From Sticking
to Clicking
The antecedents of customers' redemption of mobile coupons vs.
physical coupons.
Sebastiaan Berkman 11152192 June, 22th Final versionMSc. in Business Administration – Digital Business Track Amsterdam Business School, Faculty of Economics and Business University of Amsterdam dr. J. (Jing) Li
Statement of originality
This document is written by Sebastiaan Berkman who declares to take full responsibility for the contents of this document.
“I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.”
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Table of Contents
Abstract 6
Introduction 7
1. Literature Review 9
1.1 Digitizing processes 9
1.2 Adoption of the mobile/digital channel 10
1.3 Physical coupon redemption 12
1.4 Mobile coupon (m-coupon) redemption 13
1.5 Comparison of mobile and physical coupon redemption 14
1.6 Time Sensitivity 15 1.7 Price Consciousness 15 1.8 Brand Loyalty 16 1.9 Personal Innovativeness 16 1.10 Privacy Concerns 17 1.11 Coupon Usage 18
2. Conceptual Model and Hypotheses 19
2.1. Conceptual framework 19
2.2 Hypotheses 20
2. Data and method 23
2.1 Survey Design 23
2.2 Sample Description 24
2.3 Measures 25
2.4 Methodology 32
3. Results 33
3.1 Validity and Reliability 33
3.2 Analysis of correlation 34
3.2 Regression 36
4. Discussion (of results) 42
4.1 Theoretical Implications 42
5.2 Managerial implications 45
5.3 Limitation and future research 46
5. Conclusion 48
References 49
Appendix 1: Questionnaire English 54
List of tables and figures
Tables
Table 1: Factors Influencing Coupon Redemption (Reibstein & Traver, 1982)2) 13 Table 2: Factors Influencing M-Coupon Redemption (Danaher et al., 2015) 14
Table 3: Comparison of influential factors on coupons 14
Table 4: Sample Characteristics (n=201) 25
Table 5: Constructs 26
Table 6: Measurements 28
Table 7: Factor analysis loadings & Cronbach's Alpha scores 33
Table 8: Means, standard deviations, correlation 35
Table 9: Regression 39
Table 10: Result summary 41
Figures
Abstract
This study aims to enrich the knowledge of coupon marketing. In the past, many scholars have investigated this topic. These papers were focussed, among other things, on different factors influencing redemption rate of mobile and physical coupons. However, no study defined the differences between those two types. These differences are important, since digitization is a crucial trend in the contemporary business environment. Companies digitize their coupon platforms, but do not have the academic literature to guide them through this change.
This study aims to answer the research gap as described above: what are the differences in consumers’ redemption intention between mobile and physical coupons? In order to do so, the effects of time sensitivity, price consciousness, brand loyalty, personal innovativeness, privacy concern and past coupon usage on customers’ intention of redeem mobile vs. physical coupons is investigated.
Data was collected by a survey which gathered input from 201 respondents. The survey had two conditions (coupon type: mobile/physical) which the respondents were randomly allocated to. After the data was collected, multiple hierarchical regression was used to measure differences between the two types of coupons. The results show that only the factor ‘personal innovativeness’ showed severe difference in ‘intention to redeem a coupon’ between mobile and physical coupons. It turned out that consumers who score higher on personal innovativeness, are likely to have a higher score in ‘intention to redeem a coupon’ for mobile coupons, than for physical coupons. The other factors had no statistically significant difference in ‘intention to redeem a coupon’ between the two coupon types.
Key words: mobile coupon, m-coupon, physical coupon, coupon, mobile marketing, coupon
Introduction
Consumers love a bargain. The first sight of a 50% off sign is usually enough to get hearts to race and wallets to open. And since most people do not want to miss out on a good deal, coupons are a popular and much-used way to save an extra dime every now and then. In contrast to most other marketing tools, consumers directly benefit from the use of a coupon. 42% of the price conscious consumers save over $30 a week by using coupons (Valassis, 2016). To put this in context: these consumers save over $1,500 a year, which is enough to buy approximately two new iPhones (Apple, 2017).
Traditionally, coupons come in the form of sticker sheets and vouchers, whereas nowadays mobile devices (e.g. smartphones, tablets and smart watches) are more and more used to display coupons. In the past decade, the usage of these mobile devices has increased tremendously (Poushter, 2016). Whereas many retailers used to have coupons based on the traditional (physical) form, many vendors have transferred their coupons into a digital platform which runs on smartphones and tablets (Ieva & Ziliani, 2017). There are many reasons to transfer a physical coupon system into a digital area. Leeflang et al., (2014) explained that getting to know the customer and decreasing costs are two main reasons. As a Dutch marketing manager of a regional gas station explained:
We wanted to get to know our customers, take them out of the anonymity so to speak. Besides that, our traditional sticker sheets costed a lot of money. With digitizing our loyalty system, we saw the opportunity to solve these two problems at the same time. (Hilbrands, 2017)
There is an extensive amount of literature available on the topics of coupon redemption (Ramaswamy & Srinivasan, 1998; Reibstein & Traver, 1982), physical coupons (Inmar, 2014) and mobile coupons (Achadinha, Jama, & Nel, 2014; Gray, 2009; Haselton, 2011) but none of these scholars seem to focus on the transition from physical to mobile and the differences in
redemption behaviour that come with this change. This topic is important, concerning the big trend of digitization emerging in the business world of today (Leeflang et al., 2014). The digitization of coupons creates new factors that are important to consider when assessing the impact on coupon redemption. Privacy concern and personal innovativeness, for example, are factors which do not apply for physical coupons, but can be expected to influence mobile coupon redemption.
The research objective of this thesis is to assess the different factors influencing the transition in digitizing a loyalty system. The research direction for this paper is the difference created by the change from physical to mobile coupons:
1. Which factors influence the intention of consumers to redeem a coupon?
2. How do the factors influencing consumers’ coupon redemption differ between physical and mobile coupons?
The available literature on this topic is discussed in the first chapter. After that, the basis for this research, data and methodology, will be discussed in chapter 2. The analysis of the data, via multiple hierarchical regression is discussed in chapter 3, and subsequently it’s results in chapter 4. Finally, chapter 5 concludes this thesis and will answer and discuss the research question.
1. Literature Review
1.1 Digitizing processes
One of the biggest challenges for companies nowadays is successfully coping with the increasing importance of digital business. As stated by Leeflang et al. (2014), the increasing usage of internet across the globe made ‘digital’ an important competitive advantage. Being able to follow customers on their journey (Leeflang et al., 2014), optimizing the marketing ROI measurements (Pauwels, 2015) and improved communication with consumers via multiple channels (Labrecque et al., 2013) are a few of the many changes in the business environment that came with the digitization of our society. Because of this phenomenon, many companies digitize processes or products for a wide variety of reasons (Clancy, 2014). One could think of process optimization, increased customer satisfaction and product quality improvement (Leeflang et al., 2014).
Loyalty programs and coupons are subject to this change as well, whereas lots of companies change their hard-copy sticker sheets into a mobile app or some other form of a digital platform. The reason to do this can vary, but the article of Leeflang et al. (2014) suggests that the opportunity to generate insights about customers, such as who they are and what characteristics they have, is one of the biggest reasons to turn company activities digital. Besides the increasing customer insights, marketing accountability (Verhoef & Leeflang, 2011) and the management of the brand health are two challenges which are also defined as important by Leeflang et al. (2014).
Besides the challenges within the digital channel itself, the increase in number of channels and different forms of media are complicating the business environment even further. Multichannel customer management helps companies to cope with these challenges.
1.2 Adoption of the mobile/digital channel
Multichannel customer management (MCM) is defined by Neslin et al. (2006) as: “the design, deployment, coordination, and evaluation of channels to enhance customer value through effective customer acquisition, retention, and development.” A key point in this definition is the emphasis on ‘value creation through new channels’. With the addition of these new channels, the number of ways to communicate with customers is rising. The increasing portfolio of channels addresses new problems and strategic choices for companies to make in the near future. Besides the question which message to send, the choice of channel is a new issue, which is becoming more and more important these days. Choosing which channel (not) to use is one of the key issues in multichannel management (Neslin & Shankar, 2009). Neslin & Shankar (2009) address 13 key MCM-issues in their paper, concluding with 5 steps that help managers identify and develop a multichannel strategy (identifying customer, developing MCM-strategy, design channels, implement and evaluate). These ‘guidelines’ can help to determine important issues for the best coupon strategy.
For the management of coupons, it is most important to decide which channels to use. Most of the firms are currently employing the physical channel or are transitioning from physical to digital. Neslin & Shankar (2009) address one crucial consideration when firms choose their channels: ‘cannibalization versus synergy’. In other words: do my channels help or harm each other? Within the B2C context, which includes coupons, cannibalization is not a big issue. Synergy, on the other hand, is key in a B2C context. Marketing efforts in a particular channel can really increase the performance of other channels. As for coupons, the literature of Neslin & Shankar (2009) suggest that it would be wise to implement both the physical and the digital channel in order to maximize this synergy effect.
Traditional brick and mortar retailers are more and more transforming their coupons from hard-copy sticker sheets into digital saving applications used on mobile devices (Ieva & Ziliani, 2017). The phenomenon of a certain business channel changing medium is not new. The shift from traditionally offline business channels into online channels has been performed on different levels and in a variety of business categories throughout the past decade. Konuş, Neslin, & Verhoef (2014) conducted a research to measure the effect of the elimination and replacement of channels. After conducting an experiment in cooperation with a Dutch retailer, where half of the sample group received an online catalogue and the other half were sent a hardcopy for 28 months, Konuş et al. (2014) concluded that when the, usually quite costly to maintain, hardcopy channel gets eliminated, managers should expect a loss in revenue. In their research, the savings in costs by eliminating the hard copy catalogue cover the decrease in revenue. As for coupons, this result depends on the current costs of physical coupons a company has. Besides that, the extra value of increased customer knowledge due to mobile coupons is not taken into account in the research of Konuş et al. (2014).
When translating the conclusion of Konuş et al. (2014) into the use of coupons, one could suggest the transformation from a hard copy to an online environment will result in a decrease of participants and possibly revenue. In order to minimize this, Ieva & Ziliani (2017) propose that managers should not choose either mobile or hard-copy, but have to consider using both. Since MCM is partly about segmentation (S.A. Neslin & Shankar, 2009), managers should segment the users of their coupons and loyalty programs in hard copy-users and mobile users. To get segments ‘right channelled’, managers should use the right marketing communication and formulate optimization models to prescribe the right customer to the right channel (S.A. Neslin & Shankar, 2009).
1.3 Physical coupon redemption
Using coupons to get a discount in shops is a phenomenon that has been studied many times over the past century. The earliest researches of this topic are from ‘before the war’ and head all the way back to 1935 (Coutant, 1935). In the many years that followed, different scholars have investigated the facets that influence physical coupon redemption. Coupon redemption is defined by Reibstein & Traver (1982) as: “percentage of coupons eventually returned to the manufacturer”.
Since there are multiple types of shoppers, whom all use coupons in a different way, one can imagine that multiple factors influence coupon redemption rate (Dhar & Hoch, 1996). To provide a clear overview of all these influencers, Reibstein & Traver (1982) conducted an meta-analysis to cluster every important influencing factor on coupon redemption that had been researched until then. The paper concluded 22 factors which influenced physical coupon redemption. An overview of these factors is shown in table 1.
From all these factors, ‘method of distribution’ has received the most attention and was considered the biggest factor to influence physical coupon redemption. After developing a model to determine the relative value of each factor influencing coupon redemption, Reibstein & Traver (1982) concluded that higher face value (perceived value of a coupon) coupons and in-pack coupons generate a higher redemption rate. As a conclusion of their research, Reibstein & Traver (1982) completed a model to forecast the redemption rate of a certain coupon. This model consists of 4 normalized variables (coupon value, market share, intensity and discount rate) and 6 dummy variables (methods of distribution). Among other factors, face value is one of the main predictors of coupon redemption rate. This model explained 91,9% of the variance of coupon redemption, and thus reduces the managerial uncertainty about this topic substantially (Reibstein & Traver, 1982).
Bawa & Shoemaker (1987) agree with this vision. In their research on the effect of a direct mail coupon on brand choice behaviour, they state that face value is an important factor of physical coupons: “many of the findings relate to the deals in general and not to the coupons specifically” (Bawa & Shoemaker, 1987).
1.4 Mobile coupon (m-coupon) redemption
In the past decade, multiple studies have been conducted concerning factors that influence redemption of mobile coupons. Danaher et al. (2015) reviewed the literature on this topic up on to 2015 in order to make a consumer response model to find factors that influence m-coupon redemption. The study concluded 10 factors which might influence m-coupon redemption, which can be seen in table 2.
After checking this model while conducting a two-year field research in a large shopping mall, Danaher et al. (2015) conclude that m-coupon redemption increases when the face value of a coupon is higher. In this particular factor, m-coupons do not differ from physical coupons Reibstein & Traver (1982). Besides face value, time and place attributes are an important factor of m-coupons. Consumers are more likely to redeem an m-coupon when they are close to the store and the coupon is about to expire. Overall, the study of Danaher et al. (2015) concludes that 5 factors significantly influence m-coupon redemption: face value, intercept distance, product type, consumers’ redemption history and expiry date.
Table 1: Factors Influencing Coupon Redemption (Reibstein & Traver, 1982)2) 1. Method of distribution 12. Area of country
2. Product class size 13. Competitive activity 3. Audience reached by coupon 14. Size of coupon drop
4. Consumer's "need" for product 15. Size of purchase required for redemption 5. Brand's consumer market share 16. Level of general support promotion 6. Degree of brand loyalty 17. Consumer attitude
7. Brand's retail availability/distribution 18. Time since the coupons were distributed 8. Face (monetary) value of coupon 19. Growth trend
9. Whether new or established brand 20. Timing, seasonal influences 10. Design and appeal of coupon ad 21. Demographics
Table 2: Factors Influencing M-Coupon Redemption (Danaher et al., 2015) 1. Distance (horizontal & vertical) 6. Day of week
2. Face Value 7. Coupon price format
3. Product Category 8. Expiry length
4. Coupon order 9. Consumer redemption history
5. Time of day 10. Consumer coupon collection history
1.5 Comparison of mobile and physical coupon redemption
Table 3 shows a comparison of the significant factors influencing m-coupon redemption provided by the research of Danaher et al. (2015) and the significant factors influencing coupon redemption which Reibstein & Traver (1982) concluded in their paper.
By reviewing table 3, it is clear that there are a lot of differences between the factors that influence mobile and physical coupon redemption. Face value and timing are the only two factors mentioned in the literature on both mobile and physical coupons. Since the two types of coupons owe their existence to the value they create through a discount, this overlap was to be expected. The remaining factors concerning mobile coupons are more consumer or product related whereas the factors influencing the redemption of physical coupons are overall more focussed on the coupon itself.
Table 3: Comparison of influential factors on coupons (Danaher et al., 2015)* (Reibstein &
Traver, 1982)**
Factor Mobile coupon* Physical coupon**
1. Face Value X X
2. Intercept distance X
3. Product type X
4. Consumers’ redemption history X
5. Expiry date/time since distribution X X
6. Distribution method X
7. Class size X
8. Availability X
9. Discount X
10. Size of coupon drop X
1.6 Time Sensitivity
As can be concluded from the literature of both Danaher et al. (2015) and Reibstein & Traver (1982), timing is a factor which influences the redemption rate of coupons. This generally happens in two ways:
1. The timing of delivery of the coupon significantly influences the redemption rate of coupons. This is an influence factor which will differ per consumer and also per category type. For example Baker, Zheng, & Xueming (2014) stated in their study that advertisements for utilitarian products are more likely to succeed in the morning hours, while hedonic products will be more successful in the afternoon. Since this form of time sensitivity is so reliant on product context, it is not part of this research plan. 2. Besides the timing of the coupon drop, redemption times are an important time factor
as well. The redemption times for mobile coupons are generally shorter than physical coupons (Danaher et al., 2015). The time urgency will influence people who are time sensitive to redeem a mobile coupon faster. The urgency of time, or time pressure is generally described as: “an urgency to finish a certain task or accomplish a certain goal, often generating feelings of anxiety, haste and hurry (Saraiva & Iglesias, 2015; Szollos, 2009).” With shorter redemption times, retailers obviously try to influence people and use negative feelings like anxiety and hurry to their advantage. Consumers who are very receptive for this type of stress due to time limits, will probably be more likely to redeem such a coupon and therefore score higher in redemption intention.
1.7 Price Consciousness
As mention before, face value factor which is extensively mentioned in the literature on both redemption of mobile and physical coupons. Since coupons are all about discount, is was to be expected that the effective value of a coupon is a big influence factor on the redemption rates.
Danaher et al. (2015) states that face value “dominates m-coupon effectiveness”, and Reibstein & Traver (1982) even conclude that ‘face value’ is the biggest influential factor of the redemption rates, even twice as big as the next influencer.
Face value of coupons, and its predictive value for the redemption rate is clearly related to consumers who consider price as an important driver of their purchases. In more general literature, this is perceived as ‘price consciousness’ and defined by Lichtenstein, Ridgway, & Netemeyer (1993) as: “the degree to which the consumer focusses exclusively on paying low prices.”
1.8 Brand Loyalty
Brand loyalty can be seen as a combination of both qualitative and quantitative factors concerning the repetition in the purchase of a particular brand (Tucker, 1964). In other words, the degree of loyalty which a consumer experiences with a particular brand depends on the reoccurrence of purchases and/or the connection one has with this brand. Whether the degree of brand loyalty is based on a qualitative (connection) or quantitative (repetition in purchase) factor is based on the product type of the brand (Peckham, 1963; Tucker, 1964).
The loyalty toward a certain brand is perceived as an influence factor on the redemption rates of coupons. Neslin & Shoemaker (1983) consider brand loyalty as one of the three main items of customer response. Besides the acceleration of product category purchases and repeat purchase effects, brand loyalty is one of the major parameters is measuring the response of consumers.
1.9 Personal Innovativeness
To understand the role of personal innovativeness, a small disclaimer is necessary. Whereas a person as such can be innovative in a lot of different areas and environments, the construct of personal innovativeness in this paper is focussed on the IT domain, more specifically in mobile
devices. In other words: the degree of innovativeness one has with ‘mobile’ products such as smartphones and tablets.
Personal innovativeness is a measure which helps to identify individuals who will most likely adopt new developments earlier than others and is defined as “the willingness of an individual to try out any new information technology (Agarwal & Prasad, 1998)”.
1.10 Privacy Concerns
When implementing a digital channel, a lot of issues have to be addressed for both the company’s stake as for the consumers’. Privacy is one of the major issues of concern when turning a traditional, physical channel digital.
Privacy is a phenomenon that has been studied a lot of times in the past. Because of this extensive research, many definitions have been provided which go from as vague as “the right to be left alone” (Warren & Brandeis, 1891) all the way to “the control over information disclosure and unwanted intrusions into the consumer’s environment” (Goodwin, 1991).
A basic need for intimacy and psychological respite are the main reasons people value privacy. Besides this, protection from the control of others and social influence are factors which influence the need for privacy (Acquisti, Brandimarte, & Loewenstein, 2015). These factors are grouped in the self-determination theory of Ryan & Deci (2006). This theory states that everyone has a basic need for autonomy and wants to be ‘in control of their own life’.
The different theories concerning privacy, have ‘the right of controlling your own life’ as a big shared value. Apparently, privacy is all about consumers wanting to determine their own lives and choices.
1.11 Coupon Usage
Past behaviour as a predictor of behaviour in the future is a popular subject in social psychology. In a summary of the research on predictors of repeated behaviour, Aarts, Verplanken, & van Knippenberg (1998) conclude that regularly performed behaviour is in many times a ‘force of habits’. The classic attitude behaviour model states that a person is expected to repeat a certain behaviour if their attitude towards this behaviour is positive (Ajzen & Fishbein, 1975). In other words: people like to relive positive experiences.
2. Conceptual Model and Hypotheses
2.1. Conceptual framework
The literature as summarized in the first chapter concludes in the conceptual model of figure 1. The model basically consists of two parts: de direct relationships between the independent variables and redemption intention & the moderated relationships between those two.
2.2 Hypotheses
The literature suggests that time sensitivity is an important factor which influences redemption intention of both mobile and physical coupons. It has already been discussed that mobile coupons tend to have shorter redemption time than physical coupons (Danaher et al., 2015). Consumers who score high on time sensitivity will therefore have higher redemption intention as such, and even higher for mobile than for physical coupons. Based on these assumptions, hypothesis 1 and 2 have been made:
H1: Time sensitivity has a positive effect on redemption intention.
H2: The effect of time sensitivity on redemption intention is stronger for mobile coupon than
physical coupon.
Coupons are all about saving money, as was described in the introduction and literature review of this thesis. Based on the coupon-related theory about face value (Danaher et al., 2015; Reibstein & Traver, 1982) and the definition of price consciousness (Ofir, 2004) it is to be expected that consumers who consider themselves price conscious, will make a bigger effort to find the lowest price for their purchases and therefor have a redemption intention than people who do not, or less, consider themselves price conscious. The literature does not give cause for a reasonable indication that this effect would be higher for either mobile or physical coupons. Based on this assumption, hypothesis three and four has been made.
H3: Price Consciousness has a positive effect on redemption intention.
H4a: The effect of price consciousness on redemption intention is stronger for mobile than
for physical coupon.
H4b: The effect of price consciousness on redemption intention is stronger for physical than
As defined earlier, Neslin & Shoemaker (1983) define brand loyalty as one of the three main triggers for customer response.
Since sales numbers are, obviously, the optimal form of customer response it is to be expected that redemption intention is influenced by brand loyalty. The degree of brand loyalty of a consumer as a predictor of coupon redemption rate, is only mentioned in the literature about physical coupon redemption (Reibstein & Traver, 1982). Apparently, mobile coupons are more used to discover new brands and companies, whereas physical coupons serve to minimize the costs of familiar products and brands. In order to test this assumption, hypothesis 5 and 6 read:
H5: Brand loyalty has a positive effect on redemption intention.
H6: The effect of brand loyalty on redemption intention is stronger for physical than for
mobile coupon.
Based on the definition of personal innovativeness as defined by Agarwal & Prasad (1998): “the willingness of an individual to try out any new information technology”, the assumption is made that the positive relationship between personal innovativeness and intention to redeem a coupon is higher for mobile coupons than for physical coupons. This tends to make sense, since mobile coupons can only be used in combination with mobile devices. Based on this assumption, hypotheses 7 and 8 are as follows:
H7: Personal Innovativeness has a positive effect on redemption intention.
H8: The effect of personal innovativeness on redemption intention is stronger for mobile than
for physical coupon.
In order to use coupons, it is generally required for the user to provide some personal information. Based on this characteristic, it is to be expected that people who value their privacy a lot will score lower on redemption intention. As different studies on mobile advertising or couponing already concluded (Ahmadi & Ghahfarokhi, 2016; Okazaki, Li, &
Hirose, 2009); privacy issues and commercial mobile activities by companies are strongly related. Consumers tend to value their privacy more in these ‘mobile situations’. Based on this theory it is to be expected coupons could suffer from privacy prone consumers, especially for mobile coupons. This concludes in hypothesis 9 and 10:
H9: Privacy Concern has a negative effect on redemption intention.
H10: The effect of privacy concern on redemption intention is stronger for mobile than for
physical coupon.
If the theory of the classic attitude behaviour (Aarts et al., 1998) is translated to the coupon environment, it is to be expected that the more coupons a consumer redeemed in the past, the more likely he will be to redeem coupons in the future. Since a repetition of behaviour, as stated in the theory of Ajzen & Fishbein (1975), will increase the attitude towards this behaviour it is expected that redemption intention will also increase. The literature does not give cause for a reasonable indication that this effect would be higher for either mobile or physical coupons. Therefore, hypothesis 11 and 12 are as follows:
H11: Coupon usage has a positive effect on redemption intention.
H12a: The effect of coupon usage on redemption intention is stronger for mobile than for
physical coupons.
H12b: The effect of coupon usage on redemption intention is stronger for physical than for
2. Data and method
2.1 Survey Design
A deductive research approach will be used to answer the research question as proposed in the introduction. In order to obtain the necessary data points, a survey has been distributed among a sample of consumers living in The Netherlands. In order to ensure the reliability of the research, the survey has been distributed digitally. Since the respondents will be able to fill in the survey at their own preferred time and place, he or she will not be influenced by the researcher watching while the respondents fill in the questionnaire. This eliminates the researchers bias and error, but may enlarge a potential participants error or bias since he or she might not be aware of certain factors which alters response (Saunders, Lewis, & Thornhill, 2012, p. 192). Since it unrealistic to force every respondent to fill in the questionnaire in a neutral space with little distraction, participants were asked to participate fully focused for the few minutes it takes to complete the questionnaire. See the appendix for the complete questionnaire.
During the design of the questionnaire, construct validity has been an important factor. Since most of the respondents live in The Netherlands, the survey was translated from English to Dutch. In order to ensure the validity of both the source and the target questionnaire, the survey has been translated by back-translation (Saunders et al., 2012, p. 442). A second reader translated the survey from English into Dutch. A third-person did this core vice-versa. After comparing the first and third version, no major differences in translation came up. The analysis of the data has been done with the statistical software package IBM SPSS statistics, version 23 for mac.
2.2 Sample Description
In order to obtain the necessary data for this research, a survey is distributed via non-probability convenience sampling. The sample of respondents that filled in the questionnaire consists consumers living in The Netherlands (n=201). After one week, a valid number of people filled in the questionnaire. It has been distributed via social media (n=108), personal e-mails (n=89) and other anonymous channels (n=4).
Besides the questions measuring the main variables, the questionnaire contained some questions concerning the characteristics of the respondents. Every participant was asked about their age, gender (male, female, other or ‘do not wish to specify’), income and education. Because of the ongoing discussion about gender and the importance of defining more types than just male/female, the options ‘other’ and ‘do not wish to specify’ were added. Since nobody picked either one of these extra options, they are removed from the dataset during the analysis.
A majority of the participants is female (60,7%), younger than 30 (54,7%), and/or has a bachelor’s degree or higher (74,6%). It is hard to define the sample on income, since 25% of the respondents did not (want to) share their net income. The response for this particular question would probably have been higher, if the answer options contained groups instead of point blank asking for a salary. Full details of the sample can be seen in table 4.
The sample was randomly distributed into two groups of nigh equal size (100 vs. 101). In between the questions about the independent variable constructs and the control variables, participants answered questions about either physical or mobile coupons. They were asked about their previous coupon usage and redemption intention. Based on the data, coupon usage and type of coupon are not significantly correlated (Independent Samples Mann-Whitney U-test’s p= .60).
Table 4: Sample Characteristics (n=201) Gender Male 78 (38.8%) Female 122 (60.7%) Age 0-20 22 (10.9%) 21-30 88 (43.8%) 31-40 17 (8.5%) 41-50 44 (21.9%) 51-60 23 (11.4%) 61-70 5 (2.5%) 71- 1 (0.5%)
Educational level High School 15 (7.5%)
MBO 34 (16.9%) HBO/WO Bachelor 111 (55.2%) Master/PhD 39 (19.4%) Income (€) 0-500 38 (18.9%) 501-1000 25 (12.4%) 1001-2000 53 (26.4%) 2001-3000 19 (9.5%) 3001-4000 10 (5.0%) 4000+ 4 (2.0%)
2.3 Measures
As mentioned before, all the constructs used in this research are based on high quality literature of scholars who already have investigated the subject. As shown in the conceptual model, this research consists of 4 control, 5 independent, 1 dependent and 1 moderator variable. Table 5 contains a full overview of every constructs, its items and the reference. Most of the constructs are measures on a 7-point Likert scale, with 1 being totally disagree and 7 being totally agree. A complete overview of the different measurement scales is summarized in table 6.
Table 5.1: Constructs
Construct Items Reference
Time sensitivity 1. I always seem to be rushed/pressed for time. (Sen, King, & Shaw, 2014)
2. I am time poor; I never have enough time. (Sen et al., 2014)
3. I feel like I never have a day off. (Gray, 2009)
Price Consciousness 1. I check out the prices in more than one food store/supermarket in order to find the cheap prices.
(Ofir, 2004) 2. I am not willing to invest special effort to find cheap prices. (Ofir, 2004)
3. I shop in more than one store to find cheap products. (Ofir, 2004)
4. The time I spend seeking out cheaper products is worthwhile. (Ofir, 2004) 5. In general, the saving achieved by finding cheaper prices is not worth the time and
effort.
(Ofir, 2004)
Brand Loyalty 1. After I get used to a brand, I hate to switch. (Baumgartner & Steenkamp, 2006) 2. When another brand is on sale, I generally purchase it rather than my usual brand. (Baumgartner & Steenkamp, 2006) 3. I feel really committed to the brands I buy. (Baumgartner & Steenkamp, 2006) 4. If my preferred brand were not available at the store, it would make little difference to
me if I had to choose another brand.
(Baumgartner & Steenkamp, 2006)
Personal Innovativeness 1. If I heard about a new information technology, I would look for ways to experiment with it.
(Ha & Im, 2014) 2. Among my peers, I am usually the first to try out new information technologies. (Ha & Im, 2014) 3. I like to experiment with new information technologies. (Ha & Im, 2014)
Table 5.2: constructs
Privacy Concern 1. I am concerned that the information I submit on the Internet could be misused. (Dinev & Hart, 2006) 2. I am concerned that a person can find private information about me on the Internet. (Dinev & Hart, 2006) 3. I am concerned about submitting information on the Internet, because of what others
might do with it.
(Dinev & Hart, 2006) 4. I am concerned about submitting information on the Internet, because it could be used
in a way I did not foresee.
(Dinev & Hart, 2006)
Intention 1. I intend to use a coupon within the next 30 days. (Pavlou & Fygenson, 2006) 2. I plan to use a coupon within the next 30 days. (Pavlou & Fygenson, 2006)
Table 6: Measurements
Variable Measure
Time sensitivity (TiSe) Strongly disagree = 1, strongly agree = 7 Price Consciousness (PriCo) Strongly disagree = 1, strongly agree = 7 Brand Loyalty (BrLo) Strongly disagree = 1, strongly agree = 7 Personal Innovativeness (PeIn) Strongly disagree = 1, strongly agree = 7 Privacy Concern (PrCo) Strongly disagree = 1, strongly agree = 7 Intention (In) Strongly disagree = 1, strongly agree = 7
Gender (GEN) Male = 0, female = 1
Age (AGE) #*
Education (EDU) Primary School = 1, High School = 2, MBO = 3, Bachelor's Degree (HBO/WO) = 4,
Master's/Phd = 5.
Income (INC) #*
Coupon usage (Physical/Mobile) (CoUs) #*
Coupon Type (CoTy) Physical = 0, Mobile = 1.
* Rating scale ‘#’ means that the respondents were asked to answer the question with a
2.3.1 Time Sensitivity
In order to measure perceived time sensitivity, a combination of the work of Sen, King & Shaw (2014) was used. The measure consists of 3 items, who were not reverse coded. Since the Cronbach’s alpha of the original measure was quite high, no reverse coding is added to the measure. To measure this construct, the 3 items had to be answered by a 7-point Likert scale, ranging from ‘strongly disagree’ to ‘strongly agree’.
2.3.2 Price Consciousness
The work of Ofir (2004) was used to measure the construct ‘price consciousness’. In his original work, Ofir (2004) defined six items to measure this construct. Since the amount of total questions to be asked was relatively high and one item raised substantial questions during the pre-test, only five items were used. In order to raise reliability of the answers, 2 items were reversed coded. To measure this construct, the items had to be answered by a 7-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree’.
2.3.3 Brand Loyalty
Brand loyalty is the third construct to be measured in the survey. Four items of the work of Baumgartner & Steenkamp (2006) were used to measure this construct. One item was deleted from the definite survey, since the translation in Dutch raised a lot of questions during the pre-test. In order to raise reliability of the answers, 2 items were reversed coded. To measure this construct, the items had to be answered by a 7-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree’.
2.3.4 Personal Innovativeness
To measure personal innovativeness, the construct as defined by Ha & Im (2014) were used. The original study uses 3 items to measure this variable. Since the Cronbach’s Alpha is high, the three original items, were maintained. To measure this construct, the items had to be answered by a 7-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree’.
2.3.5 Privacy Concern
After reviewing the many options to measure privacy concern, the work of Dinev & Hart (2006) has been used as the source for this construct. Besides the high value of the Cronbach’s alpha, this construct seemed highly appropriate due to the ‘internet angle’ which this construct shares with this research. To measure this construct, the items had to be answered by a 7-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree’.
2.3.6 Coupon usage
In order to measure the previous behaviour of the participants concerning coupon usage, a construct was used from the work of Xu & Teo (2010). In their research, they ask participants to indicate their prior experience in using mobile applications for the past 6 months. This construct is changed into the participant’s prior experience with coupons for the past 6 months, in order to measure the construct ‘coupon usage’.
2.3.7 Coupon Type
The research question clearly states that the difference between ‘mobile’ and ‘physical’ is a major part of this research. In order to clearly assess what the influence of the medium is, coupon type is used as a moderator. Practically, this means that half of the respondents will
based on a physical variant. The mediator and dependent variable will be influenced by this random distribution of participants.
2.3.8 Redemption intention
The independent variables will influence the dependent variable redemption intention. This variable is rooted in the work of Pavlou & Fygenson (2006). To measure this construct, the items had to be answered by a 7-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree’.
2.3.9 Control Variables
Five control variables (gender, age, income, educational level and coupon usage) were used to describe the sample and to explain a certain part of the variance in the model. Concerning gender, respondents had to choose one out of 4 options: male, female, other or do not wish to specify. Even though the choice between just ‘male’ and ‘female’ is more standard and easier to analyse in SPSS, the addition of the other 2 option was considered very important due to the ongoing discussion about gender classification. Since it is to be expected that no respondent will pick either option 3 or 4, male/female is recoded as 0 and 1. These control variables were included in the last section of the survey.
2.4 Methodology
In order to ensure the reliability of the data, an explanatory factor analysis was conducted. This is important to do, in order to be sure that the items really represent the constructs as defined. After that, a correlation analysis was performed to check whether the independent variables do not correlate with each other. This is important to debar multicollinearity. After these tests to check the reliability of the data, multiple regression was performed on the dataset to check whether the hypotheses can be supported or not. This analysis was performed in 4 steps: first, the impact of the independent variable ‘redemption intention’ on its own was measured. Secondly, the control variables were added. To measure the direct relationships, the independent variables and the moderator were added. Finally, in step 4, the moderated effects were added to conclude the entire model. During this analysis, the model fit (r-square and AIC) were measured as well as the regression indicators of each proposed relation in the conceptual model. The results of these analyses are shown in chapter 3.
3. Results
3.1 Validity and Reliability
In order to ensure that the different constructs actually measure what they are supposed to measure, a factor analysis and a reliability analysis (model: alpha) were conducted. The results of these tests are shown in table 7.
Table 7: factor analysis loadings & Cronbach's Alpha scores
Factor Loadings Items Privacy Concern Price Consciousness Personal Innovativeness Time Sensitivity Brand Loyalty Privacy Concern 1 .827 Privacy Concern 2 .884 Privacy Concern 3 .931 Privacy Concern 4 .874 Price Consciousness 1 .616 Price Consciousness 3 .731 Price Consciousness 4 .756 Price Consciousness 2 .575 Price Consciousness 5 .730 Personal Innovativeness 1 .881 Personal Innovativeness 2 .840 Personal Innovativeness 3 .837 Time Sensitivity 1 .806 Time Sensitivity 2 .839 Time Sensitivity 3 .605 Brand Loyalty 1 .712 Brand Loyalty 3 .681 Brand Loyalty 2 .559 Brand Loyalty 4 .615 Eigenvalues 3.746 2.981 2.452 2.333 1.839 % of variance 19.71 15.69 12.90 12.28 9.68 Cronbach's Alpha .931 .810 .889 .784 .733
The factor loadings define the different factors which are measured in this survey. Based on the Eigenvalues, this dataset contains 5 factors. The 19 items are distributed across these factors, based on their rotated factor loadings. After matching the different items to the factors, it can be concluded that no items need to be extracted since every factor contains items with a relatively high factor which is minimal twice as big as the other factor loadings for a particular item (Saunders et al., 2012).
After this, the scale means of the different items belonging together were calculated. The rest of the analysis is done with these average scores.
3.2 Analysis of correlation
Table 8 contains an overview of the standard descriptive statistics (mean; M and standard deviation; S) and the correlation between the respective variables. When looking at the descriptive statistics, the high standard deviation of variable 12 (‘income’) stands out. This can be explained by the high positive skewness (1.41) (Field, 2013): a small group of respondents earns a lot more than the majority, while nobody obviously earns less than 0 euros. In the case of coupon usage, this scenario is the same. A few respondents indicated that they used a big number of coupons, while a big group of respondents did not redeem any coupon lately. The control variables ‘income’ and ‘age’ are correlated (r =.546, p <.01). This correlation was to be expected, since older people mostly have more working experience and thus a higher salary. The independent variables do not show a high significant correlation (r > 0.9), so multicollinearity can be excluded (Saunders et al., 2012, p. 524).
Table 8: means, standard deviations, correlation Variables M S 1 2 3 4 5 6 7 8 9 10 11 12 1. Coupon Type*** 50 .501 1 2. Coupon Usage 1.080 2.206 .000 1 3. Time Sensitivity 3.983 1.433 -.009 .110 1 4. Price Consciousness 3.5801 1.235 -.076 .220** .137 1 5. Brand Loyalty 3.9863 1.150 -.114 -.113 .140* -.043 1 6. Personal Innovativeness 3.7413 1.471 .076 .084 .047 .132 .057 1 7. Privacy Concern 4.8035 1.356 .045 .130 .076 .085 .125 .050 1 8. Gender*** 61 .489 .033 .012 .269** .104 .027 -.315** .195** 1
9. Educational Level: High School*** 7.5 .263 -.055 .043 .122 .075 .106 -.006 -.004 .072 1
10. Educational Level: MBO*** 16.9 .376 .029 -.047 -.063 .065 .026 -.137 .036 .116 -.128 1
11. Educational Level: Master/PhD*** 19.4 .396 .040 -.081 -.009 -.070 .047 .144* -.040 .031 -.139* -.221** 1
12. Income 1470 1184 -.031 -.187* -.074 -.172* .108 -.054 .100 -.258** -.190* .053 .095 1 13. Age 33.750 2.206 -.001 -.191** -.197** -.137 .175* -.193** .109 .012 -.007 .254** -.118 .546**
*Correlation is significant at the 0.05 level (2-tailed).
**Correlation is significant at the 0.01 level (2-tailed).
3.2 Regression
Hierarchical multiple regression was performed in order to investigate the effects of coupon usage, time sensitivity, price consciousness, brand loyalty, personal innovativeness and privacy concern on consumers’ intention to redeem a coupon, after controlling for gender, age, income and educational level and taking into account the moderating effect of coupon type. Via this way, the analysis focusses on the main question of this thesis: what is the difference in redemption intention between mobile and physical coupons? The complete results can be seen in table 9.
To explain the model fit of the different steps, both R2 and AIC have been used. R2 is
the incumbent way to describe model fit, and is therefore chosen to represent the variance explained in the different models. But, since R2 always increases when predictor variables are
added (Field, 2013, p. 324) AIC is also included in table 9. The Akaike Information Criterion (AIC) (Akaike, 1987) does not automatically improve if predictor variables are added; it punishes models for having too many variables. This characteristic of AIC makes this criterion highly usable for this thesis, since there is an extensive number of independent variables. AIC is a ‘goodness-of-fit- measure’ which is corrected for the complexity of the model. Table 9 contains a column with the AIC values. The lower the AIC value is, the better the model fit is. Since the sample size is not that large, AIC will be used instead of the more conservative variant BIC (Field, 2013, p. 826). Since AIC is not intrinsically interpretable (Field, 2013, p. 324), the values of the four steps have to be compared to one another. Since the numbers decline, the model fit improves every step and thus with every addition of new variables. Based on the AIC values, step 3 has the biggest model fit and explains the most variance.
Step 1 of the regression analysis includes the control variables age, income, gender and education. This model explains 6.5% of the total variance, and is statistically significant at the
Step 2 covers all of the independent variables, including coupon type and the control variables. The model is statistically significant (p<.01) and explains 25.9% of the variance of the model (adj. R2 =.259) which is an increase of 19.4 percentage points (adj. R2 change
=.194). The AIC also improves to 147.57.
Step 3 includes the moderating effect of coupon type, and represents the entire model as illustrated in the conceptual model in figure 1. Based on the R2, the addition of the moderator
variables made the model statistically insignificant (p>.05). The AIC, however, does improve to 134.354. Considering the usefulness of AIC for this particular thesis, as described above, step 3 is regarded as the best model of this analysis. When considering the different variables and their respective relationships to redemption intention, only 4 of those relationships are statistically significant. Gender is the only control variable which has a statistically significant relationship with redemption intention (p<.05, B=.614). When the variable gender increases with 1 point, redemption intention is expected to increase with .614 points on a 7-point Likert scale. In other words: a female is anticipated to score .614 points higher on redemption intention than a male.
Both coupon usage and brand loyalty have a statistically significant effect on redemption intention. For every extra coupon a consumer has redeemed in the past 6 months, he or she is expected to score .206 points higher on redemption intention (p<.01, B=.206). This result supports hypothesis 11: Coupon usage has a positive effect on redemption intention. The data also generates support for hypothesis 5: Brand loyalty has a positive effect on redemption
intention. For an increase of 1 point on Brand Loyalty, a consumer is expected to score .295
points higher on redemption intention (p<.05, B=.295).
The moderation effect explains the main message of this thesis: the difference in redemption intention between mobile and physical coupons. The relationship between personal innovativeness and redemption intention is statistically significant moderated by coupon type
(p<.05, B=.604). In other words: the effect of personal innovativeness on redemption intention depends on the type of coupon. The results indicate that the relationship between personal innovativeness and redemption intention is only significant in the case of a mobile coupon, since the conditional effect of personal innovativeness on redemption intention is significant for mobile coupons and not for physical coupons. This result is in line with hypothesis 8: The
effect of personal innovativeness on redemption intention is stronger for mobile than for physical coupon.
3.4 Result summary
When summarizing the results of this thesis, 3 out of 12 hypotheses are supported by the data. Table 10 shows a full overview of each hypothesis and whether or not it is supported. Half of the hypotheses (6) were based on a moderating effect, and thus meant to describe the difference in redemption intention between mobile and physical coupons. Only the moderated effect of personal innovativeness on redemption intention was statistically significant, and therefor the only ‘moderation hypothesis’ which can be supported. The direct effect between brand loyalty and redemption intention, and between coupon usage and redemption intention is also statistically significant. These are the only two ‘direct hypotheses’ that can be supported.
Table 9.1: regression step 0-2
Step Variables R2 adj. R2 Change AIC
0 - 0 - -
1 Age .065* .103 182.02
Income
Gender
Educational Level - High School
Educational Level - MBO
Educational Level - Master/PhD
2 Age .259** .194 147.57
Income
Gender
Educational Level - High School
Educational Level - MBO
Educational Level - Master/PhD
Coupon Type Coupon Usage Time Sensitivity Price Consciousness Brand Loyalty Personal Innovativeness Privacy Concern
*Model fit is significant at the 0.05 level (2-tailed). **Model fit is significant at the 0.01 level (2-tailed).
Table 9.2: regression step 3
Variable adj. R2 adj. R2 Change AIC B SE p
Step 3 .264 .005 134.354 constant .378 .961 .695 Age -.005 .013 .679 Income .000 .000 .344 Gender .614 .276 .028 High School .457 .470 .332 MBO .096 .369 .795 Master/PhD .356 .294 .229 Coupon Usage .206 .064 .002 Time Sensitivity .028 .118 .814 Price Consciousness .212 .127 .097 Brand Loyalty .295 .140 .037 Personal Innovativeness -.136 .116 .243 Privacy Concern .095 .119 .425 Coupon Type .263 .229 .252
Coupon Usage * Coupon Type .127 .229 .580
Time Sensitivity * Coupon Type .002 .252 .995
Price Consciousness * Coupon Type -.159 .232 .495
Brand Loyalty * Coupon Type -.018 .251 .942
Personal Innovativeness * Coupon Type .604 .263 .023
Table 10: Result summary
Hypothesis Supported?
1 Time sensitivity has a positive effect on redemption intention. No 2 The effect of time sensitivity on redemption intention is stronger for mobile coupon than physical coupons. No 3 Price Consciousness has a positive effect on redemption intention No 4 The effect of price consciousness on redemption intention is stronger/weaker for mobile than for physical coupons. No
5 Brand loyalty has a positive effect on redemption intention. Yes
6 The effect of brand loyalty on redemption intention is stronger for physical than for mobile coupons. No 7 Personal Innovativeness has a positive effect on redemption intention. No 8 The effect of personal innovativeness on redemption intention is stronger for mobile than for physical coupons. Yes 9 Privacy Concern has a negative effect on redemption intention. No 10 The effect of privacy concern on redemption intention is stronger for mobile than for physical coupons. No
11 Coupon usage has a positive effect on redemption intention. Yes
4. Discussion (of results)
4.1 Theoretical Implications
This research focusses on the difference in redemption intention of consumers between mobile and physical coupons. Many researchers (Bawa & Shoemaker, 1987; Danaher et al., 2015; Gray, 2009; Scott A. Neslin & Shoemaker, 1983; Ramaswamy & Srinivasan, 1998; Reibstein & Traver, 1982) focussed on the impact of different factors on the redemption rate of either mobile or physical coupons, while none managed to compare the two coupon types and identify differences. In the existing literature it was concluded that price, time urgency and brand loyalty had a severe impact on the redemption of mobile and/or physical coupons. The factors personal innovativeness (Agarwal & Prasad, 1998), privacy concern (Ahmadi & Ghahfarokhi, 2016; Okazaki et al., 2009) and previous coupon usage (Ajzen & Fishbein, 1975) came from other related researches and were considered to be important for this study on coupon redemption.
Some interesting insights were found in the data. One of the results that does not match the literature is the insignificance of the effect of time sensitivity on the intention to redeem a coupon. Time sensitivity was one of the variables which arose from the existing literature on coupon redemption (Danaher et al., 2015). It was to be expected that people who consider themselves time sensitive would have a higher intention to redeem a coupon. Since mobile coupons, in general, have shorter redemption times. The second assumption for this variable was that ‘time urgent people’ would even have a higher intention to redeem a mobile coupon, since the redemption time of mobile coupons is generally shorter. Though, both the direct and the moderated relationships were insignificant. For the respondents present in this sample, time sensitivity is not a factor which
itself is not a valid predictor. A consumer should have a basic interest in coupons, before time sensitivity could play a role. The relationship between time sensitivity and redemption intention might be statistically significant for a sample of coupon using consumers. In order to improve theoretical relevance, this might be an interesting topic for further research.
As highlighted in the literature review, face value is the only factor which is mentioned in both the literature about mobile and physical coupons (Danaher et al., 2015; Reibstein & Traver, 1982). Combining this with the fact that consumers who use coupons are expected to be price conscious, the positive relationship between price consciousness and intention to redeem a coupon was expected to be significant. As the results point out, though, this is not the case. In this dataset, there is no statistically significant evidence that price consciousness and intention to redeem a coupon are significantly related. This is the case for both the direct as the moderated relationship.
Brand loyalty is related to the degree and intensity of previous purchases or experiences of consumers (Neslin & Shoemaker, 1983; Peckham, 1963; Tucker, 1964). This resulted in the assumption that brand loyalty is positively related to redemption intention. The data supports this hypothesis. There is a statistically significant relationship between the degree of brand loyalty of a customer and their redemption intention. Apparently, the degree of knowledge one has from a particular brand improves their intention to redeem a coupon. This result might have to do with the increasing ‘ease of use’ when a consumer is loyal to a particular brand. Since the consumers is already present in (one of the) channels owned by a particular brand, it is very easy to redeem a coupon. The consumer is ‘already there’, so why not take advantage of a good deal.
Personal innovativeness was one of the variables which did not appear in the ‘classic’ literature on coupon redemption. As was described in the literature review, this factor was considered important due to the rise of mobile coupons and mobile devices which come with that.
Since personal innovativeness in IT is only a factor for mobile devices, the assumption was that people who are highly personal innovative will have a higher intention to redeem a mobile coupon than people who aren’t. Based on the data this relationship is statistically significant, and the hypothesis can therefore be supported. This insight is new, and contributes to the existing literature.
Based on the literature, it was suggested that a high score in privacy concern would lead to less intention to redeem a mobile coupon. However, the data of this research shows that this relationship is not significant. In other words, there is no significant relationship (either direct or moderated) between privacy concern and redemption intention. This could be explained by the privacy paradox (Acquisti et al., 2015), which states that people who claim to be very conscious concerning their privacy, generally show little or no privacy concern in their actions and decisions. Especially when a short-term incentive (for example a discount in the form of a coupon) is rewarded in exchange for personal information, people tend to value privacy as less important. The privacy paradox could therefor explain why privacy concern and intention to redeem a coupon are not statistically significantly related.
Coupon usage has a statistically significant relationship with redemption intention. As was already suggested in the literature review, previous behaviour is a strong predictor of future behaviour intention (Ajzen & Fishbein, 1975). This happened to be true for this thesis as well, since the number of coupons redeemed in the last 6 months, is statistically significant related to redemption intention. The data shows no statistically significant difference between mobile and physical coupons, though. A possible explanation for this result can be that the experience between a mobile or physical coupon is quite similar. The consumer gets a discount for showing a (mobile)
discount by redeeming a coupon, is the same for both coupon types. Therefore, the effect of coupon usage would be the same for both types of coupons.
As was discussed earlier, only 1 moderated relationship was found to be statistically significant. The research question as proposed in the introduction was twofold:
1. Which factors influence the intention of consumers to redeem a coupon?
2. How do the factors influencing consumers’ coupon redemption differ between physical and mobile coupons?
The answer to the first part of the research question is ‘coupon usage’ and ‘brand loyalty’. These factors influence the redemption intention of consumers. The second part of the research question, whether or not there is an overall difference in redemption intention between mobile and physical coupons, is true. The data shows a difference in redemption intention for mobile coupons, between high and low personal innovative consumers. This difference cannot be made for physical coupons. In other words, the factor personal innovativeness influences redemption intention in another way for mobile than for physical coupons.
5.2 Managerial implications
Marketing managers who are dealing with, or considering the change of medium for their coupon platforms, have to take into account whether or not their customers will score high on personal innovativeness. If this is not the case, when the target audience of a particular firm is seniors for example, one should reconsider the decision to ‘go mobile’ with their coupons. The benefits of this change, less costs and deeper customer insight, will otherwise be nullified by the decrease of participants.
Besides the presence of the personal innovativeness effect, the absence of the effect of privacy concern on redemption intention is also very interesting to consider. Since privacy is such a hot topic nowadays, it is mentioned a lot in the literature. But, when considering the results of this thesis and the theory about the privacy paradox as discussed in the theoretical implications it can be concluded that the privacy concerns of consumers can be overturned by offering a good incentive. In general, consumers are willing to give up on their personal information when a short-term incentive is offered. Marketing managers would benefit from providing extra incentives when digitizing their coupon platform.
Overall, there are a lot of pros and cons to implement mobile coupons or not. As the theories of Neslin & Shankar (2009) and Ieva & Ziliani (2017) suggested, synergy is very important among B2C marketing activities. Based on this idea: it is perfectly possible to maintain the physical coupon while at the same time transferring to the mobile variant. In this way, both consumers who score high on personal innovativeness as consumers who do not can redeem their favourite type of coupon.
5.3 Limitation and future research
The results and findings of this thesis are subject to a number of limitations. At first, the participants used to collect the data were approached by a convenience sampling method. And even though the sample was pretty diverse in terms of gender, age and income, a majority of the respondents was highly educated.
Second, the dependent variable of this thesis is redemption intention. Even though the intention of a consumer to redeem a coupon is a valuable thing to know, it is not the same as actual
According to the results of this thesis, people who had an experience with coupons in the past score higher on intention to use a coupon in the near future. This effect does not differ between mobile or the physical coupons. In order to statistically support this assumption, future research should focus on the full construct of previous experience as a predictor of coupon redemption intention. Ideally, the impact of previous coupon usage on real sales data will be measured in the future. Since this current research measures the influence of the independent variables on redemption intention, real behaviour is not part of the model. Discrepancy is present between the willingness of a consumer to do something (intention) and their actual behaviour. In general, intention only explains 28% of the variance of behaviour (Sheeran, 2002). Future research should focus on the influence of the different factors on actual behaviour, to get a full view of this topic. Besides the difference in methodology, a distinction between consumers who use a lot of coupons and those who do not use a lot of coupons would be interesting to make. Based on the significant relationship between coupon usage and redemption intention, it does matter how many coupons a consumer has redeemed in the past time. In order to analyse whether or not the number of coupons has an impact on the redemption intention, further research in required.
