Fa F a cu c ul lt ty y o of f M Ma an na a ge g em me e nt n t a a nd n d O Or rg ga an n iz i za at ti io o n n De D e pa p ar rt tm me en nt t o of f M Ma a rk r ke et ti in ng g
C C us u s t t om o me er r L L o o ya y al l t t y y P Pr ro og gr r am a m s: s : A A S S e e gm g me e n n t t at a ti io on n R R e e se s ea a r r ch c h
Au A ut th ho or r : : E Ev va an ng ge el li ia a F Fr ra ag gk ki ia ad do ou ul la ak ki i s1 s 15 50 01 16 65 58 8 S
Su up pe er rv vi is so or r: : P Pr ro of f. . D Dr r. . T T. . H H. . A A. . B Bi ij jm mo ol lt t Pl P la ac ce e/ /D Da at te e: : G Gr ro on ni in ng ge en n, , J Ju un ne e 2 20 00 05 5
T
Thhee aauutthhoorr iiss rreessppoonnssiibbllee ffoorr tthhee ccoonntteennttss ooff tthhiiss rreesseeaarrcchh rreeppoorrtt;; tthhee aauutthhoorr hhoollddss tthhee ccooppyyrriigghhtt ooff tthhiiss rreeppoorrtt
2
Fo F or re ew wo or rd d
The document on hand is the result of the research performed to finalize my education at the faculty of Management and Organization at the University of Groningen. This is why, at this point, I would like to acknowledge the people around me that have made my education to a success.
First and foremost, I would like to thank my supervisor Prof. Dr. T. H. A. Bijmolt in providing data and useful comments on the submitted draft documents.
Undoubtedly the completion of this work would be impossible without his contribution.
Furthermore, I would like to thank my family who supported me morally and financially, regardless of the fact that they all were thousands of miles away. Last but definitely not least, a lot of thanks go out to my boyfriend Bart who always kept his confidence in me and thus significantly contributed to the accomplishment of this master degree during this hard year abroad.
Evangelia Fragkiadoulaki,
Groningen, June 2005
3
Ma M an na ag ge em me e nt n t S Su um mm ma a ry r y
Over the past few years, new trends have emerged in the broad field of marketing practice and science. Worldwide, companies have adopted loyalty marketing programs, in order to establish a relationship with their best customers.
In this notion, customer relationship management and loyalty programs have received a great deal of attention from marketers, business people, consultants and academicians.
In this frame, the on hand document is an endeavor to provide a more in depth insight into the factors that strongly influence customers’ perception of and response to loyalty schemes. Analytically, this paper examines whether homogeneous subgroups of customers who participate in loyalty programs can be identified, based on the relative importance they place on decision criteria, such as their preferences towards several aspects of those schemes.
Furthermore, the relationship of the identified subgroups and the socio- demographics of the participants is also being tested.
This project was based on a research conducted by Carlson Marketing Group and the University of Groningen. The sample consists of approximately 4000 respondents who participate in loyalty programs. The data were collected through questionnaires that were sent to customers of Dutch companies via e-mail.
Questions were mainly addressed to elicit information concerning preferences, perceptions, buying behavior and intentions of customers towards loyalty programs.
The findings identify three segments of customers who show similar pattern of
preferences with regard to several aspects of loyalty programs. Moreover, the
profiling of the subgroups on additional variables, such as socio-demographics,
provides distinctive and viable segments. In particular, the first segment is
defined as “Tangible and Intangible Offers and Privacy Concerns”. In this
4
segment participants are interested in all kind of offers a company poses itself or in cooperation with partner companies (e.g. coupons, free products), while they are also interested in intangible offers (e.g. participation in events, information provided through a newsletter or a magazine). They also have great confidential and privacy concerns with regard to the treatment of their personal data. The second segment is named as “Special Treatment”, due to the fact that participants are expecting to receive prominent customer service, while their loyalty feeling is omnipresent. Lastly, the third segment is defined as “Discount, Easiness and Credit Validity”. In this segment, participants are mostly interested in direct discounts. They also prefer an easy and free participation, while they place special concern in the validity of the saved credits.
Taking everything into consideration, the results can be fruitfully applied by
business people and managers, especially in the fields of market segmentation
and target marketing. Additionally it provides valuable guidelines in designing
effective loyalty programs, since it reveals the specific aspects of the programs
that customers are most interested in. Apart from that, the general information
over the effects of a loyalty program, the incentives that stimulate participation,
the possibility to modify consumer behavior can assist in a certain degree the
design and the implementation of successful loyalty schemes, saving in this way
priceless financial and time resources. Nevertheless, additional research is
considered necessary in order to be able to generalize the findings over
alternative industries and to provide opportunities for better customer base
knowledge.
5
Ta T a bl b le e O Of f C C on o nt te en nt ts s
FOREWORD ...2
MANAGEMENT SUMMARY...3
TABLE OF CONTENTS ...5
LIST OF TABLES & FIGURES...6
CHAPTER 1: INTRODUCTION ...7
1.1.PREFACE...7
1.2.PROBLEM STATEMENT &RESEARCH QUESTIONS...7
1.3.ACADEMIC AND BUSINESS RELEVANCY...8
1.4.STRUCTURE OF THE THESIS...9
CHAPTER 2: CUSTOMER LOYALTY...10
2.1.THE CONCEPT OF LOYALTY...10
2.2.IS LOYALTY PROFITABLE?...11
CHAPTER 3: LOYALTY PROGRAMS ...12
3.1.DEFINITION AND OBJECTIVES OF A LOYALTY PROGRAM...12
3.2.HISTORY OF LOYALTY PROGRAMS...13
3.3.THE EFFECTS OF LOYALTY PROGRAMS...14
3.4.DESIGN AND PARTICIPATION IN LOYALTY PROGRAMS...15
3.4.1. Design and implementation of a loyalty program...15
3.4.2. Participation in loyalty programs and the relation with consumer behavior ...17
CHAPTER 4: RESEARCH DESIGN ...21
4.1.INTRODUCTION...21
4.2.DATA AND SAMPLE...21
4.3.QUESTIONNAIRE...21
4.4.ANALYSIS METHOD...23
CHAPTER 5: ANALYSIS AND RESULTS ...25
5.1.BASIC DESCRIPTIVE STATISTICS...25
5.2.CLUSTER ANALYSIS...30
5.2.1. Introduction...30
5.2.2. Decision Process...30
5.2.3. Conclusions of Cluster Analysis...45
5.3.LATENT CLASS ANALYSIS...48
5.3.1. Introduction...48
5.3.2. Decisions and Search Strategy...48
5.3.3. Interpretation ...50
5.3.4. Conclusions of Latent Class Analysis ...62
CHAPTER 6: CONCLUSIONS ...64
CHAPTER 7: MANAGERIAL IMPLICATIONS - LIMITATIONS & ACADEMIC RELEVANCY ...72
REFERENCES ...74
APPENDIX ...78
6
Li L is s t t O Of f T Ta a bl b le es s & & F Fi ig gu ur re e s s
Table 5.1.1. Average Number Of Loyalty And Payment Cards __________________________________ 25 Table 5.1.2. Average Number of Participation Years in Loyalty Programs ________________________ 25 Table 5.1.3. Average Number Of Participants Per Loyalty Program _____________________________ 26 Table 5.1.4. Mean Values For Active Variables (Preferences) __________________________________ 27 Table 5.1.5. Frequencies for Financial Advantages (direct discount) ____________________________ 28 Figure 5.1.1. Histogram Of Special Offers Adjusted To Your Interests And Buying Behavior __________ 29 Table 5.2.2.1. Analysis of Agglomeration Coefficient for Hierarchical Cluster Analysis ______________ 33 Table 5.2.2.2. Cluster Sizes For the Two And the Three Cluster Solutions Of The Hierarchical Cluster Analysis ____________________________________________________________________________ 34 Table 5.2.2.3. Number Of Cases In Each Cluster Using The K-Means Method _____________________ 35 Table 5.2.2.4. Crosstab Between Cases Of The Three Cluster Solution Of Ward Method And K-Means Method_____________________________________________________________________________ 36 Figure 5.2.2.1. Graphical Profile of the Three Cluster Solution of the Nonhierarchical Cluster Analysis_ 37 Table 5.2.2.5. Crosstab Between Cluster Number Of Case And Cluster Number Of Case Using Initials From Calibration Sample ______________________________________________________________ 38 Table 5.2.2.6. Crosstab between Gender and The Three Cluster Solution of the Nonhierarchical Cluster Analysis ____________________________________________________________________________ 39 Table 5.2.2.7. Chi-Square Test of Statistical Significance of Gender _____________________________ 39 Table 5.2.2.8. Crosstab between Category of Age and The Three Cluster Solution of the Nonhierarchical Cluster Analysis______________________________________________________________________ 40 Table 5.2.2.9. Crosstab between Education and The Three Cluster Solution of the Nonhierarchical Cluster Analysis ____________________________________________________________________________ 41 Table 5.2.2.10. Crosstab between Composition of Household and The Three Cluster Solution of the Nonhierarchical Cluster Analysis ________________________________________________________ 42 Table 5.2.2.11. Crosstab between Ownership of the house and The Three Cluster Solution of the
Nonhierarchical Cluster Analysis ________________________________________________________ 43 Table 5.2.2.12. Crosstab between Gross Income of the family and The Three Cluster Solution of the Nonhierarchical Cluster Analysis ________________________________________________________ 43 Table 5.3.2.1. Test Results Using Latent Gold ______________________________________________ 49 Table 5.3.3.1. Parameters Output of Latent Gold (Indicators) __________________________________ 51 Table 5.3.3.2. Parameters Output of Latent Gold (Covariates) _________________________________ 52 Table 5.3.3.3. Profile Output of Latent Gold (Indicators)______________________________________ 54 Table 5.3.3.4. Profile Output of Latent Gold (Covariates) _____________________________________ 55 Figure 5.3.3.1. Profile Plot Output of Latent Gold ___________________________________________ 57 Table 5.3.3.5. ProbMeans Output of Latent Gold ____________________________________________ 58 Table 5.3.3.6. Bivariate Residuals Output of Latent Gold______________________________________ 59 Table 5.3.3.7. Standard Classification Output of Latent Gold __________________________________ 60 Table 5.3.3.8. Covariate Classification Output of Latent Gold _________________________________ 61 Table 6.1. Summary Table of Cluster Analysis Results ________________________________________ 67 Table 6.2. Summary Table of Latent Class Analysis Results ____________________________________ 68
7
CH C HA A PT P TE ER R 1 1 : : I IN NT TR RO OD DU U CT C TI IO ON N
1.1. Preface
The past decade has seen many firms (re)adopt a customer focus – often through a formal program of Customer Relationship Management – CRM (Brown 2000; Kalakota and Robinson 1999; Peppers and Rogers 1997; cited Uncles, Dowling and Hammond, 2003, p. 4). Indeed a pure shifting from a transactional concept to a more relational concept is been noticed in many firms. The relationship marketing concept is perceived as the 360 – degree view of serving customers, since it seeks to create and sustain mutually satisfying long-term relationships not only with customers but also with other key players, such as employees, suppliers, distributors, retailers, the surrounding community, and society as a whole (Hoffman et. al. 2005, p. 8).
In this framework and over the past few years, loyalty programs have become a key tactic of CRM, that has been adopted in a wide variety of industries, varying from airlines and hotels, retailing, banks, grocery stores, car companies, telecommunications, fashion stores, universities and many other areas. The use of loyalty programs has been boomed over the last decade, and nearly 75 percent of shoppers now belong to at least one loyalty program (Young and Stepanek, 2003, p. 51). Furthermore, the importance of loyalty programs has been recognized in both the managerial and the economic modeling literature (Borenstein 1996; Kim, Shi, and Srinivasan 2001; Kopalle and Neslin 2000; cited Kivetz and Simonson 2003, p. 454). Nonetheless, much less research has been undertaken in identifying the factors that influence customers’ perception of and the response to such programs. Obviously this is why some programs are highly successful while others fail.
1.2. Problem Statement & Research Questions
The objective of this paper is summarized to the following statement:
In this respect, several research questions should be addressed in order to fulfill the above mentioned problem statement. In particular the research questions concern:
1. What is a Loyalty Program?
Getting a more in-depth insight into the factors that influence customer’s
perception of and response to loyalty programs by identifying
homogeneous subgroups of loyalty programs’ participants based on
their preferences towards several aspects of these schemes
8
2. What are the main effects of a Loyalty Program?
3. What are the crucial incentives that stimulate participation to these schemes?
4. Does a Loyalty Program affect consumer behavior and how?
5. Is it possible to identify homogeneous subgroups of customers that participate in Loyalty Programs based on their preferences towards several aspects of those schemes?
6. Granting that several homogeneous subgroups of consumers are identified, is there any relationship with the socio-demographics (e.g.
education, age, composition of household etc.)?
The aim of the thesis, thereupon, is focused on answering the above stated research questions.
1.3. Academic and Business Relevancy
An additional word with respect to the relevancy of this paper with the business and the academic fields needs to be added.
Firstly, this paper is based on a research that has been conducted with reference to the attitude of consumers towards loyalty programs. As specifically mentioned in the problem statement, the goal is to get more insight into the customers’
preferences towards loyalty programs. In this survey, approximately, 4000 customers have filled in a questionnaire, attaching different degrees of importance across several aspects of loyalty programs. The analysis of the corresponding dataset reveals homogeneous subgroups of customers that participate in loyalty programs. In this respect, the findings can be useful for managers and business people, since they create opportunities for segmentation of consumers and target marketing. Indeed, segmentation is a major way of operationalizing the marketing concept, while it also provides guidelines for a firm’s marketing strategy and resource allocation among markets and products (Wind, 1978, p. 317). Additionally, this research can be a useful source of information in designing effective loyalty programs, since it documents the aspects of a loyalty scheme that customers mostly count.
On the other hand, there are some evident limitations. Firstly, the study is based
only on one research in the Dutch retail sector. However, loyalty programs also
appear in other industries, such as airlines, lodging (Bell, et. al., 2002, cited
Leenheer and Bijmolt, 2003, p. 105), and financial services (Verhoef, Frances,
and Hoekstra, 2001, cited Leenheer and Bijmolt, 2003, p. 105). Given the
differences in market structure and product type, the drivers and the
effectiveness of loyalty programs could be different in these industries (Leenheer
and Bijmolt, 2003, p. 105). Furthermore, an analysis on the customer-specific
purchase behavior has not been conducted. However, such analysis could
9
improve the knowledge of the customer base or could also be the basis for the identification of different homogeneous subgroups of customers.
As far as the relevancy with the academic profile is concerned, literature extensively recognizes the importance of loyalty programs (Kivetz and Simonson 2003, p. 454). In the recent years there has been a rise in the academic interest in measuring loyalty (Sharp and Sharp, 1997, p. 473). However, as mentioned in the preface, little is known about the factors that influence customer’s perception of and response to such programs (Kivetz and Simonson, 2003, p. 454). On the other hand, academicians and market researchers often encounter situations best resolved by defining groups of homogeneous objects, whether they are individuals, firms, products or even behaviors (Hair et. al., 1998, p. 469). In this respect, segmentation has become a dominant concept in marketing literature (Wind, 1978, p. 317).
1.4. Structure of the thesis
This paper is organized as follows:
First of all, a discussion on what is meant by the term “customer loyalty” is presented, followed by a thorough review of loyalty programs. In this perspective, the history, the goals and the main effects of loyalty programs are reported.
Next, drawing on these conceptualizations, the design of effective loyalty programs and the incentives that evoke participation to these programs are been discussed, placing special attention to the relationship between loyalty programs and consumer behavior.
In the subsequent part, the research method is described, accompanied by a thorough presentation of the results. In this notion, alternative analysis methods are applied, discussing the similarities and differences between the various approaches and the results obtained.
Finally, conclusions, managerial implications and limitations are reported in
detail.
10
CH C HA A PT P TE ER R 2 2 : : C C US U ST TO OM ME ER R L LO OY YA AL LT TY Y
2.1. The concept of loyalty
The concept of loyalty is not new in marketing. The word “loyalty” is widely used, both in the marketing literature and in the marketing plans of practitioners (Uncles and Laurent, 1997, p. 399). Generally, loyalty is exhibited by consumers towards various things, such as: brands, services, stores, product categories (e.g.
alcohol) and activities (e.g. jogging). The term customer loyalty is used in this paper to stress the feature of people, rather than something inherent in brands which corresponds to brand loyalty.
Unfortunately, there is no universally agreed definition (Jacoby and Chestnut, 1978; Dick and Basu, 1994; Oliver, 1999; cited Uncles, Dowling and Hammond, 2003, p. 5). Instead, as Uncles, Dowling and Hammond showed in their study (2003, p. 5) there are three popular conceptualizations: loyalty as primarily an attitude that sometimes leads to a relationship with the brand; loyalty mainly expressed in terms of revealed behavior (i.e. the pattern of past purchases); and buying moderated by the individual’s characteristics, circumstances, and/or the purchase situation. Yi and Jeon (2003, p. 231) define loyalty as “the repeated purchases of particular products or services during a certain period of time”.
According to Oliver’s definition (1997; cited Yi and Jeon, 2003, p. 231) loyalty is
“the deeply held commitment to rebuy or repatronize a preferred product or service consistently in the future, despite situational influences and marketing efforts having the potential to cause switching behavior”. Personally speaking the latest definition is considered to be more complete and descriptive.
In this respect, many researchers have identified a wide range of loyalty measures where the main distinction is between attitudinal loyalty and behavioral loyalty. Attitudinal loyalty includes cognitive, affective and co native elements (Oliver, 1997; cited Yi and Jeon, 2003, p. 231). On the other hand, behavioral loyalty is more directed towards the measurement of exclusive purchases, hard- core loyalty, repeat purchase probability, share of category requirements, etc. In this paper, the key dependent variable is correlated to behavioral loyalty, since the study mainly deals with the relationship between loyalty programs and purchase behavior.
Furthermore, Yim and Kannan (1999, p. 76) in their study documented two types
of consumer behavioral loyalty: the first is hard-core loyalty, which is defined as
the proportion of a product’s purchases accounted for by its hard-core customers
or customers who almost exclusively repeat purchase the product alternative,
and the second is reinforcing loyalty, which is the proportion of the product’s
purchases accounted for by consumers who may switch among product
alternatives, but predominantly repeat purchase this product alternative to a
significant extent.
11
Apparently, consumers often do not buy only one brand. Their actual behavior of either brand switching or promiscuity leads to a phenomenon called “polygamous loyalty” (Dowling and Uncles, 1997, p. 74). In this respect, share-of-wallet is the most suitable measure for behavioral loyalty, since it actually measures the share of category expenditures spent on purchases at a certain company and integrates choice behavior and transaction sizes during a certain period of time into one single measure (Leenheer, Heerde, Bijmolt and Smiths, 2004, p. 3).
2.2. Is loyalty profitable?
In has been a controversial matter whether loyalty is truly profitable. On the one hand, loyalty advocates support that: it costs less to serve loyal customers; loyal customers pay higher prices for the same bundle of goods; and loyal customers market the company (Reinartz and Kumar, 2002, p. 88). Accordingly, Reichheld and Sasser (1990, p. 106) showed that acquiring a new customer entails certain one–time costs for advertising, promotions and the like, also companies with long-time customers can often charge more for their products or services, while another economic boon from loyal customers is the free advertising they provide.
However, there is no strong evidence that supports the above mentioned claims.
In fact, academic research documents the often poor profitability of long-standing customers in business-to-business industries. Additionally, loyal customers can be more price sensitive than occasional ones, because they expect to get some tangible benefits for their loyalty. Finally, the overall link between customer longevity and the propensity to market by word-of-mouth was not that strong (Reinartz and Kumar, 2002, p. 89).
Taking everything into account, to get strong returns from loyalty, companies need a clearer understanding of the link between loyalty and profits (Reinartz and Kumar, 2002, p. 86). Thus, it is important to find ways to measure the relationship between loyalty and profitability so as to better identify which customers to focus on and which to ignore. Characteristically, Reinartz and Kumar (2002, p. 90) stressed that in order to spot the true apostles, companies need to judge customers by more than just their actions. Nevertheless, there is no one right way to make loyalty profitable. But whatever the context, no company should ever take for granted the idea that managing customers for loyalty is the same as managing them for profits. Only by managing both at the same time, it is possible to strengthen the link between profits and loyalty (Reinartz and Kumar, 2002, p.
94).
12
CH C HA A PT P TE ER R 3 3 : : L LO OY YA AL LT TY Y P PR RO OG GR RA AM MS S
3.1. Definition and Objectives of a Loyalty Program
Loyalty programs are structured marketing efforts which reward and therefore encourage loyal behavior: behavior which is, hopefully, of benefit to the firm (Sharp and Sharp, 1997, p. 474). These programs actually operate differently from other marketing efforts, such as advertising campaigns and sales promotions. Loyalty promotions for instance differ from price promotions in their degree of defensive orientation and their long-term perspective (Yi and Jeon, 2003, p. 232).
Loyalty program adoption is a part of a new emphasis on defensive marketing, which include activities that focus on locking in customers and getting more custom from them, in contrast to activities that focus on winning new customers (Sharp and Sharp, 1997, p. 474). More specifically, Leenheer and Bijmolt (2003, p. 91) reported that the percentage of competitors with loyalty programs, the customer diversity and the strong customer orientation constitute the main reasons for a company to adopt a loyalty program.
Moreover, De Weaver & Partners (cited Young and Stepanek, 2003, p. 53) present six types of loyalty programs that provide different kind of offers to customers: the appreciation (giving customers more of a company’s product/service), the rewards (giving customers rewards unrelated to a company’s product or service), the partnership (marketing to another company’s database and allowing loyal customers to choose their rewards from either company), the rebate (giving customers money back when they buy more), the affinity (building a life-time value relationship with a customer based on mutual interests and not on the use of rewards) and the coalition (teaming up with different companies to share customer data to jointly target a specific customer demographic).
Given the bevy of offerings, loyalty programs should be designed and implemented carefully bearing in mind the following objectives:
a) Maintain the current customer base by retaining the sales levels, margins and profits (defensive outcome), increasing the loyalty and potential value of existing customers (offensive outcome) and inducing cross-product buying by existing customers (Dowling and Uncles, 1997, p. 72).
b) Increase sales revenues by raising purchase/usage levels and/or increasing the range of products bought from the supplier (Uncles, Dowling and Hammond, 2003, p. 4).
c) Dealing with fears of competitive parity by attempting to differentiate a
parity brand, pre-empting the entry of a new (parity) brand and pre-
13
emptying a competitor from introducing a similar loyalty scheme (Dowling and Uncles, 1997, p. 73). Such fears companies rarely state publicly.
d) Operationally a loyalty program should run smoothly both for employees and customers. The latest should also be excited about signing-up and using the program with such sufficient frequency that is becomes second nature to them (Grinnell, 2003, p. 24).
e) Additional peripheral goals are: furthering cross-selling, creating databases, aiding trade relations, assisting brand PR, establishing alliances, etc. (Uncles, Dowling and Hammond, 2003, p. 4).
3.2. History of Loyalty Programs
Loyalty programs that recognize and reward frequent customers have become one of the most commonly used marketing tools for retaining customers and stimulating product or service usage (Kivetz and Simonson, 2002, p. 155). In the 1970s, European researchers studying business-to-business marketing discovered that suppliers and customers can form a win–win arrangement (Dowling and Uncles, 1997, p. 72). That is, the suppliers are having “better”
customers (whom often give them a greater share of their business) and customers are also reported having “better” suppliers. Subsequent research revealed that loyal customers are more profitable to a firm.
Historically, loyalty programs were first introduced in 1981 by the American Airlines. In the early years of the programs, around 1993-1995 there was typically one pioneer retailer in each country: in the UK in was Tesco, in Italy it was Esselunga, while others displayed a ‘wait and see’ attitude. It took approximately five years – depending on the country – for a quick imitation phase to follow (Ziliani, 2003, p. 47).
Nowadays, in some markets card penetration is approaching saturation, while Schneiderman (1998, cited Kivetz and Simonson, 2002, p. 155) reports that nearly half of the U.S. population belongs to at least one loyalty program and that such programs are growing at a rate of approximately 11% a year. Furthermore, information technology that enables firms to practice individual-level marketing has facilitated the spread of loyalty programs into such diverse industries as gaming, financial services and retailing (Deighton, 2000; cited Lewis, 2004, p.
281). Characteristically, as Ziliani (2003, p. 47) stresses in her study: “In 2002 we
found 20 loyalty programs in Europe whose databases count in excess of one
million customer profiles. Six retailers have more than five million customers
profiled in their databases”. With reference to Dutch companies, Albert Hein’s
and Edah’s loyalty programs were reported to belong to the top 20 loyalty
programs in European grocery retailing in year 2002 and respectively in the 10
thand 16
thrank (Ziliani, 2003, p. 48).
14
3.3. The Effects of Loyalty Programs
In conjunction with the second research question, extant empirical research provides mixed evidence of loyalty program effectiveness. Proponents tend to focus on the psychological bonding that eventuates from membership (a customer benefit) and the enhanced customer insights that can be gained from analyzing the program database (a firm benefit) (Brown, 2000; Pearson, 1996; cited Uncles, Dowling and Hammond, 2003, p. 16). Accordingly, a loyalty program can accelerate the loyalty life cycle, by encouraging customers to become business builders by spending more with the company, paying premium prices and bringing in new customers by referrals (O’ Brien and Jones, 1995, p. 75; Dowling and Uncles, 1997, p. 77). Furthermore, loyalty programs that base rewards on cumulative purchasing can indeed enhance retention, since they provide incentives for customers to purchase more frequently and in larger volumes (Lewis, 2004, p. 281).
On the other hand, critics argue that “given the popularity of loyalty programs, they are surprisingly ineffective” (Sharp and Sharp, 1997; cited Lal and Bell, 2003, p. 181). Loyalty – both attitudinal and behavioral – for most customers is quite passive and resembles habit than serious commitment (Uncles, Dowling and Hammond, 2003, p. 16). Furthermore, those schemes do not fundamentally alter the market structure. They might help to protect incumbents and might be regarded as a legitimate part of the marketer’s harmony, but at the cost of increasing marketing expenditures (Dowling and Uncles, 1997, p. 71). What is more, the impact of those schemes might be moderated by customers’ usage levels and their assessments of their service experiences (Bolton, Kannan and Bramlett, 2000, p. 96). With regard to the retention rate of customers, a current study of Bolton, Lemon and Verhoef (2002, p. 35) reveals that the loyalty program (of a financial service organization) was rather effective in increasing cross-buying behavior but it had a relative small effect on customer retention. In addition, they reported that the operational costs of the loyalty program far exceeded its quantifiable monetary benefits (p. 32), which certainly weights negatively the effectiveness of such schemes.
Nevertheless, as stated by Koslowsky (1999; cited Leenheer, Heerde, Bijmolt and Smidts, 2004, p. 1): “While none of these (loyalty) programs result in a perfect world, each can generate that little extra that can provide the retailer market with potential tactical weapons”. Therefore, the aim of a firm should be focused on gaining this “little extra” from the loyalty programs.
Ultimately, the effectiveness of loyalty programs depend on the following
factors: the program rewards and the analysis of customer purchase data
(Leenheer and Bijmolt, 2003, p. 91).
15
Finally, according to O’ Brien and Jones (1995, p. 78) in order to be successful, a loyalty program must become a self-selecting and individually correcting program, through targeting valuable customer segments and discouraging the less valuable customers.
3.4. Design and participation in loyalty programs
Keeping in mind the central problem statement of this paper, a special emphasis should be placed on the literature concerning the design, implementation, participation in loyalty programs, and the relationship with consumer behavior.
3.4.1. Design and implementation of a loyalty program
Often loyalty programs are misunderstood and misapplied. Especially as far as their design and implementation is considered, too many companies treat them as short-term promotions (O’ Brien and Jones, 1995, p. 75) while many others adopt them mostly by fears of competitive parity (Dowling and Uncles, 1997, p. 73).
First of all, it is important to understand the various factors that influence the implementation of a loyalty program. These factors are: the senior management support, the financial evaluation, the employee input and action, the benchmarking and the customer information (Naumann, Jackson and Rosenbaum, 2001, p. 37). The existence of all factors is important in order to correct the potential problem of focusing solely on customers and instead maximize customer value.
Moreover, a loyalty program is a competitive strategy and thus it must meet certain criteria (O’ Brien and Jones, 1995, p. 79), such as: aligning with the company’s capabilities, customers will value the program and checking whether partnering can make the program more competitive.
Further, according to Johnson and Leger (1999, p. 36) generating a positive return on loyalty marketing investment requires developing a customer contact strategy that includes the following four components: selecting the right customers, delivering additional value, nurturing the relationship over time and tracking responses and fine-tuning.
In the same respect, Dowling and Uncles (1997, p. 79) present some useful
guidelines in order to maximize the program’s chances of success and in
particular they mention: design the loyalty program to enhance the value
16
proposition of a product or service, fully cost the loyalty program, design a reward scheme that maximizes the buyer’s motivation to make the next purchase and consider specific market situations in planning.
Grinnell (2003, p. 60) on the other hand, provides some specific rules in designing such programs, which generally focus on testing the concept before making the investment. These include: ensure that loyalty program’s objectives have been clearly defined and measuring what is set out to accomplish, design simple programs and conduct test programs within limited time frames, test programs in a limited geographical area, ensure that the offers ties with the brand position and track both program participants and non participants.
What is more, a straightforward implication of the findings of Kivetz and Simonson (2002, p. 169) regarding the design of loyalty programs is that as the level of program requirements increases, marketers should enhance the relative share of luxury rewards in their “reward mix”. Subsequently, the program can require more efforts or points for luxury rewards than for necessity rewards of the same monetary value.
Specifically, for packaged good brands, Roehm, Pullins and Roehm Jr. (2002, p. 211) suggested that it would be wise to seek incentives that will facilitate increased accessibility and avoid incentives that may introduce interference when designing a loyalty program.
Additionally, the results of the study of Yi and Jeon (2003, p. 238) showed that brand managers need to consider involvement as an important factor in designing a loyalty program. That is because involvement generally moderates the effects of loyalty schemes. Thus, in high involvement situations direct rewards are preferable to indirect rewards and in low involvement situations immediate rewards are more effective in building a program’s value than delayed rewards.
Finally, it is useful to consider the design elements of a loyalty program.
Leenheer and Bijmolt (2003, p. 100) provide a full list of these elements
accompanied with a detailed description, such as: saving feature (saving for
points to obtain reward in the future), promotion feature (discounts on certain
items for all program members), direct mailings (program members receive
direct mails), credit cards (loyalty card can be used as credit card), contests
(lotteries, competitions), demonstrations (trade shows, product instructions,
workshops) and multi-vendor (loyalty program can be used at different
companies).
17
3.4.2. Participation in loyalty programs and the relation with consumer behavior
The central question regarding loyalty programs is whether the members of such schemes become more loyal due to their membership. Obviously, customers are not becoming loyalty program members randomly. Instead, customers will become loyalty program members if they perceive it as worthwhile to do so (Mauri, 2003, p. 13).
Research has shown that consumer preferences are often unstable and ill defined and that consumers construct their evaluations and preferences when faced with the need to make a decision (Bettman, Luce and Payne, 1998;
Payne, Bettman and Johnson, 1992; cited Kivetz and Simonson, 2003, p.
455). Besides that, the difficulty of assessing individual options and outcomes is omnipresent (Bazerman, Loewenstein and White, 1992; Hsee, 1996;
Nowlis and Simonson, 1997; cited Kivetz and Simonson, 2003, p. 455).
However, prior research suggests that consumers seek cues to serve as proxies for the offer’s attractiveness or values to them, in order to evaluate the individual offers and options (Kivetz and Simonson, 2003 p. 455).
With regard to the incentives that stimulate the participation of a consumer in a loyalty program (research question three), substantial empirical studies and researches have been conducted.
Firstly, Leenheer, Heerde, Bijmolt and Smidts (2004, p. 9) document that a customer may compare the expected benefits and costs, in order to decide whether to participate in a company’s loyalty program. The benefits accrue from the saving reward rate (percentage of amount spent that is reimbursed as a saving rewards in money) and the discount rate (percentage direct discount on all purchases) that the program offers. The higher the saving and the discount rates, the higher the likelihood to enroll to the program.
Moreover, a saving feature creates switching costs: members of loyalty programs are losing value when stopping purchasing from the company (Zhang, Krishna and Dhar 2000; cited Leenheer, Heerde, Bijmolt and Smiths, 2004, p. 7).
For packaged goods brands, Roehm, Pullins and Roehm Jr. (2002, p. 203) reported that incentives that overlap with brand associations, which the authors term cue-compatible incentives, can prompt rehearsal that increases the accessibility of favorable brand associations, which, in turn, helps boost post program loyalty. On the other hand, incentives that are tangible or concrete can undermine post program loyalty, due to the fact that elaboration is attracted to the incentive itself and not at the brand.
Furthermore, customer’s loyalty to the store may influence the decision of
participating in a loyalty program. In particular, the proximity to the store
18
(distribution density), price attractiveness and service level (Kahn and McAlister, 1997; cited Leenheer, Heerde, Bijmolt and Smiths, 2004, p. 8) determine the store attractiveness.
In the same respect, household characteristics may influence the participation to a loyalty program. The effects of a loyalty program are expected to be stronger for larger households. Families with higher incomes also tend to have greater bills since they buy more luxurious products. However they might value the monetary rewards from loyalty programs less than families with lower incomes (Leenheer, Heerde, Bijmolt and Smiths, 2004, p. 8).
Furthermore, Leenheer, Heerde, Bijmolt and Smiths (2004, p. 10) also state that there are some additional benefits and costs which influence the decision of participation in a loyalty program, referring to them as instrumental variables. Specifically, the likelihood of enrolling to a particular program increases with an increase of economic benefits and non-economic benefits (psychological and social). In particular the psychological drivers include the feelings of being a preferred or special customer, smart shopper feelings (Kivetz and Simonson, 2002, p. 156) and pride about being economical (Chandon, Wansink and Laurent, 2000; Schindler, 1998; cited Leenheer, Heerde, Bijmolt and Smiths, 2004, p. 5). Whereas, the sociological drivers include the creation of affective commitment with the company, the fundamental human need to belong to groups and the existence of a reciprocity norm: customers evoke obligation towards those who treat them well or provide value (DeWulf, Odekerken-Schoder and Iacobucci, 2001; cited Leenheer, Heerde, Bijmolt and Smiths, 2004, p. 6).
Apart from the economic and non economic benefits, there are the economic costs (the fees for program membership) and the non economic costs (loss of privacy) that lead to a decreased likelihood of loyalty program participation (Leenheer, Heerde, Bijmolt and Smiths, 2004, p. 11). Characteristically, the reluctance to provide with personal information is caused by consumer’s need to control and a negative attitude towards direct marketing (Phelps, D’Souza and Nowak, 2001; cited Leenheer, Heerde, Bijmolt and Smiths, 2004, p. 10).
Finally the latest authors document the usage of different memberships in different industries as an extra instrumental variable, in this notion the more loyalty programs a consumer is engaged in, the more likely is to sign up for a different program in a different industry.
Research also suggests that the relative amount of customer value drives the
number of customers that a brand will acquire. The rewards that a loyalty
program offers are designed to alter this relationship. Dowling and Uncles
(1997, p. 75) proposed that there are three psychological effects that have the
potential to alter normal patterns of behavior. These include:
19
(a) the extent to which loyalty is to the brand (a direct effect) or to the program (an indirect effect); concluding that loyalty programs that directly support the value proposition and positioning of the target product better fit the goals of loyalty marketing
(b) the way buyers value the rewards offered. In this notion O’Briens and Jones (1995; cited Yi and Jeon, 2003, p. 231) suggested five elements of the loyalty program that determine the value of it, these are: the cash value of redemption rewards (cash value), the range of choice of these rewards (redemption choice), the aspirational value of rewards (aspirational value), the perceived likelihood of achieving rewards (relevance) and the scheme’s ease of use (convenience). To this Yi and Jeon (2003, p. 231) added the psychological benefits of belonging to the program and accumulating points
and (c) the effect of timing; suggesting that the more immediate rewards are preferable to delayed rewards and that direct support of the target product’s value proposition increases the chance that the program will build loyalty for the product and not just for the program.
In the value perception concept, additional views can be presented. For instance, Johnson and Leger (1999, p. 37) clearly states that the emphasis is shifted from the price of the product to value-added benefits. Thus, if the program does not deliver additional value then there is no compelling reason for a customer to interact or interplay. In order to deliver this additional value they identified several ways ranging from purely informational communications to point-based rewards programs. Kivetz and Simonson (2002, p. 156) found that luxuries as rewards are better valued than necessities as rewards. According to their point of view, on reason that loyalty programs are attractive may be their ability to increase hedonic, luxury experiences without the high psychological cost of such consumption (the guilty feeling about luxury consumption). Bolton, Kannan and Bramlett (2000, p. 95) suggested that members of loyalty programs perceive that they are getting better quality and service for their price or, in other words, “good value”.
An additional word needs to be presented with regard to the fourth research
question about the ability of such programs to modify consumer behavior. These
programs should definitely be targeted towards consumers who have the
potential to alter their behavior in a way that is desirable to the company. Further,
rewards need to be designed so as to encourage such customers to modify their
behavior (Lal and Bell, 2003, p. 201). Additionally, with regard to consumer
behavior, Ehrenberg, Hammond and Goodhardt (1994, p. 18) found that
consumers are often attracted by a bargain (without perhaps being aware of the
full price of the item in question). But they certainly do not always respond,
especially to unfamiliar brands or products, or probably to advertisements for
them, which apparently reflects the power of branding.
20
Finally, the results of a relative survey concerning consumers’ perceptions and
behavior are considered useful to be provided. In a recent survey held by Maritz
Loyalty Marketing and presented in October 2004 at the U.S. Banker (p. 14) it
was reported that 66 percent of consumers say that discounts are why they
participate in loyalty programs. In collecting rewards most consumers choose gift
certificates. After discounts, consumers say that cash back is the next most
popular participation reason (54 percent), followed by merchandize (49 percent),
gift certificates (34 percent) and travel (29 percent). Reward preferences vary
significantly with income level and age. Younger respondents (73 percent aged
18 to 34) are more likely to favor discounts than older consumer. Discounts also
appeal more to consumers with income levels under $60.000. Moreover, as it
was reported at Credit Union Journal for the same survey, programs that
customize potential rewards based on the individual customer hobbies and
interests hugely increase reward program satisfaction. Further, receiving too
much mail from the company reported to be a crucial reason for customer
defection. Finally, as advised by Sneed, the market development director of
Maritz Loyalty Marketing: “The customer has to be able to believe that they can
earn an appealing reward in a reasonable amount of time”
21
CH C H AP A PT TE ER R 4 4: : R RE ES SE EA AR R CH C H D DE ES SI IG GN N
4.1. Introduction
After reviewing the concepts of customer loyalty and loyalty programs, the definition of a loyalty program, the main effects, the incentives that invoke participation to these schemes and the relation with consumer behavior (research questions 1 to 4) have been addressed in detail. Thereinafter, the sequential research questions (5 and 6) concerning the identification of homogeneous subgroups of loyalty programs’ participants with respect to their preferences towards those schemes and the relationship with the socio-demographics are going to be addressed as well. It is worth to mention here that these questions comprise the core of the problem statement.
This chapter serves to present the research design of this project. In particular, the collected data, the used sample and the corresponding questionnaire are been discussed. Furthermore, the analysis method is reported followed by a presentation of the segmentation bases. Based on this research design the remaining research questions are going to be answered thoroughly in the next chapter.
4.2. Data and Sample
First of all, the data for the analysis were collected through questionnaires that were sent to customers of Dutch companies via e-mail. These companies operate in different business fields. One is a clothing retailer; two are insurance companies, while there is also one gas station and one food products retailer.
Approximately, for each company 10.000 e-mails were sent to their customers.
Thus, about 50.000 questionnaires were sent to be filled in by loyalty programs’
participants.
The sample that has been used in the corresponding analysis, consists of 3922 customers; each of them placing different degree of importance across several aspects of a Loyalty Program. It might be of interest to mention that the collected sample refers to the year 2005.
4.3. Questionnaire
The questionnaire that has been used in order to collect the data consists of
several questions concerning loyalty programs. The first section is about general
questions, such as to select from a list of loyalty programs, those schemes that a
22
customer participates. Additionally, there are questions asking the average number of customer and payment cards that a participant has and/or usually carries with him/her while going shopping.
Next comes the crucial question concerning the preferences of customers towards several aspects of loyalty schemes. Thus, it is important to present this question in detail due to the fact that the answers have been used as the basis for the following analysis. In particular, it is asked by respondents to place a degree of importance (ranging from very unimportant to very important, using a 5-point scale) to specific aspects of a loyalty program, such as: Financial advantages (direct discount), Special offers by the company, Special offers by partner companies, Special offers adjusted to your interests and buying behaviour, Saving program for coupons and/or free products, Possibility to save at several companies, Customer service especially for customer card holders, The feeling to be a stable/loyal customer of the company, Information about services and products by means of a newsletter or magazine, Participation in events, Free participation to the program, The program is simple and demands little energy by your side, Saved credits always remain valid, and Guarantee of privacy; confidential treatment of customer data. These aspects consist the preferences of customers towards loyalty programs and consist the input for the following analysis.
Apart from this, it is asked to report the degree of agreement or disagreement (in a 5-point scale) towards several statements concerning the participation to a loyalty program (e.g. I like participating in loyalty programs or I like saving points or saving stamps). Sequential, respondents are asked to choose between 2 different options of a loyalty program, such as: Saving for presents and/or coupons or Direct discount on the purchase. Moreover, the average number of participation years in a loyalty scheme is asked, followed by several questions related to buying behavior. Additionally, there is a section concerning the use of internet and related technologies.
Finally, general socio-demographic questions are listed. Again, it is important to thorough present these specific questions, due to the reason that the answers are used in the following analysis. First of all, the gender of the respondents is asked. Next, the category of age should be notified in a 6-point scale (24 years old or younger, 25 up to 34 years old, 35 up to 44 years old, 45 up to 54 years old, 55 up to 64 years old and 65 years old or older). Furthermore, the educational level of the respondent should be declared (ranging from LBO, Mavo, Higher General Secondary Education/VWO, MBO, HBO to WO). Next, the composition of the family should be reported (choosing one of the following options: living with parents, living alone, without children, living alone, with children, living together or married without children living at home, and living together or married with children living at home). Additionally, there is a question regarding the ownership of the house (possibility to choose between 2 options:
Bought the house or apartment and Rent an apartment or house). Finally, it is
23
asked to indicate the gross income of the family (in a scale that ranges from Up to 20,000 euro, 20,001 up to 40,000 euro, 40,001 up to 60,000 euro, 60,001 up to 80,000 euro, 80,001 up to 100,000 euro to more than 100,000 euro).
Lastly, it is important to mention that the questionnaire was designed by Carlson Marketing Group and the University of Groningen.
4.4. Analysis Method
Concerning the analysis method, various Cluster Analysis and Latent Class Analysis methods were applied in order to derive the most effective result. In particular, firstly an extensive Cluster Analysis has been implemented, followed by a respectively thorough Latent Class Analysis. In this perspective, SPSS and Latent Gold computer’s programs have been utilized for the analyses.
Before presenting the results it is important to indicate the clustering variables that are used as segmentation bases in the subsequent analysis. Literature documents that some variables are “better” than others as a basis for segmentation (Wind, 1978, p. 320). In this paper preferences for several aspects of a loyalty program are used as the segmentation basis (active variables) which is justified by the following reasons. First, preferences contain sufficient information in order to permit the clustering of the customers (Milligan and Cooper, 1987, p. 331). Secondly, as generally is considered, the most effective bases are the unobserved (subjective) and product/domain specific variables, such as psychographics, benefits, perceptions, elasticities, attributes, preferences, intentions etc. Thus, the preferences that are used as active variables are the following:
1. Financial advantages (direct discount) 2. Special offers by the company
3. Special offers by partner companies
4. Special offers adjusted to your interests and buying behaviour 5. Saving program for coupons and/or free products
6. Possibility to save at several companies
7. Customer service especially for customer card holders 8. The feeling to be a stable/loyal customer of the company
9. Information about services and products by means of a newsletter or magazine
10. Participation in events
11. Free participation to the program
12. The program is simple and demands little energy by your side 13. Saved credits always remain valid
14. Guarantee of privacy; confidential treatment of customer data
24
Finally, socio-demographics are used as descriptors of the clusters. Thus, there are six passive variables:
1. Age 2. Gender
3. Educational level
4. Composition of the household
5. Ownership of the house
6. Gross income
25
C C HA H AP PT TE ER R 5 5: : A AN N AL A LY YS SI IS S A AN N D D R RE ES SU UL LT TS S
This chapter serves to present the analyses conducted and the results obtained in order to identify homogeneous subgroups of customers who participate in loyalty programs based on their preferences towards these schemes and the possible relationship of these subgroups with the socio-demographics. In this notion, the remaining research questions (5 and 6) which consist the cornerstone of the problem statement are addressed thoroughly.
5.1. Basic Descriptive Statistics
Before applying any clustering method some basic descriptive statistics are provided as a first illustration of the data that are analyzed afterwards.
Table 5.1.1. Average Number Of Loyalty And Payment Cards
N
(cases) Mean Std.
Deviation How many customer cards do you own? 3853 4,57 1,888
How many customer cards do you take with you when going shopping?
3835 4,07 1,875
How many payment cards do you take with you when going shopping?
3830 2,71 1,439
Table 5.1.1. shows that customers approximately own 4 to 5 loyalty cards of which 4 they usually carry with them. Additionally, they take approximately 3 payment cards with them when going shopping.
Table 5.1.2. Average Number of Participation Years in Loyalty Programs
Years of Participation Frequency Percent Valid Percent
Cumulative Percent Valid
(cases)
Less than 1 year 222 5,7 5,8 5,8
1 to 3 years 669 17,1 17,6 23,4
3 to 5 years 872 22,2 22,9 46,3
5 to 8 years 991 25,3 26,1 72,4
More than 8 years 1050 26,8 27,6 100,0
Total 3804 97,0 100,0
Missing (cases) 118 3,0
Total (cases) 3922 100,0
26
Table 5.1.2. shows that the majority of the customers (27,6%) is participating in the program for more than 8 years, while the percentages of the customers that participate less than 8 years in a loyalty program are illustrated accordingly.
It might also be of interest to check which program(s) is/are the most popular in terms of participation. Table 5.1.3. shows the accurate number of participants in each of the programs and the corresponding percentages. It is obvious that Air Miles loyalty program accumulates the most participants of the whole range of loyalty programs (2986=22,85%), followed by the AH Bonus card (2390=18,29%) and the Douwe Egberts loyalty program (1325=10,14%).
Table 5.1.3. Average Number Of Participants Per Loyalty Program
Loyalty Program Number of
participants Percentage
Air miles 2986 22,85
AH Bonus kaart 2390 18,29 Douwe Egberts 1325 10,14
Rocks 707 5,41
Hout-Brox 631 4,83
Free Bees 602 4,61
Flying Dutchmen (KLM) 521 3,99 BIJcard (Bijenkorf) 460 3,52
Douglas card 428 3,28
Shell zegels 360 2,76
Esso Golden Tiger Kaart 320 2,45 Edah Spaarkaart 314 2,40 Konmar Klantenkaart 266 2,04 V&D klantenkaart 266 2,04 Esprit e*club card 259 1,98 Fina/Total Power Points 224 1,71 AH Koopzegels 206 1,58 MEXX connect card 179 1,37 Setpoint @vantage card 174 1,33 Duthler Kadokaart 152 1,16 Klant-is-Koning kaart (Hoogvliet/Vomar) 95 0,73
Kluskaart 76 0,58
Savers card (Shoeby Fashion & Jilly Mitch) 69 0,53 Q8 Cadeaukaart 56 0,43
SUM 13066 100
27
Next, some general descriptives of the active variables (preferences) are reported. First, Table 5.1.4. shows that customers overall place more importance to the following active variables: Financial Advantages, Saved credits always remain valid and Guarantee of privacy, since they have greater mean values in comparison to the rest active variables.
Table 5.1.4. Mean Values For Active Variables (Preferences)
Active variables (Preferences) N
(cases) Mean Std.
Deviation Financial advantages (direct discount) 3861 4,14 ,863
Special offers by the company 3850 3,73 ,898
Special offers by partner companies 3816 2,88 1,014
Special offers adjusted to your interests and buying behaviour
3841 3,46 1,061
Saving program for coupons and/or free products 3828 3,46 1,123
Possibility to save at several companies 3826 3,16 1,192
Customer service especially for customer card holders
3812 2,95 1,079
The feeling to be a stable/loyal customer of the company
3818 2,78 1,175
Information about services and products by means of a newsletter or magazine
3824 2,79 1,101
Participation in events 3821 2,42 1,037
Free participation at the program 3835 3,37 1,197
The program is simple and participation demands little energy by your side
3816 3,70 1,051
Saved credits always remain valid 3838 4,21 ,953
Guarantee of privacy; confidential treatment of customer data
3841 4,32 ,950
Valid N (listwise) 3577
28
Additionally, some frequency tables are provided where the distribution of the importance of each variable is shown. For instance, Table 5.1.5. shows that the majority of the customers (46,3%) rate the Financial Advantages variable as an important aspect of participating in a loyalty program, while the 37% regard it as very important, and respectively the 12,7% regard this variable as neutral, the 1,7% as unimportant and the rest 2,2% as very unimportant.
Table 5.1.5. Frequencies for Financial Advantages (direct discount)
Frequency Percent Valid Percent
Cumulative Percent Valid
(cases)
Most unimportant 85 2,2 2,2 2,2
Unimportant 67 1,7 1,7 3,9
Neutral 490 12,5 12,7 16,6
Important 1789 45,6 46,3 63,0
Most important 1430 36,5 37,0 100,0
Total 3861 98,4 100,0
Missing (cases) 61 1,6
Total (cases) 3922 100,0
Finally for illustration purposes histograms are also provided. For example,
Figure 5.1.1. depicts the histogram of the Special offers adjusted to your interests
and buying behavior shows that most of the customers regard this variable as
important, followed by a portion who regards it as neutral to their decision to
participate in a loyalty program.
29
Figure 5.1.1. Histogram Of Special Offers Adjusted To Your Interests And Buying Behavior
**
The frequency tables and histograms of the rest active variables are provided at the appendix.6 5
4 3
2 1
0 2.000
1.500
1.000
500
0
Mean = 3,46 Std. Dev. = 1,061 N (cases) = 3.841
Frequency Very Unimportant Unimportant Neutral Important Very Important
