low response rate the sample size in this study should be have been 2602. It would have been better to further enhance the response rate, for example by enhancing the convenience to the respondents by sending them invitations by e-mail with a direct link to the online survey, instead of inviting them to participate in writing.

Looking back on this study, the service-clues which were measured were formulated too broadly. It is to be recommended that the service-clues are measured on a narrower basis. For instance, the item measuring contact quality of the operator should be divided into: accurate information, tone-of-voice, friendliness, helpfulness, etc. The items used in this study contain more than one type of service clue. For example, the item the operator of the central telephone number was friendly and

helpful contains functional service-clues (is the right information given?) and humanic service-clues

(30)

REFERENCES

Anderson, E.W. (1996). Customer satisfaction and price tolerance. Marketing Letters, 7(3), 265–274. Anderson, E.W., Fornell, C. and Lehmann, D.R., 1994, "Customer Satisfaction, Market Share, and

Profitability: Findings from Sweden". Journal of Marketing 58, 2, 53-66.

Anderson, E.W. and Sullivan, M.W. (1993), "The antecedents and consequences of customer satisfaction for firms". Marketing Science, vol. 12(2), pp. 125-143

Anseel, F., Lievens, F., Schollaert, E. and Choragwicka, B. (2010) "Response Rates in Organizational Science, 1995-2008". Journal of Business & Psychology, vol. 25, pp 335–349

Ajzen, I. and Fishbein, M. (2008), Scaling and Testing Multiplicative Combinations in the Expectancy– Value Model of Attitudes. Journal of Applied Social Psychology, 38: 2222–2247.

Bearden, W.O., & Teel, J.E. (1983). "Selected determinants of consumer satisfaction and complaint reports". Journal of Marketing Research, 20(1), 21–28.

Berry, L.L., Parasuraman, A., and Zeithaml, V.A. (1994). "Improving service quality in America: Lessons learned". Academy of Management Executive, 8(2):32– 44.

Berry, L.L., Seiders, K., & Grewal, D. (2002). "Understanding service convenience". Journal of

Marketing, 66, pp. 1–17.

Berry, L.L.; Wall, E.A.; Carbone, L.P, "Service Clues and Customer Assessment of the Service

Experience: Lessons from Marketing". Academy of Management Perspectives, May2006, Vol.

20 Issue 2, p43-57

Bolton, R.N., & Drew, J.H. (1991). A multi-stage model of customers’ assessments of service quality and value. Journal of Consumer Research, 17(4), 375–384.

Bolton, R.N., Lemon, K.N. and Verhoef, P.C. (2004). The Theoretical Underpinnings of Customer Asset Management: A Framework and Propositions for Future Research. Journal of the Academy of

Marketing Science July 2004 vol. 32 no. 3 pp. 271-292

Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences. Mahwah, New Jersey: Lawrence Erlbaum Associates. Converse, J. & Presser, S. (1986) Survey Questions. Handcrafting the Standard Questionnaire. London:

Sage.

Deluxe Financial Services (2005). 2005 Deluxe knowledge exchange expo redefines relationship

between financial institutions and customers. Deluxe Financial Services Press Release. 25

April 2005.

Flick, U. (2006), “An Introduction to Qualitative Research”, Sage, third edition

Fornell, C. (1992). A national customer satisfaction barometer: The Swedish experience. Journal of

Marketing, 56(1), 6–21.

Fornell, C., Johnson, M.D., Anderson, E.W., Cha, J. and Bryant, B.E. (1996). "The American Customer Satisfaction Index: Nature, Purpose, and Findings". The Journal of Marketing Vol. 60, No. 4, pp. 7-18

Gubrium, J. F. and Holstein, J. A., eds. Handbook of interview research: context & method. Thousand

Oaks, CA: Sage, 2002. xiii

Haeckel, S.H., Carbone, L.P., & Berry, L.L. 2003. How to Lead the Customer Experience. Marketing

Management, 12(1), 18-23.

Keaveney, op.cit. and Mohr, L.A., & Bitner, M.J. 1995. The Role of Employee Effort in Satisfaction with Service Transactions. Journal of Business Research, 32:239-52.

Kotler, P.(1991), "Marketing Management - Analysis, Planning, Implementation and Control", 7th Ed., Englewood Cliffs, NJ: Prentice-Hall, Inc.

Kotler, P. and Armstrong, G. (2006), "Principles of marketing", Prentice Hall, 11e editie Kumar, P. (2002). "The impact of performance, cost, and competitive considerations on the

relationship between satisfaction and repurchase intent in business markets". Journal of

(31)

How do service clues influence customer satisfaction? 31 LaBarbera, P.A., & Mazursky, D. (1983). A longitudinal assessment of consumer satisfaction/

dissatisfaction: The dynamic aspect of the cognitive process. Journal of Marketing Research,

20(4), 393–404.

Lietz, P. (2010), "Research into questionnaire design – A summary of the literature" International

Journal of Market Research 52.2, pp. 249-272.

Malhotra, N.K.(2008), ‘Marketing Research : An Applied Orientation’, 5e editie, Prentice hall Meuser, M. and Nagel, U., (2002), “ExpertInneninterviews – vielfach erprobt, wenig bedacht. Ein

beitrag zur qualitativen Methodendiskussion” in Bogner, A., Littig, B. en Menz, W (2002), “Das Expertinterview”, Opladen: Leske & Budrich, pag. 71-95.

Mittal, V., and Kamakura, W., (2001), “Satisfaction, RepurchaseIntent, and Repurchase Behavior: Investigating the Moderating Effect of Customer Characteristics,” Journal of Marketing

Research, vol. 38, pp 131–42

Mukhopadhyay, I., Bandyopadhyay, S. K. and Chatterjee, A. (2011) "Prioritisation of the determinants of customer satisfaction: A simultaneous equation approach in ordinal endogenous set-up". Total Quality Management & Business Excellence, 22: 1, 117 — 130 Oliver, R.L. (1980). A cognitive model of the antecedents and consequences of satisfaction decisions.

Journal of Marketing Research, 17(4), 460–469.

Oliver, R.L., & Swan, J.E. (1989). Consumer perceptions of interpersonal equity and satisfaction in transactions: A field survey approach. Journal of Marketing, 53(2), 21–35.

Oliver, R.L. (2009), "Satisfaction: a behavioral perspective on the consumer", 2nd ed. M.E. Sharpe, Inc.

Parasuraman, A., Zeithaml V., & Berry L. (1985). A conceptual model of service quality and its implications for future research. Journal of Marketing, Vol 49, pp 41-50

Parasuraman, A., Zeithaml V., & Berry L. (1988). SERVQUAL: A Multiple-item Scale for Measuring Consumer Perceptions of Service Quality. Journal of Retailing, vol 64 Issue 1, pp 12-40 Presser, S. & Blair, J. (1994) Survey pretesting: do different methods produce different results?

Sociological Methodology, 24, pp. 73–104.

Ravald, A. and Grönroos, C. (1996), “The value concept and relationship marketing”, European

Journal of Marketing, Vol. 30 No. 2, pp. 19-30.

Reichheld, F.F., & Sasser, W.E., Jr. (1990). Zero defections: Quality comes to services. Harvard Business Review, 68(5), 105–111.

Reichheld, F.F. (2003), "The one number you need to grow", Harvard Business Review, dec. pp. 46–54 Rust, R.T., & Keiningham, T.L. (1995). Return on quality (ROQ): Making service quality financially

accountable. Journal of Marketing, 59(2), 58–70.

Rust, R.T., Zahorik, A.J., & Keiningham, T.L. (1994). Return on quality: Measuring the financial impact

of your company’s quest for quality. Chicago, IL: Probus Professional Pub.

Yi, Y. (1990). A critical review of consumer satisfaction. In V.A. Zeithaml (Ed.), Review of Marketing

(pp. 68–123). Chicago, IL: American Marketing Association.

Zaltman, G. 2003. How Customers Think: Essential Insights into the Minds of the Market, Boston, MA: Harvard Business School Press

Zaltman, G. 1996. Metaphorically Speaking. Marketing Research, 8 (2): 13-20;

Zaltman, G. & Coulter, R. H. 1995. Seeing the Voice of the Customer: Metaphor-Based Advertising Research, Journal of Advertising Research, 35(4):35-51.

(32)

APPENDIX I - INTERVIEW GUIDE EXPERT INTERVIEWS

The interviews were conducted following the steps in this interview guide. 1. Thanking the expert for their time

2. Introduction of the research by interviewer 3. General questions:

o What experience do you have with customer satisfaction in your industry? o What aspects are in your view the antecedents of customer satisfaction? 4. Introduction of service-clues by interviewer

5. Specific questions:

o Given the service-clues what would you consider to be functional clues for your industry? o Given the service-clues what would you consider to be mechanic clues for your industry? o Given the service-clues what would you consider to be humanic clues for your industry? o Which category of clues would you expect to be the most important?

(33)

APPENDIX II - THE QUESTIONNAIRE

The questionnaire follows the stages of the customers’ journey through the process.

To start your questionnaire, enter your license plate number.

Making an appointment

1a. It easy for me to make an appointment for repair.

1b. I find it important that it is easy to make an appointment for repair. 2a. I can make an appointment at short notice.

2b. I find it important to be able to make an appointment at short notice. 3a. The operator at the central telephone number was friendly and helpful.

3b. I find it important that the operator at the central telephone number is friendly and helpful. 4a. The workers at the repair location were friendly and helpful.

4b. I find it important that the workers at the repair location are friendly and helpful.

The repair

5a.The mechanics are very skilful.

5b. I find it important that the mechanics are skilful. 6a. The repair of my car is of high quality.

6b. I find it important that the repair of my car is of high quality.

7a. The parts that are used for the repair of my car are of the highest (Original Equipment) quality. 7b. I find it important that the parts that are used for the repair of my car are of the highest (Original Equipment) quality.

8a. The car is repaired using the most innovative repair methods. 8b. I find it important that use is made of innovative repair methods.

9a. The repair was carried out as quickly as possible so that the waiting time was limited. 9b. I find it important that repairs are carried out as quickly as possible.

Repair location

10a. The repair was executed in a clean and dry workshop.

10b. I find it important that repairs are executed in a clean and dry workshop. 11a. It was easy to find to the repair location.

11b. I find it important that the repair location is easy to find.

12a. The distance between my house and the repair location was not too long.

12b. I find it important that the distance I have to travel to get my car repaired is not too long. 13a. The repair location was easy to recognize from the outside.