• No results found

A research into the influence of customer surveys and service calls on long term customer loyalty Increasing Customer Loyalty: a Continuous Process

N/A
N/A
Protected

Academic year: 2021

Share "A research into the influence of customer surveys and service calls on long term customer loyalty Increasing Customer Loyalty: a Continuous Process"

Copied!
66
0
0

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

Hele tekst

(1)

Increasing Customer Loyalty: a Continuous Process

A research into the influence of customer surveys and service calls

on long term customer loyalty

(2)

Increasing Customer Loyalty: a Continuous Process

A research into the influence of customer surveys and service calls on

long term customer loyalty

N.J.

R

EITH

University:

Rijksuniversiteit Groningen

Faculty:

Faculty of Economics and Business

MSc:

MSc Business Administration

Profile:

Marketing Research & Management

Date:

April 2013

Supervisor RUG:

Dr. J.E.M. van Nierop

Second reader:

Dr. H. Risselada

Address:

Lange Wateringkade 35A

2295 RP Kwintsheul

E-mail:

n.j.reith@student.rug.nl

Telephone:

+31615153090

(3)

MANAGEMENT SUMMARY

The present research was initiated by an online market research agency to determine the effect of handling customer complaints via a phone call on the likelihood of customers to recommend the company to others. In other words, what the effect of complaint handling is on customer loyalty. When a customer is attached to a brand or company and repeatedly uses it, this is called customer loyalty. Attitudinal loyalty, measured via the recommendation likelihood, is an often used term when it comes to loyalty. Long-term customer relationships are more beneficial for companies than short-term relationships. Therefore, managing long-term relationships is important. This can for instance be done via feedback research, in which customers are asked about their experiences with the company. This can be done via an online survey. The insights obtained via the survey are used to maintain and improve customer satisfaction and to increase customer loyalty. In some cases generating the customer’s experiences does not end after an online survey about a contact moment. Companies can decide to also make a service call followed by a follow-up survey. The follow-up survey uses the same measurement method as the first survey and the outcomes can be used to measure the change in attitude. The service call is made to provide the customer the opportunity to share complaints with the company. This way, the customer might become more satisfied with the company and how it treats his customers which might lead to a higher likelihood to recommend the company to others. Improved service quality will lead to more satisfied customers and thus an increase in customer loyalty.

In the current research it was analyzed whether or not complaint handling via a survey and phone call increases the likelihood of customers to recommend the company to others. The realization of the aim is done via quantitative research: two loyalty researches were conducted with a period of six months between the two measurements. Besides, customers have been evaluating contact moments with the company via an online customer survey. The study is divided into two parts: part I of the paper focuses on the effect of extra attention or complaint handling directly after a contact moment with the company on customer loyalty. This is measured via an online customer survey. Part II focuses on the effects of follow-up research: contact the customer again via telephone to provide the opportunity to share complaints with the company. This telephone call is followed by a second online survey in which the recommendation likelihood is asked again. Based on the two surveys a company is able to see the change in attitude thanks to the follow-up service call.

(4)

are clear and assume that it contributes to higher customer loyalty, in the current research the effect of complaint handling via surveying customers after a contact moment on customer loyalty cannot significantly be proven. The results may be insignificant because only a limited amount of respondents were available for the research; this hampers the significance. The limited amount of respondents available for the research is due to the fact that a respondent for this study should have participated in the two loyalty researches in May and in November, and in the first research (customer survey). In the initial dataset most respondents participated only in one of the researches. Only 35 respondents were left that did participate in both loyalty researches and in a customer survey. These respondents were used to determine the effect of on customer loyalty. Although not significant, the direction of the beta suggests positive results of feedback research on customer loyalty. Further research is required to provide significant prove for the effect.

Most important is the effect of follow-up research, in which the customers received a service call on the occasion of a low recommendation likelihood score in the first survey. Follow-up research is proven to have a positive effect on customer loyalty. People are more likely to recommend the company when they have participated in follow-up research after the first survey, which might be due to the extra attention they have received and the handling of complaints via the service call. For people who are not very likely to recommend the company to others follow-up research can be a solution because thanks to the service call they might also become more likely to make recommendations about the company. In the end this leads to an increase in customer loyalty.

The most important limitation for this research concerns the limited amount of respondents. Therefore, in future research a company should make sure it takes enough time for the research. This allows for multiple researches without customers becoming tired of filling in surveys. Future research could also be focused on the reasons why people become more likely to recommend the company. But most important the research should contain a large amount of respondents so that the significance of the results is not hampered by the limited amount of people.

(5)

PREFACE

After almost six years of studying and enjoying the student life, the time has arrived to make these great years a memory. My master thesis, in which I put a lot of effort, is ready and now lies in front of you. In this thesis I have summarized as much as possible from what I have learned during my studies. I am now ready to finish studying and move on to the next stage in my life.

I started at The Hague University in September 2007 with the bachelor European Studies. After three years of hard working I graduated and decided to continue studying. I applied for the pre-MSc program Business Administration at the University of Groningen. The program was a real challenge and I was glad I made it within a year. Now I had to choose a master program, which was not a hard decision. I started the master Marketing Management & Marketing Research. This was one of the best decisions during my time as a student in Groningen. The courses were very interesting and I never had such interesting and professional teachers.

Luckily, I got the opportunity to start a graduation internship at an online market research agency. I had to leave Groningen, but without regret: I had a great time. I would like to thank my supervisors for the great help and support while I was writing my thesis. Next, I would also like to thank my university supervisor Erjen van Nierop, for his critical and extensive feedback. Although at times it was a rather tough research he came up with new points of view and really helped me finalizing my thesis.

Furthermore, I would like to thank my friends at home and in Groningen, who made my study time one to always remember as one of the greatest times of my life. Last, but definitely not least, I would really like to thank my parents and sisters who supported me all the way through my studies. I really appreciate it and I could not have made it without you!

I hope you are encouraged to read my thesis and if you find it interesting, spread the word and recommend it to others as well. I might come with some follow-up information in the future.

(6)

TABLE OF CONTENTS

1 Introduction ... 8

Introduction of the Company ... 9

1.1 The Aim of the Research ... 11

1.2 Theoretical and Social Relevance ... 13

1.3 Structure of the Thesis... 13

1.4 PART I: FEEDBACK RESEARCH ... 15

2 Theoretical Background ... 16

Drivers of Customer Loyalty ... 16

2.1 2.1.1 Customer Loyalty ... 16

2.1.2 Components of Attitudinal Loyalty ... 18

2.1.3 Customer Experience ... 20

Complaint Handling via Surveying Customers About Contact Moments ... 21

2.2 2.2.1 Service Recovery ... 21

The Relationship between Complaint Handling via Surveying Customers and Customer 2.3 Loyalty ... 23

2.3.1 Positive or Negative Experiences with the Company ... 24

2.3.2 Effect over Time ... 24

3 Research Design: the Feedback Product ... 27

Data Collection... 27

3.1 Analysis Plan ... 28

3.2 4 Results: Feedback Research ... 29

Descriptive Statistics ... 29

4.1 Hypotheses Testing ... 31

(7)

4.2.2 Positive or Negative Experiences with the Company (H2) ... 35

4.2.3 Effect of Time Passed After the Research (H3) ... 35

4.2.4 Customer Characteristics ... 37

Summary of the Results ... 38

4.3 PART II: FOLLOW-UP RESEARCH ... 39

5 Theoretical Background ... 40

Follow-up Research ... 40

5.1 5.1.1 Positive or Negative Experiences with the Company ... 41

6 Research Design ... 43

Data collection ... 43

6.1 Analysis Plan ... 44

6.2 7 Results: Follow-up Research ... 45

Descriptive Statistics ... 45

7.1 Hypotheses Testing ... 46

7.2 7.2.1 The Relationship between Feedback Research and Follow-up Research (H4) ... 48

7.2.2 Positive or Negative Experiences with Feedback Research (H5) ... 48

Latent Class Analysis ... 50

7.3 CONCLUSION ... 54

8 Conclusion and Recommendations ... 55

Summary and Recommendations ... 55

8.1 Limitations and Directions for Further Research... 58

8.2 9 References ... 60

(8)

1 INTRODUCTION

For companies these days it is important to determine the most important customers and to decide how to approach them. Only a small percentage of the total amount of customers is responsible for a large part of the revenue of a company. This is also known as the 80/20 rule or Pareto principle (Sanders, 1987; Lammers and Tvetkov, 2008). It is interesting to determine who these customers are, because they are important for a company. Traditionally, marketing activities were often focused on the physical aspects of services or products such as price and quality (Mascarenhas, Kesavan, and Bernacchi, 2004). Currently, the trend is shifting towards more emotional attachment with the brand. Due to the fact that customers are expecting more of a company today, the most successful companies are those that are able to offer a customer more than just an efficient or satisfactory product or service: companies should focus on the customer experience (Dickinson, 2012).

(9)

In some cases generating the customer’s complaints based on a contact moment with the company does not end after an online survey about the contact moment. Companies can decide to also perform follow-up research. This means that based on the online survey after the contact moment, the company contacts the customer by telephone to provide an opportunity to share for instance complaints about the contact moment. This way the customer can express his feelings and the company can prevent him from switching. A failure in service recovery efforts can lead to negative word of mouth (Rothenberger, Grewal, and Iyer, 2008). In contrast, satisfaction with the process of complaint handling is likely to result in positive behavioral intentions, such as the likelihood to recommend the company to others (Rothenberger et al., 2008). This effort of complaint handling can be seen as an extra service moment. Moreover, improved service quality will lead to more satisfied customers (Bitner, Booms, and Mohr, 1994; Anderson, Fornell, and Lehmann, 1994). It is important to take this into account.

I

NTRODUCTION OF THE

C

OMPANY

1.1

For this research a company case has been used. An online market research agency provided the information of a financial services company and the customer data of this company were used. The added value of this approach is that it is a real case and multiple types of researches are conducted for this company. Besides this, the outcomes of this study can be used for future research.

(10)

Figure 1.1

The Feedback Product: Customer Survey, Service Call, and Follow-up Research

It works as follows: a customer of a company might have a complaint, or he might want to end his contract. In order to get this done, he has to contact the company for instance by telephone. After contact with the company (for example via a call center), respondents receive an online customer survey in which they are asked to rate how likely they are to recommend the company to others based on that specific contact moment and they indicate why. This provides insights into what specific processes a company should improve in order to increase the likelihood of customers to recommend the company to others.

With the customer survey the process can end; however, a company can also ask customers to participate in follow-up research. The respondents that filled in a customer survey will be asked whether or not they want to participate in follow-up research. The customers that indicate that they are willing to participate will then receive a service call. In this telephone call the customer will be asked why he is not likely to recommend the company, so he can share his feelings. After this contact moment, the customer receives a second online survey (follow-up survey) in which the likelihood to recommend the company to others is asked again, now based on the phone call. It can then be analyzed whether complaint handling via a service call has an effect on the recommendation likelihood of a customer.

(11)

T

HE

A

IM OF THE

R

ESEARCH

1.2

The purpose of this research is to analyze whether or not extra attention via a customer survey and complaint handling via a phone call increases the likelihood of customers to recommend the company to others. Currently, the positive effects on customer loyalty are visible, but have not yet been proved. In this research it is attempted to determine the effect of handling with customer’s complaints after a contact moment on customer loyalty. Besides, it is analyzed what the effect of follow-up research is. Below, the research problem can be found.

“What are the effects of complaint handling via surveying customers concerning a contact moment with the company on customer loyalty? And, what is the effect of follow-up research on customer loyalty?”

Sub questions that will be answered in this study are the following: 1. What is customer loyalty and what does it comprise?

2. How can companies improve the experiences of a customer with the company? 3. How are customer loyalty and dealing with customer’s complaints via surveys related? 4. Does customer loyalty differ between customers who used to be positive about a

company and customers who used to be negative?

5. Does customer loyalty depend on how many months have passed after filling in a customer survey about the contact moment with the company?

6. What is the effect of follow-up research on customer loyalty?

(12)

Figure 1.2

Amount of Respondents Available for the Customer Survey

Furthermore, the loyalty of customers after the follow-up research is measured. While analyzing the sample it turned out that only a small part of the sample can be used to analyze the effect of up research on customer loyalty. The amount of respondents that participated in follow-up research combined with the two loyalty researches in May and November is too low to statistically draw conclusions, see Figure 1.3. Because of the low amount of respondents (n=1), it has been decided that the part of the follow-up research will be dealt with in a separate part. The loyalty researches in May and November will not be taken into consideration for follow-up research.

Figure 1.3

Amount of Respondents per Group – Follow-up Research

(13)

T

HEORETICAL AND

S

OCIAL

R

ELEVANCE

1.3

Several studies have discussed customer loyalty (Hofmeyr, Goodall, Bongers, Holtzman, 2008; Dick and Basu, 1994; Fitzgibbon and White, 2005; Backman and Crompton, 1991; Day, 1969; Ganguli and Kumar, 2008; Hepworth and Mateus, 1994; Nordman, 2004). The current study elaborates on previous studies because the effect of complaint handling via a customer survey followed by a phone call and a follow-up survey on customer loyalty will be researched. Up until now this has not been researched extensively. In light of these unexplored effects on customer loyalty, the key contribution of the research is to investigate the effect of complaint handling via surveying customers about contact moments with the company. The results can be used to show companies that evaluating customer interactions with a company are important and are contributing to loyalty. Besides this, it is assumed that performing follow-up research is also important in making or keeping customers satisfied. This has not been researched yet, so the current research can be used to fill this gap in academic literature. The most important possible managerial contribution is that, as said previously, management can decide to perform both loyalty research and the feedback product (customer survey and follow-up research). This provides insights on both strategic and customer level.

S

TRUCTURE OF THE

T

HESIS

1.4

(14)
(15)
(16)

2 THEORETICAL BACKGROUND

In order to set up the research, it should be clear what the components of the research are and what they mean. Therefore, a theoretical background is useful in which relevant aspects of this study are described. First, loyalty research will be elaborated on, followed by an explanation of providing extra attention to customers after a contact moment. Also, the hypotheses are provided in this chapter.

D

RIVERS OF

C

USTOMER

L

OYALTY

2.1

For companies these days loyalty research is important because it provides strategic insights into the overall business of a company. Furthermore, thanks to loyalty research companies are able to determine the most important customers. According to the Pareto Principle or 80/20 rule, only a small percentage of the total share of customers is responsible for a large part of the revenue (Eisenberg, 2002). Whether or not a customer is beneficial to the company depends on for instance the purchase history of a customer (Reinartz and Kumar, 2000). In general it can be said that long-term customer relationships are more beneficial for companies than short-term customer relationships. Loyalty research is also important because companies are becoming more customer-centric, as they recognize the importance of cultivating loyal customers (Zeithaml, 2000; Zeithaml, Berry, and Parasuraman, 1996). Also, customer experience is becoming more important, which means that companies should do something extra to get an emotional connection with the customer (Dickinson, 2012). This chapter describes customer loyalty, including the types of loyalty drivers. Moreover, it describes the possible effects of paying extra attention to customers.

2.1.1 Customer Loyalty

(17)

positive. Besides that, compared to behavioral loyal customers, attitudinal loyal customers tend to be more profitable and they are less price sensitive. Due to the fact that these customers are less price sensitive, providing incentives to generate repeat purchases is not necessary, which for the company leads to higher profitability. Furthermore, it is likely that the positive attitude towards the brand results in positive word of mouth (WOM) advertising, which brings customer acquisition benefits (Fitzgibbon and White, 2005). Furthermore, the composite approach of loyalty includes both behavioral and attitudinal loyalty. A customer should purchase products but he should also have affection with the brand in order to be truly loyal (Day, 1969).

Based on the conceptualized figure of Dick and Basu (1994) different loyalty groups can be identified, see Figure 2.1. A combination is made between repeat patronage and relative attitude towards the brand or service. Repeat patronage represents behavioral aspects of loyalty, whereas relative attitudes address the attitudinal aspects of loyalty (Nordman, 2004). Four categories of customer loyalty can then be identified: loyalty, latent loyalty, spurious loyalty, and no loyalty. The category loyalty illustrates the ideal situation when a company is striving for loyal customers (Dick and Basu, 1994). Latent loyalty exists when there is no or low behavioral loyalty and relative attitude is high (Dick and Basu, 1994); this can be caused by situational factors and social norms. Customers might buy from a competing service provider whereas their relative attitude to the old service provider is higher. With spurious loyalty repeat purchasing is high and the attitude is low. Customers in this situation continue to patronize a service provider whereas they do not have a positive relative attitude towards the service provider (Dick and Basu, 1994). Last, when both relative attitude and behavioral loyalty are low, there is no loyalty (Dick and Basu, 1994). So, it is important to come to the situation with high relative attitude combined with repeat patronage.

Figure 2.1

(18)

Now that the types of loyalty have been described, the drivers of loyalty will be explained. Due to the fact that only attitudinal aspects of loyalty are taken into account in this study, behavioral loyalty will not be elaborated on in the literature review. The components of attitudinal loyalty will be explained next.

2.1.2 Components of Attitudinal Loyalty

When a customer is loyal, this indicates that he will purchase the same brand, he most of the time increases the amount of purchases, and he is willing to recommend the brand to others (Hepworth and Mateus, 1994). This leads to (among others) two components that can be used for attitudinal loyalty: the repurchase intention or purchasing additional products or services from the same company and the willingness to recommend the company to others (Akbar and Parvez, 2009; Ganguli and Kumar, 2008). In other literature, the repurchase intention and positive recommendations are indicated as the key manifestations of attitudinal loyalty (Vázquez-Casielles, Suárez-Álvarez, and Del Río-Lanza, 2009). They are described below.

One way to measure attitudinal loyalty is via measuring the repurchase intention of a customer. Repurchase intention is very important for a defensive marketing strategy and for business success (Cronin, Brady, and Hult, 2000). The focus is mostly on attracting new customers which brings an increase in costs for attracting customers as well as for competition. Most companies use the repurchase intention of customers to make forecasts about the adoption of new products or repeat purchases and thus about customer loyalty (Jamieson and Bass, 1989).

A second component, which is often used together with the repurchase intention and seems to work well across industries, is the recommendation likelihood. According to Vázquez-Casielles, Suárez-Álvarez, and Del Río-Lanza (2009), positive recommendations are one of the key manifestations of attitudinal loyalty. Besides, Bowman and Narayandas (2001) measured WOM via a survey and found that word-of-mouth increases customer loyalty. In this research the recommendation likelihood will be used as a measure for customer loyalty.

(19)

product] to family, friends or colleagues?” Respondents can indicate their recommendation likelihood on a 0 to 10 rating scale. Based on the rating they can be categorized into three groups: promoters (mark nine or ten), passives (mark seven or eight), and detractors (mark zero to six). The NPS is calculated by subtracting the share of detractors from the share of promoters, see Figure 2.2 below.

Figure 2.2

Net Promoter Score (Philips, K. 2012)

Contradictory statements have been published about the NPS. On the one hand, researchers are critical about the metric. It is argued that calculating NPS only is not sufficient, because it is only a single loyalty question (Hayes, 2008). A customer might be very satisfied with a firm and can have a high likelihood to recommend it to others. However, he might equally like or even prefer a competitor which will in the end lead to a decrease in sales (Keiningham, Aksoy, Buoye, and Cooil, 2011). On the other hand, it is the most used traditional metric to improve customer loyalty (Keiningham et al., 2011). Moreover, asking one question is simple and can be a useful barometer for the advocacy and loyalty of existing customers (Samson, 2006). The NPS is not only used to determine the likelihood that a customer will recommend the company to others (Farooqi and Rehmaan, 2010), but it can also determine how to boost future loyalty and thus profitability (Garrity, 2010). It is a metric that is easy to analyze and it is easy to explain the results (Farooqi and Rehmaan, 2010). The NPS can be used across the whole company for measuring progress of processes or performances.

(20)

2.1.3 Customer Experience

The last 25 years the focus of marketing practice and research has undergone transformations. It started as creating fast-moving consumer product brands, and shifted to building customer relationships through service marketing. Currently, the focus lies on creating compelling customer experiences (Maklan, and Klaus, 2011). The transformation can be found in the framework introduced by Pine and Gilmore (1999, 2000), see Figure 2.3 below.

Figure 2.3

Experience Economy: Progression of Economic Value (Pine and Gilmore 1999, p. 22)

(21)

the customer and this way increase the experience. Companies should always try to ‘delight’ customers by striving to exceed the customer’s expectations (Parasuraman, Zeithaml, and Berry, 1988).

Now that customer loyalty is described, the focus will be on the effect of complaint handling via surveying customers about a contact moment with the company.

C

OMPLAINT

H

ANDLING VIA

S

URVEYING

C

USTOMERS

A

BOUT

C

ONTACT

M

OMENTS

2.2

In this section it will be attempted to take a closer look into the extra attention paid to customers about contact moments with the company and what the effect is on customer loyalty.

Companies are requesting insights in customer experiences on day-to-day business activities throughout the entire customer life cycle in order to maintain and improve customer satisfaction and to increase loyalty. This is because the actions and behaviors of employees towards customers may influence customer loyalty (Crosby and Johnson, 2006). Therefore, it is important that companies are aware of a customer’s opinion about the contact they had with the employee or company. Moreover, employees should understand and experience what role they play in the customer loyalty strategy in that it can be evaluated what customer loyalty drivers (for instance the customer’s perceptions, likelihood to buy more or make recommendations) are applicable to them (Crosby and Johnson, 2006). This part discusses why it is interesting to continuously measure how customers evaluate a contact moment with a company.

2.2.1 Service Recovery

(22)

Next to doing things right, it is important that problems of customers, whenever they occur, are taken care of effectively (Hepworth and Mateus, 1994). Negative experiences with for instance service, product quality, or value for money can lead to dissatisfied customers. One problem experience can cause a 30 percent decrease in customer loyalty (Hepworth and Mateus, 1994). Most dissatisfied customers do not explain why they are dissatisfied and when asked they provide responses which are logical or socially acceptable, such as poor service or too expensive products (Hepworth and Mateus, 1994). According to this study, key in obtaining information from dissatisfied customers is to provide them the opportunity to talk about their problems (Hepworth and Mateus, 1994).

Thus, in order to obtain the real explanation, customers should be triggered to express their complaints or problems and they should have the opportunity to do this. And afterwards, it is important that their problems are taken seriously (Hepworth and Mateus, 1994). This can be done by finding out how customers evaluate the employees and service of a company. It is important to ask customers for feedback. This can be done for instance via a survey asking customers for experiences with the company (based on a contact moment). Once a company knows what problems customers are facing, serious and systematic efforts for correcting the problem (service failure) should follow. This is called service recovery (Lovelock and Wirtz, 2007: 395). The most important outcome of service recovery is maintaining customers that would otherwise switch company (Griffin and Lowenstein, 2001). A complaint management system could solve problems of customers, as well as improve the system based on customer’s complaints (Lovelock and Wirtz, 2007: 395). Satisfaction with the process of complaint handling is likely to result in positive behavioral intentions, such as the likelihood to recommend the company to others (Rothenberger, Grewal, and Iyer, 2008). In contrast, a failure in service recovery efforts can lead to negative word of mouth (Rothenberger et. al., 2008).

(23)

The next section describes how surveying customers about their experiences with the company relates to customer loyalty.

T

HE

R

ELATIONSHIP BETWEEN

C

OMPLAINT

H

ANDLING VIA

S

URVEYING

C

USTOMERS AND

2.3

C

USTOMER

L

OYALTY

The effects of extra attention and complaint handling on customer loyalty can be measured, for instance via an online survey that requests for customer feedback. Feldman and Lynch (1988) have initiated the self-generated validity theory, which is defined as a strengthened relationship between latent intentions and actual behavior caused by the measurement of intentions, so it explains the reactive effects of measurement on behavior (Chandon, Morwitz, and Reinartz, 2005). They argued that respondents that filled in a survey about a company form judgments which they had otherwise not formed or accessed in memory. The survey causes people to think and that makes them more likely to form judgments (Feldman and Lynch, 1988). Offering the possibility to customers to express themselves via an online survey might bring more positive customers. On the other hand, customers might also have negative experiences with the company. For instance, an employee at a call center might receive large call volumes and repetitive questions, which might cause an employee to be falling into an assembly line mode of dealing with customers (Evans, Arnold, and Grant, 1999). This might affect customers’ perspectives of service quality (Brady and Cronin, 2001).

(24)

likely to recommend the company to others (Markey et al., 2009). It is assumed that these customers are more likely to be loyal customers. The following hypothesis has been set up set test whether there is an effect of the handling of a complaint on customer loyalty.

H1: Customer loyalty is higher when a customer’s complaints about a contact moment with the

company are handled via an online survey than when they are not.

2.3.1 Positive or Negative Experiences with the Company

Every customer will evaluate the contact moment differently. Some of them might have had very negative experiences which will lead to a customer being very unlikely to recommend the company to others. Others might also have had negative experiences, but can still be rather likely to recommend the company. Both groups will be asked to participate in the survey about the contact moment. The effect might be stronger for the group that was very unlikely to recommend the company than the group that was more positive. So there might be a difference between the groups (promoters, passives, and detractors) and the effect on customer loyalty. In general, very satisfied customers are most often more forgiving than less satisfied customers, which is why attention should be paid to especially less satisfied customers (Van Doorn and Verhoef, 2008).

H2: The effect of handling a complaint (via an online survey) on customer loyalty is stronger for a

customer who is not likely to recommend the company (based on loyalty research in May) than for a customer who is already likely to recommend the company.

2.3.2 Effect over Time

(25)

be either positive or negative), which will lead to a critical evaluation. However, when the time between the survey and the loyalty research increases, the information becomes less accessible. This can also be called ‘regression to the mean’. Regression to the mean refers to a statistical phenomenon that occurs when on the same subject of observation repeated measurements are made (Barnett, van der Pols, and Dobson, 2004). This occurs because there is variation in the observed values around a true mean (such as random measurement error). This can also be described as things evening out, when a random variable different from the norm returns to normal, to the mean.

So there might be a difference between a customer that was surveyed six months ago and a customer that was surveyed one month ago. The first customer might have changed his opinion because the feeling he had after the contact moment has ‘faded away’, so there might be an effect over time. The second customer might still feel this way because his experiences occurred recently. It will be investigated whether or not the effect of complaint handling via an online survey changes when time between the survey and the loyalty research in November increases. This can be summarized in the following hypothesis:

H3: The effect of complaints handling (via an online survey) on customer loyalty decreases when

the time between the online survey and the loyalty research in November increases: for a customer who recently had a contact moment with the company the effect of the online survey is stronger than for a customer who contacted the company six months ago.

(26)

Figure 2.4

Conceptual Framework Customer Survey

Below the dependent variable, independent variable and moderator are described based on the conceptual model.

Dependent Variable

The dependent variable concerns the change in customer loyalty, measured via the recommendation likelihood (RL) from time t to time t+1. Time t occurred in May 2012, time t+1 in November 2012. This concerns a time span of six months. When it comes to customer loyalty the RL is related to the company in general and not specified to a contact moment.

Independent Variable and Moderator

It is analyzed whether participation in the research product has an effect on customer loyalty. Therefore a variable participation in customer survey yes/no is used as independent variable. This variable concerns the likelihood to recommend a company after a contact moment with the company. To test whether an effect exists between positive or negative customers, the RL is used as moderator. Besides this, also the effect of the amount of months that have passed after the research is included as moderator. Furthermore, variables are included that can be used to describe the respondents.

(27)

3 RESEARCH DESIGN: THE FEEDBACK PRODUCT

In this chapter the data collection and analysis plan are provided. The data collection is described first.

D

ATA

C

OLLECTION

3.1

This subchapter contains the design of the study and the description of how the data has been collected. An online quantitative loyalty survey is conducted among the customers of an insurance company. Customers are contacted via email during the first three weeks of November 2012. In order to analyze whether complaint handling via surveying customers influences customer loyalty, two measurement points are necessary: the research in November 2012 and a loyalty research that serves as a benchmark to see whether changes have occurred. A loyalty research in May 2012 will serve as benchmark. A sample of 255 respondents is available with respondents that participated in both loyalty researches. Of the sample, 46 respondents were invited for the feedback research and 35 completed a customer survey. The other part of the sample (n=209) is used as control group for this research.

Figure 3.1 illustrates the amount of respondents per group that can be taken into consideration for this study. As can be seen, the 209 concern the respondents that participated in both loyalty researches in May and November, but did not participate in the research. The 46 respondents participated in both loyalty researches in May and November and have had a contact moment with the company. After the contact moment they were invited for the customer survey. 36 of this group started the survey, and 35 of them completed the survey. This means that only one of them did not complete it. The total sample contains 255 respondents.

Figure 3.1

(28)

In the next section the analysis plan is described including the tests that will be performed.

A

NALYSIS

P

LAN

3.2

(29)

4 RESULTS: FEEDBACK RESEARCH

This chapter contains the results of the first part of the study. First, the descriptive statistics are provided. Linear regression analyses will show whether or not the variables are related and if the variables have a positive or negative effect on the change in loyalty from May to November. In this chapter, the abbreviation RL is used for the recommendation likelihood. First, the descriptive statistics are provided.

D

ESCRIPTIVE

S

TATISTICS

4.1

In total 255 respondents are included in this research. These respondents participated both in loyalty research in May and in November. As said previously, this selection is made due to the fact that it is possible to see the change in recommendation likelihood (RL) from May to November. This way it can be determined whether feedback research influences the change in RL from May to November.

In total the sample consists of 56% consumers, 25% business people, and 19% entrepreneurs. The table below contains information about how long the respondents have been a customer (so the duration or length of the relationship) and what their future intentions are. Concerning the duration of the relationship, most respondents have been a customer for ten to fifteen years or even longer, see Table 4.1

.

Future intentions might be to extent the number of products with the company, to retain an equal amount of products, decrease the amount of products or quit the relationship with the company. Most respondents indicated to retain an equal amount of products with the company (90%) and will not expand or stop them.

Table 4.1

Summary Characteristics of the Sample

Type of customer 142 consumers (56%) 64 business people (25%) 48 entrepreneurs (19%) Relationship duration 1-3 years: n=11 (4%)

(30)

20-25 years: n=25 (10%) 25 years or more: n=56 (22%) Future intentions Extent number of products: n=7 (3%)

Retain an equal amount of products: n=230 (90%) Decrease amount of products: n=15 (6%)

Quit products: n=3 (1%)

The mean score that respondents gave for the loyalty research in May is 7.09, see Table 4.2. 18% of the respondents were promoters of the company, 58% passives, and 24% detractors in May, which means that the NPS is -6 (percentage promoters 18% minus percentage detractors 24%). In November the mean score is 7.13. This time, 15% is promoter, 64% passive and 22% detractor of the company, which means the NPS is -7. Of the 255 respondents that were included in this research, 46 respondents had contact with the company and were invited for the customer survey. Of this group, 36 actually started filling in the survey and just one of them did not finish the survey, so 35 respondents (76%) completed the survey. Of the 35 respondents that completed the survey the mean RL is 7.3. The respondents are divided among the three RL groups as follows: 29% is a promoter of the company, 54% is a passive, and 17% detractor, which gives an NPS of 12.

Table 4.2

Summary Descriptive Statistics (Amount of Respondents, Mean RL, RL Groups, and NPS)

May Customer Survey November Respondents n=255 n=35 (completed) n=255

Mean RL 7.09 7.29 7.13

Promoters (RL score 9&10) n=45 (18%) n=10 (29%) n=37 (15%) Passives (RL score 7&8) n=148 (58%) n=19 (54%) n=163 (64%) Detractors (RL score 6-10) n=62 (24%) n=6 (17%) n=55 (22%)

NPS -6 12 -7

(31)

Table 4.3

Summary Descriptive Statistics for the Participants of the Customer Survey (n=35)

May Customer Survey November

Mean RL 7.37 7.29 7.14

Promoters (RL score 9&10) n=8 (23%) n=10 (29%) n=6 (17%) Passives (RL score 7&8) n=20 (57%) n=19 (54%) n=22 (63%) Detractors (RL score 6-10) n=7 (20%) n=6 (17%) n=7 (20%)

NPS 3 12 -3

The amount of promoters for loyalty research has decreased from May to November from 18% to 15%, while for the customer survey the amount of promoters is higher (29%), which was shown in Table 4.2. It is now checked how many promoters in November have participated in the customer survey. It turns out that the amount of promoters is higher among the participants of the customer survey than among the respondents that did not participate in the survey, see Table 4.4. This assumes that the respondents that participated are more likely to be a promoter of the company. However, the difference in characteristics of the respondents can be due to self-selection problem, because the survey is put on the web and respondents are those that have access to Internet and decide to participate in the survey. There is no control of the selection process, which makes it hard to determine the accuracy of the estimates (Bethlehem, 2008). On the other hand, this cannot really affect the results because the loyalty research is also done online, so the group without customer survey also needs access to the Internet.

Table 4.4

Difference RL Groups with/without Customer Survey

Loyalty research November Promoters Passives Detractors

Customer survey YES 17% 63% 20%

Customer survey NO 14% 64% 22%

H

YPOTHESES

T

ESTING

4.2

(32)

experimental or the control group, and a pretreatment measure is taken on each group. The experimental group is exposed to the treatment (X), and finally a post treatment measure is taken on each of the experimental and control groups (O2) (Malhotra, 2010, p. 259). Also, a post

treatment measure is taken on each of the experimental and control groups. It is symbolized as follows: O1 X O2.

The treatment effect is computed as O2 - O1. The treatment effect concerns the change in loyalty from May to November (which can be computed as the RL in November subtracted by the RL in May). First, the correlation between the two variables is checked. Due to the fact that these are two variables with a value between 0 and +1, the measure of association that can be used is the Cramér’s V (φc). The association between the RL of loyalty research in May and the RL of loyalty research in November is significant (p=.000), with a value of .464 which indicates that there is an association between the two. Also the Pearson’s r (r=.728) and the Spearman Correlation (ρ=.652) are significant (p=.000), see appendix Table A. From the t-test it turns out that the means of the two variables differ (mean loyalty May: 7.09, mean loyalty November: 7.13, p=.000). Now, in order to be able to analyze the change in RL from May to November, a variable is computed which is called the RL difference: the RL of loyalty research in November minus the RL of loyalty research in May. This indicates the change in RL between the two periods.

A linear regression analysis is performed with the RL difference as dependent variable. In order to be able to answer the first hypothesis, the variable participation in customer survey yes or no is included. Next to this, for the second hypothesis a moderation effect is added, which concerns the variables customer survey yes / no multiplied by the RL of the customer survey to determine whether the

(33)

Furthermore, the reason for recommending the company, the type of customer, future intentions, and duration of the relationship are included. Table 4.5 below shows the output of the linear regression. The overall model is significant (p=.001), the R2 is .128. The variable RL at customer survey (main effect for the second hypothesis) is excluded from the analysis due to multicollinearity: the tolerance criterion is not met (the default tolerance level is .0001).

Table 4.5

Linear Regression Output: Effects on the Change in Loyalty from May to November (1)

Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error (Constant) .899 .586 1.535 .126

H1 Customer survey yes/no .343 1.090 -.087 -.315 .753

H2 Customer survey yes/no * RL at customer survey

.050 .101 -.108 -.494 .622

H3 Months passed after customer

survey

.081 .121 -.099 -.670 .504

H3 Customer survey yes/no * Months

passed after survey

-.142 .193 .163 .736 .463 Customer Type .007 .108 .004 .063 .950 Customer Duration .009 .049 .012 .189 .850 Future Intentions -.474 .232 -.126 -2.039 .042 Products 1.114 .467 .173 2.385 .018 My Relationship With -.071 .230 -.020 -.311 .756 Accessibility .232 .127 .124 1.826 .069 Behavior Employee .282 .098 .222 2.859 .005 Communication .000 .080 .000 -.002 .998 Service -.108 .069 -.125 -1.561 .120

ANOVA overall model sig.: .001 R2: .128

Notes: Significant p-values are in bold

(34)

Variables were excluded step by step, starting with the variables with the highest p-value, in order to see whether more variables turn out to be significant. However, this does not lead to more significant variables. Moreover, another linear regression analysis is performed without the reasons for recommending the company (which are: products, my relationship with, accessibility, behavior of the employee, communication, and service). The new regression output can be found in Appendix Table B.

It turns out that this model is not significant (p=.279). Furthermore, the R2 is low (R2 =.034), which means that it is hard to explain the model with the variables included in this model. The first model (Table 4.5) will be used to answer the hypotheses. This is done next in section 4.2.1, 4.2.2, and 4.2.3.

4.2.1 The Relationship between Complaint Handling via Surveying Customers and Customer Loyalty (H1)

The first hypothesis concerns the relationship between participating in a customer survey and the change in loyalty (RL) from May to November. It is tested whether customer loyalty is higher when a customer is able to express his complaints about a contact moment via an online survey than when he is not.

From the linear regression analysis it turns out that the relationship between participating in feedback research and the change in RL from May to November is not significant (p=.753), see Table 4.5. This means that there is no significant association between participating (yes or no) and the change in RL from May to November, which can be caused by the low amount of respondents that participated in the research (n=35). The effect should be further researched in future studies.

(35)

4.2.2 Positive or Negative Experiences with the Company (H2)

It might be the case that respondents of loyalty research in May become more positive thanks to positive experiences. This is analyzed in this section. The RL score of the respondents for the research is taken as moderator. The relationship between the RL given at a customer survey and the change in RL from May to November is analyzed. It is tested how respondents behave for instance after a

negative contact moment with the company. For instance, if the RL of a respondent increases due to the research, does the change in RL from May to November increase as well?

The main effect of the RL on the change in customer loyalty from May to November is excluded from the analysis due to multicollinearity. The moderator variable shows no significant effect on the change in customer loyalty from May to November (p=.622). This indicates that for participants of the research an increase in RL does not significantly contribute to a higher RL score in November compared to May. The groups for feedback research as presented in Table 4.6 are very small, which might cause the insignificant relationship for this variable.

Table 4.6

Amount of Promoters, Passives, and Detractors for Customer Survey

Customer Survey Promoters (RL score 9 & 10) n=10 (29%) Passives (RL score 7 & 8) n=19 (54%) Detractors (RL score 0 – 6) n=6 (17%)

In the next section, the relationship between the amount of months between the research and loyalty research on the change in customer loyalty is analyzed.

4.2.3 Effect of Time Passed After the Research (H3)

(36)

the customer loyalty research. Table 4.7 shows how the respondents that participated in the research are divided over the months. As can be seen, the amount of respondents per month is rather small so it is not allowed to draw conclusions from the analyses.

Table 4.7

Division of Respondents per Month

May June July August September October November

6.5% (n=3) 8.7% (n=4) 21.7% (n=10) 30.4% (n=14) 13.0% (n=6) 15.2% (n=7) 4.3% (n=2)

Figure 4.2 below illustrates the change in customer loyalty from May to November per month passed. From the figure it can be seen that there is no clear structure between the months that have passed after the research and the change in RL for May compared to November.

Figure 4.2

Effect Months Passed on the Change in Customer Loyalty

(37)

negative about the company. From May to November the RL has increased from 5.7 to 6.3, whereas the RL for non-participants has decreased from 7.9 to 7.5. This indicates that although the RL does not seem to have increased, it did however increase with regard to loyalty research in May.

A linear regression with the variable Months passed as main effect and the change in loyalty as dependent variable shows no significant effect (p=.504), see Table 4.5. Also, the moderation effect (the months passed multiplied by whether or not the respondent participated in the research) is also not significant (p=.463). In sum, it can be said that in this study the effect of the time that has passed on customer loyalty cannot be proven significantly.

4.2.4 Customer Characteristics

It can be analyzed whether the change in customer loyalty relates to the type of customer, future intentions of customers, and the duration of the relationship. From the linear regression it turns out that only the variable ‘future intention’ has a significant effect on customer loyalty (p=.042, β=-.474). The direction of the beta indicates that an increase with one unit (which means that the customer becomes less likely to extent the amount of products with the company) decreases the change in loyalty, so the customer becomes less likely to recommend the company. The other way around it says that the more likely a customer is to stay with the company (extend the amount of products), the higher his change in loyalty. This is also what can be expected: namely that a customer who is willing to increase the amount of products is also more likely to be a loyal customer.

(38)

S

UMMARY OF THE

R

ESULTS

4.3

The results of the analyses described in this chapter will shortly be summarized. First, the effect of participating in feedback research on the difference in RL from May to November was analyzed (H1). Here, it turned out that no relationship can be found between participating in the

research and the change in customer loyalty from May to November.

Second, the effect of the research was analyzed per RL group (H2). Unfortunately, the RL groups

are rather small, so it is not statistically allowed to draw conclusions from the analysis.

Third, the number of months passed after the research does not have a significant effect on customer loyalty (H3). So the change in customer loyalty is not caused by the moment the

customer survey was performed. The non-significant results might be due to the low amount of respondents spread over the months. This hampers the

significance of the analyses.

In sum, none of the three hypotheses can be accepted. This is probably due to the low amount of respondents that participated in the research.

(39)
(40)

5 THEORETICAL BACKGROUND

Part II of this paper includes the follow-up research and is organized in the same way as part I. This chapter will provide the literature review of the follow-up study. It is described what it comprises and what effects of follow-up research on customer loyalty can be expected. Here, also the abbreviation RL is used for the recommendation likelihood.

F

OLLOW

-

UP

R

ESEARCH

5.1

As stated previously, follow-up research concerns calling back customers that, via a survey, gave remarks about a contact moment with the company. These call back moments are initiated to provide customers the opportunity to talk about the complaint or remark given in the survey. As stated by Bitner, Booms, and Stanfield Tetreault (1990), human interaction for service delivery is essential for determining satisfaction. This indicates that customers are expected to become more satisfied when the service is delivered via human interaction. So adding a telephone call in addition to an online survey might lead to more satisfied customers.

(41)

So, based on the telephone call a customer might become more positive about the company thanks to the extra attention. After the telephone call, the customer will receive another survey in which he can evaluate the contact moment again and indicate its likelihood to recommend the company based on the service call. The effect of follow-up research on customer loyalty will be tested with the following hypothesis:

H4 Customer loyalty will be higher when, after feedback research, a customer receives a

follow-up phone / service call to elaborate on the complaint than when he does not.

5.1.1 Positive or Negative Experiences with the Company

As stated before, customers will evaluate the contact moments differently. Some of the customers could have had very negative experiences which will lead to a customer being very unlikely to recommend the company to others, while others might be rather likely to recommend the company to others. Thanks to the service call of follow-up research a customer might become more positive which affects the recommendation likelihood (RL). But for a customer that is already likely to recommend the company, the effect of follow-up research is expected to be less strong. In general, very satisfied customers are most often more forgiving than less satisfied customers, which is why attention should be paid to especially less satisfied customers (Van Doorn and Verhoef, 2008). This leads to the following hypothesis:

H5: The effect of a follow-up phone / service call on customer loyalty is stronger for a customer

who is not likely to recommend the company after the first contact moment with the company than for a customer who is already likely to recommend the company.

(42)

Figure 5.1

Conceptual Framework Follow-up Research

Below the dependent variable, independent variable and moderator are described based on the conceptual model provided (Figure 5.1).

Dependent Variable

The dependent variable is the change in customer loyalty, which is the change in recommendation likelihood (RL) at follow-up research compared to the RL at the customer survey. This concerns the likelihood to recommend the company based on the second contact moment (service call) with the company compared to the recommendation likelihood a respondent gave at the customer survey (based on the first contact moment).

Independent Variables and Moderator

It is analyzed whether the follow-up phone /service call has an effect on customer loyalty. After the phone call the likelihood to recommend a company has been measured via a follow-up survey. In the conceptual model the variable follow-up research yes/no indicated whether a customer has received a phone call yes or no. It will also be tested whether a difference exists between positive or negative customers for follow-up research; this is analyzed with a moderator variable. The moderator variable includes the different RL groups promoters, passives, and detractors.

(43)

6 RESEARCH DESIGN

After feedback research the respondents that filled in a customer survey will be asked whether or not they want to participate in follow-up research. The customers can indicate that they are willing to participate in follow-up research. They will receive a follow-up service call. In this telephone call the customer will be asked why he is not likely to recommend the company, so he can share his feelings. After this contact moment, the customer receives a second online quantitative survey in which the likelihood to recommend the company to others is asked again, now based on the phone call. The data of the online survey will be used in this research.

D

ATA COLLECTION

6.1

The data for this part of the study is thus conducted via online quantitative surveys; sent to respondents after they have received a follow-up phone call. Furthermore, the RL those respondents gave for feedback research is also taken into account. This way, it is possible to analyze the change in customer loyalty (the RL for follow-up research minus the RL for feedback research).

Figure 6.1 below shows the amount of respondents for the part of the follow-up research. Without taking into account the loyalty researches in May and November, 576 respondents are left. This means that for the follow-up research only the customer survey and follow-up survey are used. The last row in the figure contains the respondents used; they have had a contact moment with the company, were invited and started the customer survey; they completed the survey and participated in follow-up research.

Figure 6.1

(44)

Due to the fact that the effect of participating in follow-up research is analyzed, a control group is required. The control group consists of 35 respondents that did not participate in follow-up research. Instead of the RL for follow-up research, here the RL provided in November is included. This way it is possible to determine a change in loyalty without participation in follow-up research.

Now that it is known what data will be used, the analysis plan is described.

A

NALYSIS

P

LAN

6.2

This subchapter describes the analyses that can be performed with the data available for follow-up research. It will be analyzed if follow-follow-up research has an influence on customer loyalty. Besides that, it will be analyzed whether differences exist between respondents that are very positive after feedback research and respondents that are negative. The respondents are divided into groups (promoters, passives, and detractors) and the groups will be used in the analysis. First, descriptive statistics will be provided, including the type and division of data in the sample. In order to test the effects of follow-up research on customer loyalty and what differences exist between the groups, linear regressions will be performed.

(45)

7 RESULTS: FOLLOW-UP RESEARCH

This chapter contains the results of the second part of this study concerning follow-up research. First, the descriptive statistics are provided. Furthermore, a regression analysis shows whether or not the variables are related and if they have a positive or negative effect on loyalty. Besides, a latent class cluster analysis is provided which shows how the data can be clustered and what cluster is most interesting when it comes to follow-up research. In this chapter also the recommendation likelihood is abbreviated to RL.

D

ESCRIPTIVE

S

TATISTICS

7.1

As was concluded in Figure 1.3 in paragraph 1.3 only one respondent participated in both loyalty researches in May and in November. Due to this low amount of respondents, the two loyalty researches in May and November will not be taken into consideration in this part of the research, so only the respondents that participated in the customer survey and follow-up research are included. In total 576 respondents are included in this part of the research. 72% of them were invited for follow-up research, because the company wants to provide the customer the opportunity to share positive and/or negative experiences. Of the respondents that were invited for follow-up research, 67% actually completed the follow-up survey. This means that 385 respondents are left that participated in follow-up research. The mean recommendation likelihood (RL) of the respondents in this data is 5.2, whereas the mean RL for follow-up research is 7.4, see also Table 7.1. This indicates that on average the RL for follow-up research has increased. There are a lot more promoters for follow-up research and the percentage of detractors has decreased a lot. This shows that the respondents became more positive thanks to follow-up research. The next section analyzes the effect of follow-up research on loyalty.

Table 7.1

Descriptive Statistics Follow-up Research (Amount of Respondents, Mean RL, RL Groups)

Total Follow-up Respondents n=576 n=388

Mean RL 5.2 7.4

(46)

H

YPOTHESES

T

ESTING

7.2

When analyzing the effect of follow-up research (the phone call) on customer loyalty, it is tested whether customer loyalty will be higher when a customer’s complaint about a contact moment with the company is dealt with via a phone / service call, than when they are not. Compared with the feedback research, the RL of 298 respondents has increased for follow-up. For 72 respondents it has remained the same, their RL did not change. Only 18 respondents gave a lower RL score for follow-up research. Furthermore, from a One-Way ANOVA test it turns out that the mean RL of feedback research significantly differs from the RL of follow-up research (p=.000).

The effect of follow-up research on customer loyalty will be tested with a linear regression analysis. The dependent variable that will be used in the linear regression analysis is delta customer loyalty: subtracting the RL at feedback research from the RL at follow-up research. The control group consists of 35 respondents that did not participate in follow-up

research. Instead of the RL for follow-up research, here the RL in November is included. This way it is possible to determine a change in loyalty without participation in follow-up research.

(47)

Table 7.2

Regression Output: Effect of follow-up on Delta Customer Loyalty (RL of Follow-up Research minus the RL of Feedback Research) Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error (Constant) -.143 .388 -3.69 .713 H4 Follow-up yes/no 2.356 .405 .274 5.821 .000

ANOVA overall model sig.: .000 R2: .075

Notes: Significant p-values are in bold

Table 7.3

Regression Output: Moderating Effects on Delta Customer Loyalty (RL of Follow-up Research minus the RL of Feedback Research) Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error (Constant) -.333 .979 -.340 .734

H5 Passives at customer survey -6.384 1.879 -1.263 -3.398 .001

H5 Passives at customer survey *

Follow-up yes/no

.797 .173 1.462 4.601 .000 H5 Detractors at customer survey -3.093 1.101 -.616 -2.808 .005

H5 Detractors at customer survey *

Follow-up yes/no

.796 .061 1.361 12.976 .000

ANOVA overall model sig.: .000 R2: .482

Notes: Significant p-values are in bold

(48)

7.2.1 The Relationship between Feedback Research and Follow-up Research (H4)

First, the main effect of the follow-up research on the change in RL from feedback research to follow-up research (delta customer loyalty) is analyzed. The regression output can be found in Table 7.2 and Table 7.3.

The main effect of the follow-up on the change in loyalty from feedback research to follow-up research is significant (p=.000, β=2.356). This indicates that follow-up has a positive effect on the change in customer loyalty: the change in loyalty increases significantly when a customer has participated in follow-up research. Next, it is analyzed whether the effect of follow-up differs per RL group.

7.2.2 Positive or Negative Experiences with Feedback Research (H5)

Next to the effect of follow-up research on customer loyalty, it can be tested whether a difference exists between the changes in RL between three groups. As was stated in paragraph 2.1.2, based on the RL score respondents can be categorized into three groups: promoters (mark nine or ten), passives (mark seven or eight), and detractors (mark zero to six). To see whether

the effect differs between the three groups, they are included in the model as dummy variables. As said in the beginning of the chapter the group of promoters is very small (n=4) and therefore excluded from the regression analysis. It is expected that the chance that the RL of detractors increases is higher than that the RL of promoters increases. This is due to the fact that more improvements can be made among detractors; the passives are already more likely to make recommendations.

Referenties

GERELATEERDE DOCUMENTEN

The tri-dimensional concept customer brand engagement (based on cognitive-, emotional- and intentional brand engagement) was used to understand what motivates customers

The reasoning behind the outcome of the first hypothesis (H1) is that when the relationship quality between the customer and the service provider is perceived

What is the influence of the following factors: price perception, quality, image and perceived channel integration on the online and offline loyalty, and what are the differences?.

The effect of channel integration, service quality, assortment variety, price perception and image is examined in both online and offline channels to see whether there are

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

This study focuses on the effects of customer-firm relationship characteristics – depth, length and breadth – and the effect of bundle completeness (the extent

The first test is conducted with the variable in which the discount is already subtracted from the spending amount. Table 4.7 contains the output of this test. In order to see whether

Based on this expected effect and the research of Liu and Brock (2007) the assumption is that relationship length positively moderates the expected effect