• No results found

Promises versus reality : customer experience among the internet service industry

N/A
N/A
Protected

Academic year: 2021

Share "Promises versus reality : customer experience among the internet service industry"

Copied!
62
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Master Thesis

Author: R.F. Grobbe Student number: 0588393

Master thesis Business Studies variant Marketing Supervisor: dr. H. Güngör

First reader: dr. A.E. Keegan Second reader: prof. dr. N.R.Anderson FEB, Universiteit van Amsterdam

Amsterdam, August 2007

Promises versus reality;

(2)

Acknowledgments

This thesis study is conducted following the business research methods of Cooper and Schindler (2003). It presents the results of six months of work researching the customer experience by Internet service providers. The same research is done for the mobile telecom industry and the health insurance industry; together they deliver an overview of the current customer experience situation.

The subject of this thesis fits the followed master courses; Consumer behaviour and marketing strategy, and ICT and services. Furthermore, it directly relates to my working experience at BBned, where I, when I graduate, could have a career. This thesis research gave me the opportunity to challenge myself to compare the theories with practice.

Without the supervision of dr. H. Güngör, of the University of Amsterdam this thesis would not have been possible. He provided me with the right feedback at the right time. His constructive criticisms during the process were well-founded and worked as a stimulating factor to me. My thesis team members, Patricia Lakeman and Anneloes van der Wijden were also very supportive and helpful. The discussions and support during the research brought this study to a higher level.

I owe a great deal to all my fellow colleagues at BBned whom provided me with market information and the right contacts. Also, I’d like to mention the great willingness to help and aid given to me by all the employees of Internet service providers, whom provided me with the information I needed.

Finally, I would like to thank my family and friends for their understanding in that I was busy the last six months, but also for their support, feedback and editorial remarks.

Rajesh Grobbe

(3)

Abstract

Customer experience is an economic and psychological term which does not receive enough attention in the service industry (Johnson & Fornell, 1991). The internet service industry is a typical example of an industry with a strong focus on fulfilling customers’ needs. To create some competitive advantage firms should keep the promises that they make to their customers (Grönroos, 1998). If they fail in keeping these promises, customers get dissatisfied and will search for another provider, also known as switching behaviour (Keaveney, 1995). This report presents the significant negative gaps between expectations and promises (0.9243), a positive gap between perceptions and promises (0. 6985) and a negative gap between expectations and perceptions (0. 7104). The study is based on a literary search and significant testing by

interviewing (n is 5), surveys (n is 100) and mystery shopping (n is 100). The collected data is analysed for significant correlations which Internet service providers should face and use if they want to improve their customer experience. Also could be stated that Internet service providers are sales-orientated. With the conducted dataset, it is possible to compare these figures with other service industries.

(4)

Table of contents

1. Introduction ... 6

1.1 The Internet industry ... 6

1.2 Problem statement & research purpose ... 7

1.3 Research questions ... 8

2. Literature review ... 10

2.1 Service delivery process... 10

2.2 Customer experience ... 13 2.2.1 Servqual-model ... 13 2.3 Constructs... 15 2.4 Promises ... 15 2.5 Expectations ... 16 2.6 Perceptions ... 17 2.7 Costs ... 18 2.8 Experience... 18 2.8.1 Staff attitude ... 18 2.8.2 To busy to respond ... 19 2.8.3 Reliability ... 19 2.8.4 Courtesy ... 19 2.8.5 Empathy ... 20

2.8.6 Understanding the customer... 20

2.9 Expertise... 21 2.10 Speed of help ... 22 2.11 Satisfaction ... 24 2.12 Word-of-mouth... 24 2.13 Loyalty ... 25 2.14 Conceptual model... 27 3. Methodology ... 28 3.1 Sample design ... 28 3.2 Methods ... 28 3.3 Data analysis ... 29 3.4 Resources ... 30 4. Findings ... 31

(5)

4.1 Promises ... 31

4.2 Expectations ... 32

4.3 Perceptions ... 33

4.3.1 Mystery shopping vs. Survey ... 33

4.4 Gap analysis ... 35 4.4.1 Gap1 ... 35 4.4.2 Gap2 ... 36 4.4.3 Gap3 ... 37 4.5 Correlation analysis... 38 4.5.1 Remarkable correlations... 40 4.6 C-sat ... 42 5. Conclusions ... 43 5.1 Conclusions ... 43 5.2 Recommendations ... 45 5.3 Future research ... 46 6. Appendices ... 47

Appendix 1; form for interviewing ... 47

Appendix 2: survey ... 49

Appendix 3; form for mystery shopping ... 52

Appendix 4; gap and correlation analysis ... 53

Orange ... 53 Scarlet... 54 Tele2... 55 Tiscali ... 56 XS4ALL ... 57 7. Bibliography... 58

(6)

1. Introduction

1.1 The Internet industry

The Dutch internet industry is divided in two separated markets which are responsible for a total of 5.042 million broadband connections. On the one side there are the DSL internet service providers with a total responsibility of 3.081 million connections and on the other side there are the cable internet service providers with a responsibility of 1.961 connections. The Dutch broadband market grew by 20 percent, or 891.000 connections, to 5.042 million

connections on 31 December 2006, versus 4.15 million on 31 December 2005. The number of DSL connections grew 23 percent, with 581.000 lines to 3.081 million connections on 31 December 2006 versus 2.5 million on 31 December 2005. The number of cable connections did not grow as fast as DSL with an annual growth of just 18%, adding only 311.000 during the year 2006 to end with 1.961 million cable broadband connections versus 1.650 million on December 2005 (Telecom paper, 2006).

KPN addresses the broadband market with four major brands (Planet Internet, Het Net, XS4ALL and Direct ADSL). The combined KPN internet service provider customer base, including customers’ from acquired ISP’s (Speedlinq and Demon), grew by 579.000 year on year to 2.044 million on 31 December 2006. KPN’s market share on the broadband market grew to 40,5 percent. On the DSL market, the share of KPN-related ISP’s increased from 58.6 percent on 31 December 2005 to 66.3 percent of 31 December 2006. Remarkable is the fact that for the first time since 2002, none of the top four broadband internet service providers had a quarterly growth higher than the average market growth. Het Net and Chello reported 4,3 percent and 4 percent growth, while the market growth for the quarter was 4,7 percent. Orange Broadband (previously known as Wannadoo) continued to loose customers, seeing its market share dropping below 10 percent at the end of 2006 (Telecom paper, 2006).

The market is expected to reach around 5,6 million connections at the end of 2007. The main growth drivers will be VoIP and IPTV service offers that will push people to use broadband internet connections for other reasons than internet access only. At same time increasing competition between ISP’s offering cheap introduction prices will see the last dial-up users transferring to broadband during the coming quarters. The cable and DSL network operators will face increasing competition from fiber-to-the-home initiatives around the country, especially in newly built suburbs and large cities like Amsterdam (Citynet) and Rotterdam (OntwikkelingsBedrijf Rotterdam). The increasing download speeds offered by mobile internet technologies as HSDPA and mobile WiMAX will also put pressure on the

(7)

operators. The new technologies will initially only put pressure on the business market. The broadband consumer market is reaching a saturation point, where every household in the Netherlands that currently wants a broadband connections, has one. The broadband

penetration per household increased to 70,8 percent at the end of 2006 versus 58,5 percent on the end of 2005, representing a year-on-year growth of 12,3 percent. The OPTA estimates that the saturation point lies around the 85 percent, which is reachable in the year 2007 (Telecom paper, 2006).

The consumer benefits from the increased competition in the markets for communication. The market for broadband and internet access is leader in this issue (Heliview, 2007). It is now-a-day possible for the consumer to choose for one connection with different services, like internet, television and telephony. This convergence increased the competitive pressure on the service providers. As a result of that pressure the consumer is now dealing with low pricing and better services. Heliview (2007) reported that there is a switching behaviour of 24 percent in the internet industry. This figure consists of five different groups, there was a group that already switched (7%), there was a group that was in the switching process (2%), there was a group that tried or had the intention to switch (2%) and there was a group planned to switch the next year (13%). This is a sign of healthy market dynamism (Kwaku Appiah-Adu, 1998).

Switching behaviour is a result of encounter failures (Kelley, Hoffman and Davis, 1993). Service encounter can be defined as “the personal interaction between employees and customers of service providers” (Keaveney, 1995). Encounter failures were attributed to behaviour and attitudes of employees of the service firms, like willingness to help, being too busy to respond, trust, politeness, personal attention and knowledgeableness (Zeithaml et al, 1988). According to research of the OPTA (2007) there is a strong switching behaviour among customers in the Internet industry. The main reasons of this switching behaviour are the strong price competition and the service quality. Also the front office is one of the main factors for switching to another provider. Reasons for not switching are the possibilities for technical and administrative problems (NSS, 2007).

1.2 Problem statement & research purpose

The main focus of this research will lie on the interaction aspect of internet providers and their customers. Through the divided market internet providers are winning their competitive advantage on a service based level through encounter services (Keaveney, 1995). To

(8)

investigate what the customer experience is based on, it is necessary to identify the promised services of the firm and compare them with the expectations and the perceptions of the customer and see of it fits. If a difference appears, it becomes possible to give some recommendations to improve the focus on customer experience among the internet service industry. Managing the customer experience will be a new challenge among contact centers. Dissatisfied customers are dangerous for the continuation of the Internet service providers. This research is focusing on the DSL providers with a primary focus on the consumer market. The business market and cable retailers are excluded.

The purpose of this research is to answer how it is possible to improve the customer experience among internet service providers. The recommendations are conducted by an analysis of the constructs that together forms the current customer experience situation. And these recommendations are based on a comparable dataset which could be interpreted to other service industries. The correlation between customer experience and word-of-mouth,

satisfaction and loyalty should be tangible. Based on these figures it is possible to produce specific recommendations were service providers should focus on creating a better customer experience and receive more competitive advantages. This research has a practical relevance and is valuable for the Internet service providers, but also indirectly for the customers of the providers. The research outcomes can be used for improvements to customer contact centers in the Internet industry. This research is ethically acceptable, since the collecting of data is based on public academically literature. Customers of the different Internet service providers are answering on a voluntary basis. The results are analyzed anonymously and the results of the research will not hold any negative consequences for the questioned customers.

1.3 Research questions

Based on the problem statement, the following research questions were formulated for this descriptive study. The main research questions are stated as;

• To what extent is there a difference between the promised service and the expectations and perceptions of the customer experience among the 'Dutch DSL service providers'?

• Is there a correlation between the constructs of customer experience, word-of-mouth, satisfaction and loyalty based on the perceptions of the customer among the 'Dutch DSL service providers'?

To answer the general question, the outcome of the research questions should be combined. The recommendations’ can only be valuable with some statistical prove. Therefore there are

(9)

investigation questions formulated, to make this research empirical. The following

investigative questions are answered to draw a conclusion about the research questions. Some could be answered by the study of available literature, others by surveying or mystery

shopping. These questions are formulated by splitting the research questions in parts into more specific questions to gather data. The questions are used as the foundation for the data collection instrument.

The investigative questions that are answered by literature study can be seen as the descriptive part of this research. With the outcome of these questions is it possible to divide the constructs in parts and makes them measurable. The questions that are answered by literature study were;

• Why is it important to focus on the interaction process for the internet service industry?

• What is customer experience? • What does word-of-mouth mean? • What does satisfaction means? • What drives loyalty?

The questions that are answered by survey or mystery shopping are questions that are based on Servqual (Zeithaml et al, 1985). By dividing the constructs in parts it became possible to measure the expectations and the perceptions. By mystery shopping it became possible to define the reality in details. The questions that are answered by these two methods were;

• What service quality promises are made by the Internet service providers to their customer?

• What are the service quality expectations of the customer? • What are the service quality perceptions of the customer? • How is the service quality experienced by the researcher?

(10)

2. Literature review

2.1 Service delivery process

The service industry in general is characterized with the providing of services. Service should be seen as intangible (Grönroos, 1997). That means that the service quality is highly

perceptive (Berry & Parasuraman, 1991,), the services can not be stocked (Normann, 1991) and has a non-transferable ownership (Normann, 1991). Another way that service is

interpreted is the production and consumption simultaneity (Grönroos, 1990). It contains that the client participates in the service delivery process (Normann, 1991), risk management (Heskett, 1990), that the service cannot be exported (Normann, 1991) and perishability (Fitzsimmons, 1997). Service can be defined as “a set of processing operations carried out by a service provider on behalf of a client, in a medium held by the client, and intended to bring about a change in this medium” (Gadrey, 1992).

Through the almost saturated market of Internet users, the focus on service marketing became more important. As the aspects of service showed, the intangibility of services could become a problem when the customer expects something that could not be delivered. In that way the interaction of the client and the company is essential for providing services (de Jong & van Bemmel, 1992). As the figure shows, the interaction is based on aspects of the client and the competences of the firm.

De Jong & van Bemmel (1992) separated three different forms of interaction;

• Regular sales, the service provider offers the customer a standard product with fewer specifications. A sort of take-it-or-leave-it situation for the customer.

• Sparring, the customer is able to modify the service with certain forks. This happens a lot by mass customization.

• Jobbing, the customer has a high influence on the provided service. Through this specification the service become a tailor-made job.

The interaction process offers the service provider the possibility to build relationships and to register their buyer’s behaviour, the communication behaviour and customers wishes and

(11)

privileges. In the specification process is it possible to fill in the market mediation function. The delivery process stipulates the nature of the service and the communication with the customer. For the intangible services, the results and quality of the delivery process should be visible for the customer with the help of call agents. The quality of tangible services will be presented in tangible products (E. de Vries, 1999).

The moment of truth is based on the promises that are made (Grönroos, 1998). He puts forward the service marketing triangle which states the requirements that are needed to fulfil the customer’s expectations. The importance of this service-orientated triangle is the relation between giving promises and keeping promises. This triangle presents the factors that influence the process consumption and marketing.

The interaction between customers and the

firm takes place in different forms. “Multi-channelling is seen as a coordination problem where by the content of service processes (in terms of communications and products) needs to be made available to different market segments in different phases of the service delivery process through different media” (Anderson, Day & Rangan, 1997). Information media are useful for activities where customers need to understand the difficulty and complex cases or to come to a mutual agreement. Communication tools are useful for daily activities. The more characteristics, like feedback possibilities and language variety, it has, the better the tool is in function. The following enumeration presents the medium with the applicability of the channels;

• Face-to-Face; highest communication possibilities. Synchronises time and place. “The process and capacity is dependent on the human memory. The effects of status and hierarchy can influence the communication process” (Dijk, 1994). It is useful for negotiations (Suh, 1990).

• Ear-to-Ear; “telephone calls are less useful than face-to-face communication, because visualization is not possible and it emphases on the pure context of the arguments” (Suh, 1999). On the other hand, the telephone is priceless for firms. According to Broek (1998) the telephone has high response ratios, possibilities to receive direct

(12)

feedback, it has a high flexibility, it could have combinations with other media and is cheap per contact.

• Sound and vision; video is more useful then audio. Videoconferencing could be a good and a realistic substitute for face-to-face communications. Videoconferencing should be very useful in the future, like also multimedia call centres and vision telephony could be (Suh, 1999).

• Face to Interface; is not that useful for negotiations and decision-making or conflicts and the exchange of confidential information (Bordia, 1997). In contrast, Suh (1999) puts forward that it is useful for negotiations; it could take some time.

• Print; one-way channel, like marketing advertisements. It has a high reach and the costs are low per reached person.

• Radio & television; one-way channel with a high tangibility. Through commercials it is possible to build an identity and to create a brand image.

The scheme of Schmenner (1986) presents the different types of service providing that have a degree of interaction and customization versus the degree of labour intensity. These four different types have a strong relation with multi-channelling. Like in a service shop there will be more face-to-face communication than what you can expect from mass services, which most of the time has a strong

focus on a wide field and is characterized as a one-way channel (Schmenner, 1986).

The following table presents the influence of the different parts of the service delivery process. Among the internet service industry, which has a high focus on pure service, the interactive process is the most important factor (Shaw, 1990). The interactive process is the basis for the customer experience.

(13)

2.2 Customer experience

As the service delivery process showed customers are acting as participants or behave as clients in the service system. This participation is a unique characteristic of the service delivery process. Positive experiences among customers are accomplished in the contact between customers and the firm employees. The employees are responsible for the

operational tasks and the environments and tools which are accessible for the customer. The total experience of the customer is called the customer service experience. This relays how the interaction moment occurs (Bateson, 1995). Customer experience has two different faces. On the one hand, there is the economic view and on the other hand there is the psychological view (Johnson & Fornell, 1991). Customer experience is the internal and subjective response that customers have to any direct or indirect contact with a company or front office. Direct contact occurs in the course of purchase and service and is initiated by the customer. Indirect contact involves unplanned encounters with clients about company’s products, services, or brands. This in the form of word-of-mouth or recommendations like criticisms, advertising campaigns and papers. (Meyer, 2007).

The service quality can be conceptualized as “the degree and direction between consumers’ perceptions and expectations” (Parasuraman et al, 1988) or criticised (Boulding et al, 1993) as “service quality is directly influenced only by perceptions”. Within the marketing literature there is no hard definition of expectations, mostly they appear as service quality or as consumer satisfaction (Brown & Swartz, 1989; Parasuraman et al, 1985). The most accepted conceptualization of expectancy-based service quality is the gap model (Parasuraman et al, 1985). The model shows the type of subtractive disconfirmation approach found in different studies of customer satisfaction (Oliver, 1977; Olson & Donver, 1979). As critics of this model Johnston and Lyth (1991) stated that “if an attribute was not expected then the

performance of that attribute cannot be judged against an evoked expectation”. Service quality can be seen as a result of perceived performance. Because of the subjectivity of perceptions and expectations, this is not easy to accurately measure, since there is not a direct relationship between service and quality (Buttle, 1996).

2.2.1 Servqual-model

Zeithaml, Parasuraman and Berry (1985) developed a model for measuring the quality of a service; they called it the Servqual model. Servqual is a short name for Service Quality and should be interpreted as the framework for measuring the quality of a service. It is useful for

(14)

identifying the shortcomings of a service. By measuring the gaps between the expectations and the perceptions of a service this model is also known as ‘gap analyse’. Following this model, it is permissible to define the service quality as the difference between expectations and perceptions. The difference in quality is visible as the gap between the expectations and the perceptions of the service.

In the first SERVQUAL researches (Zeithaml et al, 1988), which were held in the telecom and credit card industry, they found through factor analysis, ten different dimensions of the quality of service; reliability, responsiveness, competence, courtesy, communication, credibility, security, understanding customer and tangibles. Zeithaml et al (1990) reduced the ten different dimensions to five, as known as RATER;

• Reliability, perform promised service dependably and accurately • Assurance, knowledge, courtesy, trust and confidence

• Tangibles, appearance of physical evidence of the service • Empathy, provision of caring, individualized attention

• Responsiveness, willingness to help and to provide prompt service

Zeithaml et al (1988) identified five different gaps which are determinative for the quality perception of the service. If the perception fulfils the expectation a gap does not exist. Otherwise it should be one of the five gaps which Zeithaml et al defined;

Gap 1; the difference between actual customer expectations and managements’ idea of the

perception of customer expectations. Management does not interpret the wishes of their customers correctly or the customers wishes are not correctly translated as basis for policy and quality.

Gap 2; mismatch between manager expectations of service quality and service quality

specifications. Even when the wishes and demands of the customer are visible, it does not mean that the strategic points of view are the same to improve the quality of service.

Gap 3; poor delivery of service quality. The service is not delivered the first time right. There

exists a gap between the service specifications and the performance. Call agents should be able to fulfil customer needs by having the right knowledge and tools to act well.

Gap 4; difference between the service delivery and the external communication with the

customer. Employees should not promise more than they can deliver. This is a lack in communication between different departments. It would be better if they under promise, but over deliver.

(15)

Gap 5; difference between expected and perceived quality. The customer should not observe

a gap between the expectations and perceptions. It is fine for the customer if the expectations are exceeded, otherwise the company must improve his service.

2.3 Constructs

This study will approach the quality dimensions in a different way. So this research is purely focused on CEES, which stands for Costs, Expectations, Expertise and Speed of help. These are constructs that are based on the Servqual (Zeithaml et al, 1988) model. CEES is used for measuring the gap between promises and expectations, promises and perception and for the gap between expectation and perception. The exploratory part is to find the correlation between the CEES constructs and word-of-mouth, satisfaction and loyalty.

To answer the main research question the CEES constructs will be explained in a way that they are measurable for the gap analysis and to compute the correlation among them. In that way it is also important to explain the meaning of word-of-mouth, satisfaction and loyalty in this research context. But first of all, the expectations and perceptions will be highlighted for the gap analysis, after which the literature study continues with highlighting the other constructs for the correlation analysis.

2.4 Promises

As the service triangle of Grönroos (1998) showed, keeping promises are an important factor for the relationship between companies and their customers. Keeping promises is the basis for creating loyal customers and works as the foundation for a service relationship (Grönroos, 1990). Customers evaluate a firm on how well they delivered on their promises. There are three forms of promises among the customer services;

Personal promises; these promises are made on an individual level, but the results reflect the whole company. Customers evaluate these promises carefully and are aware of the fact that when they do not trust a call agent, they should not trust the company itself. So companies must train their call agents carefully to guarantee that their agents understand the importance and consequences of personal promises. Empty promises and unrealistic timelines are frustrating for the customers. These promises have a negative impact in the trust for the companies and would harm the brand image (Bitner, 1995).

Organizational promises; these promises are made by the organization itself to the customers. These promises include statements about service, product quality, terms and conditions etc. These promises can be published by websites, advertisement campaigns or other marketing activities like commercials with a certain slogan in it (Bitner, 1995). Like

(16)

with the slogan of Tele2 `why pay more`, they are claiming that they are the cheapest with offering the total service package.

Expected promises; these promises are made by the industry. Like what is expected from providers in a specific industry. Customers are expecting something even when it is not stated. Like for the Internet service, customers are expecting to surf on the World Wide Web when they ordered a connection. Some companies are acting harder then they are promising. They like to create a competitive advantage with their extra service (Bartlein, 2006). The OPTA is the controller of the telecom sector, who defines what is allowed and what is not according to the given promises.

To create a great customer service, companies should remember the rule of “under promising and over delivering”. If the customer precepts more than they expect, they will be loyal to the company (Newsom, 2005).

2.5 Expectations

Expectations have a central role in measuring service quality. There is not a hard definition for what expectations are and what they really do. Most of the time expectations relate to service quality or satisfaction (Brown & Swartz, 1989). The two-sides of this construct extend to their role in the evaluation process. According to Parasuraman (1988) expectation should be conceptualized as “the degree and direction between consumers’ perceptions and

expectations”. Boulding (1993) disagreed with this citation and stated that “service quality is directly influenced only by perceptions”.

According to Boulding (1993) there is the belief that a “person’s expectations colour the way he or she perceives reality”. Two different expectations can be identified, the “should” and “will” for measuring the service quality. The predictability of the customer about expectations works as positive on reality. If the perception increases by front offices. Commenting on the “will” and should” theory, the general idea is that it influences

perceptions and that service quality can only be evaluated subjectively. The evaluation of the service is not that easy for customers. It can only be done after consumption of the service, so pre-purchase evaluation can’t be done (Cowel, 1984). Through this effect, providers should also be handling with past experience of customers. Loyal customers’ expectations increase (Tellis, 1988). But they will also place more reliance on the promises from internal sources. Foulkes (1994) states that there is a reliance on an internal source, “the consumers’ memory has a considerable influence on the formation of expectancies”. Hasher (1984) presented the results that the encode information increases when the frequency of encoding increases. By

(17)

implementing this information about frequencies of the customers, it becomes possible for the providers to identify and to analyse the difference between the “should” and “will”

expectations (Underwood, 1971). This is valuable information for providers to manage the customer experience.

2.6 Perceptions

According to Bowen (1992) customer perceptions are influenced by the attitudes and behaviours of the call agents when evaluating the delivered quality of service. Call centre agents are customers’ first point of connection with a service provider. The agents are responsible for selling the service of the provider, keeping the promises offered by the provider and have to create a positive image (Bettencourt, 1997). To create a competitive advantage, providers should see their call agents as more important than the service itself (Pfeffer, 1994). To full fill customers desires, call agents need to know what the customers needs to help them right and effectively. By face-to-face encounters is it possible to create quality perceptions using physical characteristics of the environment. By telephone encounters is it only possible for the consumer to evaluate the quality of service on their perceptions of how the call agent trades (Burgers, 2000). Currently, all service providers own a call centre for managing the customers’ service operations. To understand customers’ expectations in voice-to-voice encounters’, the encounters became much more important.

The perceptions of constructs like price, quality and added value are variables of consumer behaviour(Bishop, 1983). Quality and value are almost equal to each other and form the same constructs like perceived utility and perceived worth (Zeithaml, 1988). Zeithaml (1988) defined perceived quality as “the customers’ judgment about product’s overall excellence or superiority”. According to Zeithaml (1988) perceived quality is “different from the actual quality, it has a higher abstraction level and is a judgment made within a consumers’ evoked set”. From the consumers’ point of view the perception of price can be defined as “what is given up or sacrificed to obtain a product” (Ahtola, 1984).

According to Zeithaml (1988) the interpreted price can be seen as the price encoded by the consumer. According to research of Zeithaml (1988), perceived value is constructed out of four consumer definitions; value is a low price, value is interpreted as what ever I want in a

H1: Customers of Internet service providers are expecting more then providers are

(18)

product, value is the quality I get for the price that I pay and value can be defined as value is what I get for what I give.

2.7 Costs

The costs of the service are dependent on the focus of the front offices. There is a high amount of service centres that have a strong focus on their sales service. With these centres the sales number is usually a free number and the service number is usually an expensive number. According to research by Keaveney (1995) costs are strong factor that has a high influence on the switching behaviour. Costs are encoded by the customers (Zeithaml, 1988). What is seen by one as cheap is seen as expensive by another. The difference between the costs of sales and service numbers originates from the idea that complaining customers are an expense for providers. But without complaining customers, providers do not receive feedback of their quality of service. So providers must see complaining customers’ as valuable, because that is an honest way to receive “free” feedback in the lack of service providing. A familiar rule among call centre is the 80-20 rule, 80 percent of the incoming calls should be answered within 20 seconds. To reduce the costs for the valuable customers call centres should invest in more call agents who can handle more calls. An expensive service number could not only be frustrated for some customers; it could also work negatively for the brand image. That was one of the main reasons that KPN decided for their internet services to invest 20 million in their call centre (Coopmans, 2007).

2.8 Experience

In this study, experience is approached with six different variables which the customer is facing when they are dealing with front offices;

2.8.1 Staff attitude

According to Boshoff (1999) call agents should be friendly in conversation and should be aware of the fact that they are communicating with customers. Aspects like personal problems are influencing the attitude of the call agents; this could lead to a decrease of the quality of service. One problem of a front office is the large amount of incoming calls. For instance, an

H2: Internet service providers are not keeping their promises that they made to their

customers.

(19)

agent who is dealing with a very angry customer does not have a great deal of time to “cool down” because the next customer is already ringing. This could leave a negative influence on the call that follows up. The customer will always recognize and remember an unwilling agent. “Willingness to help, involves timeliness of service by mailing immediately, calling the customer back and giving prompt service” (Zeithaml, 1985).

2.8.2 To busy to respond

Boshoff (1999) stated that “the speed with which the customer complaint or wish is resolved”. According to Wilson (1995) is time an important factor that could influence the relationship between buyers and sellers. Customers should not feel pressured, but also should not be held waiting unnecessary longer, what most of the time happens in a clothing store. Anton (1997) believed that time is an important variable for measuring performance. Variables for

measuring the call centre performance are in this study talk time and hold time.

2.8.3 Reliability

Reliability is based on delivering promises dependably and accurately (Zeithaml & Bitner, 2000). Trustful call agents are expected to keep their promises and should also be reliable (Wilson, 1995). Trust can be seen as the foundation for a long term relationship (Crosby et al, 1990). Especially for Internet call centres, trust is seen as the critical variable of customer expectations of the call agents’ behaviour. The risk that call agents are facing is that

customers are judging their reliability only by their word and not by expression or attitude. By e-mail or fax services the customer is not dependent on the reliability of the call agent,

because there is not only verbal contact (Burgers, 2000). For front offices reliability means; “accuracy in administration, keeping records correctly and performing the service at the designated time” (Zeithaml et al, 1985).

2.8.4 Courtesy

The style of communication stands for the handling of the call agents in which the customer is addressing the service needs (Boshoff, 1999). Boshoff (1999) poses two types of

communication styles; the convergence style and maintenance style. The convergence style is a communication style in which the call agents adapt the tone of voice in the conversations held with the customers. In the maintenance style no effort is made to adapt to the customer needs. Crosby (1990) found that there is a similarity in the customer perception of the sales persons. This should also be happening in the voice-to-voice service context in which

(20)

founding the call agent should be able to adjust the language to different customers (Parasuraman et al, 1985). For call agents it basically means that they should consider consumers’ property and that they should have a clean and neat appearance in public contact with customers (Zeithaml et al, 1985).

2.8.5 Empathy

Parasuraman (1988) stated that empathy is a characteristic defined as “the ability to provide the customer with caring, individualized attention”. The call agents are responsible for the customers’ personal involvement. Call agents should give the customers the feeling that they care about them. To create a more personal atmosphere, call agents should focus on the communication of responsiveness and assurance (Bearden, 1998). The voice-to-voice front offices have a high level of contacts. In that way, empathy is expected as an important attribute by these contacts. According to Bearden (1998) is the expected empathy less important than less personally involving contacts, so they should found the balance in that.

2.8.6 Understanding the customer

Understanding the customers’ problem, means that the call agents should make the effort to understand the specific wishes of the customers. The call agent should be able to learn how to read customers for identifying what they require and what they expect form the call agent. Call agents must listen to the customers closely and are responsible that the customer feels like an individual (Parasuraman, 1985). The current available technology in front offices should be used to extract information about the customer. Call agents should log their calls in databases, in that way providers collect essential historical customer information. For

knowing the customer, a database alone is not enough to achieve that target. Like the information about current needs, which will not be registered in the database yet. Burgers (2000) are stating that it is important that call agents can listen and that they are able to retrieve information of the customers’ desires. Understanding the customer could be summarized as; “learning the specific requirements, providing individual attention and recognizing the regular customers” (Zeithaml et al, 1985).

H4: Customers have the highest expectations on the reliability of call agents from Internet

(21)

2.9 Expertise

Call agents must develop the skills and knowledge necessary to perform the service delivery (Zeithaml & Bitner, 2000). Call agents must have the required knowledge about the content of the service description, but they should also be able to perform the multiple activities that are needed to execute the job. Skills that are needed are a pleasant telephone voice and the ability to perform more activities at the same time, like listening or searching for information. The knowledge level can be reached by following some courses. However, personal trades are innate and can not be attained through training. When expectations are not met the first time, customers will call for an explanation. Call agents should know the level of knowledge and skills of their operational support personnel.

Froehle (2004) defined three task-orientated call agents’ characteristics which are required for a good communication relationship with the customers;

Knowledgeableness; call agents should be trained and skilled with the information that is needed to execute their function (George, 1991). These call agents must own the skills to ensure that incident are solved well and other operational support will be performed correctly. According to Darian et al (2001) highly skilled call agents produce a higher predicting

performance. Knowledgeable call agents are valuable for companies, “knowledge is power” (Kettinger & Lee, 2005).

Preparedness; an informed call agent could serve the customers better with their problems. Some preparation could address to a better understanding of customer desires. By using some CRM systems, call agents could prepare themselves better. Even the knowledge about how to gather the needed data is required. A reliable customer database is very valuable for the front office; this could be seen as a good investment (Reichheld & Schefter, 2002). To ensure that call agents are prepared to perform their tasks well, they must follow courses to use the CRM system (Berry & Parasuraman, 1997). The call agents must have access and authorization to the right tools and applications to deliver right assistance to their customers. Preparation is an important factor for front offices, because it could reduce the expensive conversation times (Parasuraman et al, 1988).

Thoroughness; this concept includes the desire to ensure that a incident is solved well and that the supporting tasks are completely performed. Asking the customer questions about their issues is a sign of positive communication style for the customer service behaviour (George, 1991). Thoroughness could reinforce the different factors of the service quality of reliability and assurance (Parasuraman et al, 1988). The call agents must be able to identify the problem and must take care of the problem so that it does not occur again. Call agents

(22)

with characteristics of thoroughness own operational and service quality skills and have a positive word-of-mouth for the service provider.

When the service is not delivered right the first time, customers will want explanations. It is a critical satisfactory service encounter incident when the response of the customer is a

complaint (Bitner, 1990). When customers call in with an official complaint, they are trying to find the truth about what happened and why it occurred. The content of the explanation of the call agent has a high impact on the satisfaction of the customer and the issue should not happen again (Bitner, 1994). Following Boshoff (1999), customers will be more satisfied and happy when they receive a clear and honest explanation. It is the responsibility of the call agents to transfer the complaining customer into a satisfied customer. So it is important that call agents are trained on these skills and develop them further and those they have access to information to provide a clear explanation to their customers (Burgers, 2000).

2.10 Speed of help

Waiting is a regular and frustrating process in our lives. It also happens in call centres; more than 70% of service customers are dealing with waiting time (Jones & Dent, 1994). Waiting is one of those daily occurrences that cause frustration and stress. It could happen that customers do not purchase some valuable service because the waiting line is discouraging them, like what most of the time happens by large stores as media markt of the V & D. Studies proved that waiting time is in strong relation with dissatisfaction. The quality of a service will be interpreted lower, when the expected waiting does not fit with the perceived waiting time (Chebat, 1993). Graham (1981) found the relation between waiting time and the loyalty of the customer. Graham (1981) stated that “waiting time directly affects the mood of the customer, which can increase or decrease customer loyalty”.

The perceived waiting time has a strong correlation to reality. Customers compare their waiting time with what they expected and what the actual waiting time was. The differences influence customer satisfaction positively or negatively depending on how long they wait or stand in the queue. To grade the performance of a service, the expectations of the customers should be visible and should be seen as ideal (Miller, 1977). According to Zeithaml

H5: There is a strong and significant positive correlation between the experience and

(23)

(1993) et al. is it possible to categorize the expectations that customers have among waiting time. The first category is the desired and anticipated level, and the other is the adequate and tolerable level. The desired and anticipated level of the expected waiting time stands for; Customers will not wait longer if it does not fit with their expected waiting time. The

adequate and tolerable level could be seen as the minimum expectation of waiting time that is acceptable for the customer.

But there is also the perception issue. Psychologically seen, there is significant evidence to state that customers believe that they waited longer than they really did (Jones, 1995). Waiting time is defined as passively passing time, while interaction is noticed as actively passing time. The perceptions of these two durations are different (Hornik, 1984). It is stated that passive duration is overestimated and active duration is underestimated. The increase of waiting time will lead directly to a decrease of satisfaction. Katz et al (1991) conclude out of their research that consumers tend to overestimate the amount of waiting time by an average of 1 min. They empirically tested that an overall waiting time of 5 minutes is reasonable

Additionally there will be a new construct introduction, called C-Sat. C-sat stands for

Customer Service Acid Test ratio. Some service providers have separated phone numbers, one for sales and one for service. Whereby there is a differentiation in waiting time or speaking directly to a call agent or facing an IVR system. C-Sat is a new ratio that is computable by dividing the average waiting time of a sales number by the average waiting time of a service number ( rvice tingtimese Averegewai les tingtimesa Averegewai

). The closer this ratio is to one, the higher the service-orientation of the provider. Providers with a ratio above one are more service-oriented. The ideal situation is that providers have a C-sat ratio equal to one (Güngör, 2007).

H6: There is a weak significant correlation between the speed of help and the C-sat ratio

among the encounter services of Internet service providers.

H7: Internet service providers are more focused on the sales point of view in comparison

(24)

2.11 Satisfaction

Nowadays Internet service providers understand that customer satisfaction is important and that it adds value to the transaction (Marsico, 1996). To satisfy the customer call agents are working much harder to fulfil the customers’ needs. Internet service providers are hardly trying to build long-term relationships with their customers (Prabhakar , 1997). Focussing on customer satisfaction is a marketing strategy that Internet service providers are implementing to differentiate from each other. Call agents are following the theory that a “satisfied customer is more likely to repurchase, purchase more, and promote positive word-of-mouth” (Walker, 1995). The level of satisfaction is dependent on the interpretation of psychological and physical variables. This level could also be dependent on past experiences and the comparison with other services of the concurrent. Berry (1991) defined ten “quality values” on which the customer satisfaction behaviour is dependent; quality, value, efficiency, timeliness, ease of access, environment, inter-departmental teamwork, front line service behaviours, commitment to the customer and innovation. Parasuraman et al (1988) stated that customer satisfaction is measurable by the gap between customers’ expectation of performance and their perceived experience of performance.

2.12 Word-of-mouth

Word-of-mouth is defined as “communication that occur when consumer share information with friends about product, services or promotions” (Peter & Olson, 2005). It is directly related with informing friends about an event that has created a certain level of satisfaction (Söderland, 1998). Fiske et al (1990) found the asymmetrical effects of positive and negative feelings. Their research claims that under certain conditions, negative events are producing a stronger response than positive events. Taylor (1991) explains this finding by his theory that negative emotions signal that action needs to be taken, by positive emotions it won’t. The less satisfied we are, the more we are talking about negative feelings (Peeters & Czapinski). In contradictory of this pose Johnston (1995) showed that social factors bias the individual towards spreading positive rather than negative information. People won’t like to talk about their dissatisfaction by a service provider, because they are afraid that others form a negative opinion about them, they should seek better in the first place (Peterson & Wilson, 1992). So there are several arguments which suggest that satisfied customers are more likely to communicate about their experiences than dissatisfied ones.

Woodruff et al (1983) found a relation between the expectation prior to a purchase and perceived performance after a purchase matched. If everything occurs as expected, there will

(25)

be less reaction on that. But when things occur not as expected, it will result in a high level of satisfaction or high level of dissatisfaction. This could be influencing someone’s switching behaviour (Keaveney, 1995). “Word-of-mouth communication could spread awareness beyond those consumers who come into direct contact with the services” (Bayus, 1985). It is a powerful and cheap form of communication that could work in a positive but also in a

negative way (Peeters & Olson, 2005).

2.13 Loyalty

By a high level of satisfaction consumers will be loyal to their current Internet Service

Provider. A loyal customer could influence the renewal of service contracts and could create a change in the buying behaviour of the customer; this could lead to a positive word-of-mouth (Hirschman, 1970). Internet providers receive feedback of their customers in two different ways, the exit way, also known as switching behavior, or the voice way. Exit implies that the customer won’t continue his contract anymore, while voice is customer complaints expressing the dissatisfaction directly to the company. The leaving customers have a strong impact on the long-term revenue. To keep customers firms should be aware of the fact that customers won’t switch if there are high switching barriers of no competitive alternatives (Reichel & Sasser, 1990). The Internet industry is facing a high barrier for switching to another provider. According to the complain database of the OPTA consumers recognize three major issue during the switching process;

• There is a lack of support. The passive support isn’t supporting enough and the active support that is expected will not be delivered to the consumer.

• There is an administrative mess and a lack of technical knowledge among the service providers helpdesk.

• The consumers facing always a period with double contracts, because the contract of the leaving provider isn’t ended yet.

Customers may also be loyal when they are satisfied. Satisfied customers will stay longer by the same firm and this could lead to a stronger relationship (Coyne, 1989). Zeithaml et al (1996) stated that there is a “zone of tolerance” that captures the range of service within a provider faces the customer expectations. The lower bound of this zone could be seen as

H8: There is a strong significant positive correlation between word-of-mouth and

(26)

“adequate” and the upper bound as “desired” service. The relationship between the service and loyalty changes depending on the position of the customer in the zone of tolerance. Through the different views of loyalty should loyalty be stated as “the intention to purchase again from the Internet service provider who has produced a certain level of satisfaction”.

Fournier (1998) argued that to much loyal customers is not that good as service providers’ actual beliefs. At a certain point they will cost more than they produce. This is explainable by the relationship theory. How longer the customer is loyal to their provider, how more they expect of their provider. This leads to an increasing service competition against itself for the providers (Oh, 1987). The relationship is build up out three factors, the psychological, the socio-cultural and the relational. The psychological relationship could be specified in three different views; it could be seen as a daily activity (Csikszentmihalyi & Beattie, 1979), it has an added value for life projects (Caspi, 1987) and it could be seen as a daily activity (Klinger, 1987). According to the socio-cultural relationship there are five attitudes and behaviours; age, life cycle, gender, social network and culture (Dion, 1996). These factors influencing the strength of the relationship with which relationships are initiated and terminated, but is has also a certain level of commitment (Thompson, 1996). Park (1991) poses that the context for an individual relationship is the most important. Through other relationships are consumers able to compare their relationship with the Internet service provider with other service providers. Customers can marry a service, but at certain level they will divorce of it, because the current service doesn’t fit anymore in the factors that build relationships (Peter & Olson, 2005).

H9: There is a strong significant correlation between satisfaction and loyalty among the

(27)

2.14 Conceptual model

The literature study for this research could be conceptualized in the following model;

Whereby promises, expectations and perceptions are based on the CEES factors and customer experience is created by a small or positive gap between promises and expectations, promises and perceptions and expectations and perceptions. The constructs are interpreted as follows;

Promises; “psychological contract indicating a transaction between two persons whereby the

first person undertakes in the future to render some service or gift to the second person or devotes something valuable now and here to his use” (Bitner, 1995).

Expectations; “the act or state of expecting or looking forward to an event that has about to

happen” (Zeithaml et al, 1993).

Perceptions; “the customers’ judgment about product’s overall excellence or superiority”

(Parasuraman et al, 1988).

Customer experience; “the internal and subjective response that customers have to any direct

or indirect contact with a company” (Meyer, 2007).

Word-of-mouth; “communication that occurs when consumers share information with

friends about products, services or promotions” (Peter & Olson, 2005).

Satisfaction; “the provision of goods or services which fulfils the customer’s expectations in

terms of quality and service, in relation to price paid” (Reichel & Sasser, 1990).

Loyalty; “the intention to purchase again from the Internet service provider who has

(28)

3. Methodology

3.1 Sample design

This research is focused on five Internet service providers. These are Scarlet, Orange, XS4ALL, Tiscali and Tele2. These providers are selected because they are using a different copper infrastructure and are not dependent on a wholesaler. Together they are responsible for more than a 50% market share of DSL users. For collecting the promises made by providers, interviewing one manager of the customer services of each provider was sufficient. The results were compared with the analyses results of the website and annual report (2006) of the providers. To gather the expectations of the different DSL users, a survey was held among one hundred respondents. 20 respondents per provider completed the survey. To receive information perceptions of the respondents, the survey was also constructed with perception questions. There were no limitations on age or gender. In total the survey was filled out by 73 men and 27 women, the average age was 28 years. The one hundred responses were

complemented with one hundred results of mystery shopping. The mystery shopping was done by three persons to guarantee the reliability of the interpretation of the perceptions. Otherwise it would be subjective. To compute the C-sat ratio, there were ten calls made to the sales number and ten to the service number for each provider. To filter out the time

differences, these calls were held in the morning and in the afternoon over five working days.

3.2 Methods

Tracking customer experience is divided in three different groups with their own aspects (Meyer, 2007). The opportunities for tracking customer experience are tracking past patterns, present patterns or potential patterns. The empirical part of the research is done according to the methodology of tracking present patterns, this tracks current relationships and experience issues with an eye toward identifying future opportunities. It collects the data for this user experience study through surveys and direct contact with front offices. This last form is called mystery shopping, which is introduced for testing customer experiences.

To collect the data that is needed for answering the research questions this study tested the CEES constructs with a universal form (appendices 1-3). This form is based on the

questionnaire of the Servqual model (Zeithaml et al, 1988). The outcomes of interviewing, surveying and mystery shopping are comparable to each other. The interviews (appendix 1) are used for collecting the promises made. Via a questionnaire providers gave their promises

(29)

scaled in a way that it was useful for this research. The survey (appendix 2) was an online survey whereby participants first gave their expectations and later their perceptions among their service provider. The mystery shopping (appendix 3) technique was also used for measuring the perceptions.

The questionnaire was constructed via CEES. The costs were measured by the total time of calling the service times the costs of the service number. The experiences were measured by asking six scaled questions related to experience and the average of these questions stand for the figure of experience. Expertise was measured by asking two scaled questions over this construct and the average of these outcomes stands for expertise. From the mystery calls, expertise is computed by the average of the answers on five medium high level questions. Speed of help is computed by variables, the first one was timing the time that a call starts till the time an actual person answered that call and the second one was, sending e-mails out and see how long it took before the responds was received. The average of these scaled variables stands for speed of help.

Word-of-mouth, satisfaction and loyalty are only measured by surveying. By a single question of each construct the overview of the participants could be evaluated. The C-sat ratio is measured by calling to a service number and calling to a sales number and dividing the average waiting time from sales to the average waiting time of service. The total waiting time is measured with the aid of a stopwatch to collect significant data measured in seconds. All this data is scaled and imported in the software program SPSS for computing gaps and correlations.

3.3 Data analysis

The collected data of all constructs is analysed in a way that answering the research questions became possible. Gap 1 and gap 2 are computed by the independent-samples T test. The

Independent-Samples T Test procedure compares means for two groups of cases. Ideally, for this test, the subjects should be randomly assigned to two groups, so that any difference in response is due to the treatment (or lack of treatment) and not to other factors. Gap 3 is computed by the paired sample test in SPSS. The Paired-Samples T Test procedure compares the means of two constructs for a single group. It computes the differences between values of the two constructs for each case and tests whether the average differs from zero. This test is applied for the mean of each variable. The assumption is; the test requires that the

observations for each pair are made under the same conditions and the mean differences are normally distributed. Gap 1 is computed by the constructs of Promises – the constructs of

(30)

Expectations. Gap 2 is computed by the constructs of Promises – the constructs of Perceptions and Gap 3 is computed by the constructs of Expectations – the constructs of Perceptions. These analyses are first done for all the five service providers in total, and further on, they are specified on a single provider level.

To find out if there are significant correlations between the constructs to answering the second research question, the bivariate correlations procedure is followed. This procedure measured how variables or rank orders are related. The spearman rank correlation, which is used for this research, is a measure of association between two constructs. Constructs with ordered categories are applied for using the spearman rank correlations. Correlations coefficients ranges in a value from -1 (a perfect negative relationship) and +1 (a perfect positive relationship). A value of zero indicates no linear relationship. Because of the unknown direction of the correlations, the test is performed with the probabilities of a two tailed test. The correlation coefficients are significant at the 0.05 level, these are identified with a single asterisk, and those significant at the 0.01 level are identified with two asterisks. The correlations are first computed for all the five providers in total and followed up by a specified level of a single provider. The correlations were computed based on the data that was collected for the perceptions.

3.4 Resources

To conduct this research some resources are needed. This research made use of an online survey tool, called “www.thesistools.com”. This tool made it possible to set the questionnaire online. It was much easier to collect the data and to find respondents who were willing to participate in this research. Also the use of a stopwatch for measuring the time related questions was important to have. Without a telephone this whole research would not have been possible. All mystery calls were done by mobile phone. The storage of the data is done with the use of excel. The gap analysis and correlation tests were computed by SPSS 15.

(31)

4.

Findings

The results of the gap analysis and correlation analysis are bundled per provider and are attached in the appendices (appendix 4).

4.1 Promises

For analysing the data found, the Cronbach’s alpha is computed for presenting the reliability of the items which form the constructs. Cronbach’s alpha is a measurement method to see what the reliability is of a survey with a multi-dimensional structure. A value closer than 1 stands for a high reliability of the items. Cronbach’s alpha is stated as the function of a couple of test items and the average intern-correlation between the items. The constructs experience, expertise and speed of help are computed by the average of the value of the concepts. Based on the findings by analysing websites, reading term and conditions, questioning sales representatives and interviewing some public relation officers, the following promises could be presented of the Internet Service Providers (N is equal to 5);

The table above is created by different kind of sources which provided the information about the promises. This study does not make a difference on the sources and as assumption can be stated that the different sources are promising all the same.

Based on the results above, it is remarkable to mention that Scarlet is a typical service provider with less promises. The other providers are almost equal to each other with their promises. The computed mean should be interpreted as an overall figure for the Internet service industry, which is also used for the gap analyses later on. The means are around the average of four on a seven point Likert-scale. Whereby providers are reserved by promising the items related to speed of help. The costs are computed on a 10 point Likert-scale, so it is also near the average, like the other constructs.

(32)

By interpreting the Cronbach’s alpha it can be stated that the questions about experience are closely related to each other and that the output is very reliable. For the other constructs there is less reliability. Remarkable is the fact that providers make some differentiation on the experience items by grading them, while the literature stated that they should be used as equal. Only the providers Orange and Tiscali are doing it well.

4.2 Expectations

With help of an online survey tool the expectations of customers about their Internet providers was collected. 100 participants took the opportunity to fill in their expectations of their

internet service providers (N is equal to 20 per provider). Based on these expectations Internet service providers could match if their promises are fulfilling the expectations of their

customers. The following data was generated;

Based on these findings, customers of internet service providers are expecting a lot from the service of their providers. Every construct scores above the average. The Cronbach’s alpha of the constructs Experience and Expertise are above 0.8, which means that this data is very reliable. The data of speed of help is less reliable, but is useful for analysing gap 1 and gap 3 of the conceptual model.

Customers of Orange and Tiscali do not worry about the cost of the service. They graded it below the average. For the other constructs it is allowed to say that the customer expectations of the different providers are almost equal to each other. Customers prefer a quicker response by e-mail than a shorter waiting time when they are on the phone.

H4: Staff attitude and courtesy are the most important items whereby customers have

(33)

The chart bar shows in detail that there is a little difference between Staff attitude (5.84) and Courtesy (5.74). It obviously shows that customers do not expect much of the

reliability (2.8) and the willingness to help (3.65) of their service provider.

4.3 Perceptions

The data for the perceptions were collected by two independent methods; a customer survey and via mystery shopping. For both is N equal to 100, so in total there is data generated of 200 perceptions. These data is useful for measuring Gap 2 and Gap 3 of the conceptual model. The collected data is as follows;

The perceived costs are just above the average. Tiscali could be seen as a provider with an expensive service and Scarlet, Tele2 and XS4ALL are offering a cheaper service. The perception of the experience is almost the same for the five providers. The construct experience has a Cronbach’s alpha of 0.873; this data is very reliable and useful. The found perceptions of expertise are also almost equal to each other. Customers of Tele2 are talking to call agents with less knowledge in comparison with the other providers. Expertise has a Cronbach’s alpha of 0.756, this is not a very strong coefficient, but for this study it is useful enough. Speed of help has a range of one point. XS4ALL scores the lowest and Scarlet, Tele2 and Tiscali are handling with the same speed. The Cronbach’s alpha is very low (0.01), but for the gap analyses this data should be comparable to the expectations and the promises.

4.3.1 Mystery shopping vs. Survey

To see if there were differences by the methods used, surveying and mystery shopping, the following section will analyse the results of both. By using the independent t-test it is allowed

(34)

to compare the data of two different sample groups. Both methods have a N equal to 100. The results were as follows;

The Cronbach’s alpha of the experience data is more reliable than the survey data, but on the other hand, the coefficient is higher for the expertise and speed of help results of mystery shopping. So to generate more reliable data, a combination between surveying and mystery shopping is applicable. It is also remarkable that the survey has a modus of four and that mystery shopping has a modus of five. A similarity between surveying and mystery shopping is that the construct costs have the highest average

By using the independent sample test, it is significantly allowed to put forward that there is a mean difference with a maximum range op 0.79 (speed of help) between the survey and mystery shopping results. Overall it is seen that the method of mystery shopping is

(35)

experience the nearest to the perceptions of experience by the customers. The data does not differ much from each other, what should be interpreted as positive. As explanation for the higher results by mystery shopping could be given that there was not a real irritation or need for help when calling the front offices. It was more for collecting data for this study. But the method of surveying and mystery shopping was useful for collecting the promises.

4.4 Gap analysis

To say something about the gaps between promises and expectations, promises and

perceptions and expectations and perceptions, the gaps were tested following the procedures of the “independent-samples t-test” and the “paired sample t-test”. These t-tests (Snedecor & Cochran, 1989) are used to stipulate if two population means are equal to each other. These tests make it able to determine if there is a gap between two constructs. The results should be interpreted as; the closer the mean is to zero, the smaller the gap really is.

4.4.1 Gap1

The findings of the gap between promises and expectations are presented below:

Based on the results, there is enough significant evidence to pose that there is an average gap between expectations and promises of 0.9243. This means that customers of Internet service providers’ expect more than Internet providers are promising. The largest gap is computed for the construct speed of help. The promised speed of help does not fit the expectations of their customers. The gap between the expected costs and the promised costs must be interpreted as negative for the Internet providers. Customers expect to pay less for a service than providers are really promising.

Referenties

GERELATEERDE DOCUMENTEN

Predictors: (Constant), Inkomenssituatie, Supermarkttrouw (0-10), Winkel, CHOCTASTE, PENPICK, Dag, Boodschapper, Aantal personen in huishouden, NULCONDITIE, Frequentie

By comparing the standardized beta coefficients of the dummy variable for the highest quality ratings (excellent (5)) of all three models, we can compare the different

When performing a regression analysis on the whole model, with all the factors and their components and the moderating effects included, only responsiveness of

Twice a year and no more. Because if you send it more often it feels for the customer that he has to invest time in it. Besides, sending it more often results in sending

• Provides insights into the effect of customer satisfaction, measured through online product reviews, on repurchase behavior!. • Adresses the question whether the reasons for

Besides investigating the overall effect of the five different customer experience dimensions (cognitive, emotional, sensorial, social, and behavioural) on customer loyalty, I

impact of average satisfaction levels during prior experiences on the current overall customer experience is mediated by the level of pre-purchase satisfaction. H4 Customers

From this research, it became clear that it pays-off to investigate the effect of CFMs in different industries since the effect of the customer experience on sales growth rate