The effect of coupon characteristics
on coupon redemption.
by
N. Lettinga
The effect of coupon characteristics
on coupon redemption.
by
N. Lettinga
University of Groningen
Faculty of Economics and Business
Department of Marketing
July, 2011
Course name: Master thesis
Program: Marketing management and Marketing research Supervisors:
Dr. Sonja Gensler Dr. Jia Liu
Niels Lettinga Eerste Hunzestraat 23a
9715BJ Groningen +31 (0) 614393500 N.Lettinga@student.rug.nl
Abstract
The main objective of this study was to examine the effect of coupon characteristics on coupon redemption. The effect of the face value, the regular price, the redemption distance, the type of product and the expiration date were analyzed with actual data from the Groupon website. The results show that a higher face value, a lower
regular price and a longer expiration duration lead to a higher number of coupons redeemed. Furthermore, hedonic products and coupons that can be redeemed in one’s own city are redeemed more often. The effect of framing was analyzed via an experiment.
The results show that the way the discount on the coupon is framed has no effect on coupon redemption.
Niels Lettinga
The effect of coupon characteristics on coupon redemption.
Table of content
1. Introduction ... 6
2. Conceptual background ... 10
2.1 Face value ... 11
2.2 Regular price ... 12
2.3 Distance to the redemption location ... 13
2.4 Type of product ... 14
2.5 Expiration date ... 16
2.6 Framing ... 16
2.7 Conceptual model ... 19
3. Actual data from coupon website ... 20
3.1 Methodology ... 20 3.1.1 Data ... 20 3.1.2 Operationalization of variables ... 21 3.1.3 Methodology ... 22 3.2 Results ... 23 3.2.1 Descriptive statistics ... 23 3.2.2 Model formulation ... 24 3.2.3 Importance characteristics ... 27
3.3 Discussion and implications ... 33
3.4 Limitations ... 36
4. Experiment framing ... 37
4.1.1 Design ... 37 4.1.2 Subjects ... 37 4.1.3 Dependent variables ... 38 4.1.4 Procedure ... 39 4.1.5 Analysis... 40 4.2 Results ... 41
4.3 Discussion and implications ... 43
4.4 Limitations ... 44
5. General discussion ... 45
References ... 48
Appendices ... 56
Appendix 1: Questionnaire ... 56
1. Introduction
There are numerous definitions of coupons. A general definition is from Coupon
Information Corporation (2008): “A coupon is a certificate with a stated value if it is used when purchasing a specific product. The consumer presents the coupon to the retailer selling the product at the time of purchase. The retailer will reduce the cost of the product
by the specific amount on the coupon”. The definition on Wikipedia is “a coupon is a ticket or document that can be exchanged for a financial discount or rebate when purchasing a product” (http://en.wikipedia.org/wiki/Coupon). Coupons can be organized
in different manners but there is not a clear categorization of different types of coupons. For example, there are coupons offered by manufacturers or coupons offered by retailers. Another example, there are coupons that can be cut out of a magazine or coupons that have to be bought online. Coupons have been a popular promotional tool since the late
is crucial for determining increases in sales (Dickson and Sawyer, 1990). Also, coupon promotions can speed category purchases (Blattberg et al., 1981; Shoemaker, 1979) because consumers purchase a given brand sooner (Neslin and Shoemaker, 1983).
Furthermore, coupon redemption can have significant effects on repeat purchases (Jones and Zufryden, 1981; Kuehn and Rohloff, 1967), as coupons can attract new consumers and influence their subsequent purchase patterns (Ben-Zion et al., 1999; Neslin and
Shoemaker, 1983). Thus, coupon redemption is vital for the success of a product.
The low redemption rate has historically been observed with coupons that were mainly distributed through free-standing inserts (e.g. magazines, newspapers), direct mailings and in stores (Hutton and Francis, 2002). The rise of the internet has paved the way for another kind of coupon distribution: on-line coupons. Electronic communication is an increasingly important platform for marketing and the potential for innovations such as e-coupons to spur interest in and redemption of coupons is profound (Blundo et al., 2005). Han and Sung (2006) forecasted a positive future for online coupons with increasing amounts of exposure and familiarity with the technology involved. E-coupons might eliminate or reduce some of the problems associated with traditional coupon methodologies, i.e. low redemption rate (Sullivan, 2008). Thus, the research question
which will be answered is:
What is the effect of coupon characteristics on coupon redemption in an on-line setting?
as mentioned above, coupons were traditionally distributed via free-standing inserts, direct mailings and in stores. Consumers received the coupon without purchasing them and the discount was subtracted from the regular price when the purchase was made in the store. This fits well with the definitions of coupons stated above. In an on-line setting, coupons are purchased immediately online and have to be redeemed subsequently offline. Consumers thus purchase the product when they purchase the coupon, not when they
purchase the product itself. This gives rise to some new interesting variables that might affect the coupon redemption. Second, in this study actual coupon data will be used. Experiments usually only take into account a limited number of variables and lack ecological validity. Actual behaviour gives the most realistic view of what coupon
characteristics affect coupon redemption. Third, for the relevant characteristics where this is not possible an experiment will be conducted. Even though actual data is preferred over experiments, some important variables cannot be analysed with actual data.
Therefore, if the literature states that a variable is important but the exact influence of that variable of coupon redemption can not be determined via actual data, an experiment will be conducted. This study thus contributes to the coupon redemption literature. In addition, this study also has managerial relevance. The number of websites that offer coupons has
increased over the last past years to around 500 and the leading companies in this market have revenues in the millions of dollars. (www.wikipedia.org/wiki/groupon). When these companies know what coupon characteristics can increase coupon redemption can result
in increased sales (Dickson and Sawyer, 1990).
regular price and a longer expiration duration lead to a higher number of coupons redeemed. Furthermore, hedonic products and coupons that can be redeemed in one’s own city are redeemed more often. Finally, the results show that the way the discount on
the coupon is framed has no effect on coupon redemption.
This paper is structured in the following way. First, the conceptual background is given which is followed by the hypotheses and the conceptual framework. Second, the
2. Conceptual background
The foregoing discussion showed that coupon redemption is of vital importance to
marketers. Researchers have investigated which factors have an influence on coupon redemption. Five factors have been found that influence coupon redemption: coupon characteristics, brand characteristics, product category characteristics, consumer
characteristics and distribution characteristics (Wang, 1998; Nickels and Wood, 1997; Reibstein and Traver, 1982). The primary characteristics are the coupon, brand, product category and consumer characteristics (Wang, 1998).
Coupon characteristics include the face value, the expiration date and the framing of the coupon (Yin and Dubinsky, 2004). The face value is the amount of discount offered on a coupon and the expiration date is the duration that a coupon can be redeemed. The framing is the way the discount is displayed. There are several ways to frame a
discount. In practice a common method is stating a face value in cents-off, percentage-off or reduced price (Yin and Dubinsky, 2004). For example, the regular price of a can of soda is 1.20 euro and it is on sale for 80 euro cents. The cents-off is then 40 euro cents,
the percentage-off is 33.3 percent and the reduced price is 80 euro cents.
Brand characteristics include the regular price and the redemption location (Yin and Dubinsky, 2004). The regular price is the price that one normally has to pay for the
product and the redemption location is the location where the product can be redeemed. For example, the redemption location can be in one’s own town or in a neighboring city.
Product type characteristics include what type the product is (Yin and Dubinsky,
dinners and theme park admissions. Utilitarian products are more functional and an example is driving lessons.
Consumer characteristics include among other attitude toward coupons and
coupon proneness (Yin and Dubinsky, 2004). Consumers who are more responsive to coupon promotions are coupon prone (Lichtenstein et al., 1990). However, consumers now purchase coupons online, and therefore it is impossible to determine consumer
characteristics.
Distribution characteristic is the way the coupon is distributed (Yin and Dubinsky, 2004). The focus of this study is the online distribution of coupons. Therefore, there is no variance in this characteristic and it will not be included.
Concluding, in this study the focus is on the following six characteristics: face value, regular price, redemption location, type of product, expiration date and framing. Each of these characteristics will be discussed next.
2.1 Face value
Face value is the amount of discount offered on a coupon (Blattberg and Neslin,
1990). Of those coupon characteristics that influence coupon redemption, coupon face value is particularly important to marketers because it affects the cost of the coupon. Previous research has found that higher face values lead to higher redemption rates (Leone and Srinivasan, 1996; Reibstein and Traver, 1982). This makes sense because
(1991) showed that higher face values provide more information to consumers. They suggest that consumers who use coupons will view face values as positive pieces of information about the product. They found that higher face values provide more
information about the brand than lower face values, resulting in a higher redemption rate for higher face values. A final concern regarding the face value is that when a discount becomes to high, consumers may become suspicious (Chen et al., 1998). Consumers can
suspect that the regular price is inflated or a permanent price promotion might follow. However, these suspicions can be easily countered by offering regular price information next to the face value, which is the topic of the next paragraph. Concluding, higher face values lead to higher coupon redemption rates. Therefore, the following hypothesis is
proposed:
H1: A higher face value leads to a higher coupon redemption rate.
2.2 Regular price
Regular price is the original price of the product or service that the coupon entails. Regular price information gives consumers the possibility to calculate their savings and it
gives consumers a reference point for the value of the product (Yin and Dubinsky, 2004). Although regular price is mentioned as a characteristic that can influence coupon redemption (Wang, 1998; Nickels and Wood, 1997; Reibstein and Traver, 1982), there is no direct empirical evidence to indicate whether consumers are more likely to redeem a
might be that coupons with a higher priced product are redeemed more often. An explanation would be that higher prices will lead to higher absolute face values, i.e. a higher discount. For example, a 50% discount on a 1000 euro product will lead to a
higher discount than that same discount on a 50 euro product. Second, another possible result might be that coupons with a lower priced product are redeemed more often. An explanation would be that prepurchase evaluation is very difficult because consumers
first have to redeem the coupon before they experience the product or service. Therefore, there is considerable prepurchase risk (Lovelock and Gummesson, 2004). When the regular price is lower, the consumer can lose less, i.e. money. This will decrease the prepurchase risk and increase the probability that the consumer will buy the coupon. Thus,
there are reasons to support both directions. Therefore, the following two contradicting hypotheses are proposed:
H2a: A higher regular price leads to a higher coupon redemption rate.
H2b: A lower regular price leads to a higher coupon redemption rate.
2.3 Distance to the redemption location
Distance to the redemption location is also a determinant which can explain
coupon redemption. The rise of the availability of coupons in an online-setting paved the way for this new variable. The precise effect that this variable might have can be explained by a conceptual framework proposed by Bawa and Shoemaker (1987). The net benefit for consumers when using coupons can be stated as followed:
Economic benefits come from the received discount. Psychic benefits come from for example the fact that the consumers thinks of oneself as a shrewd shopper.
Substitution costs are the costs of not gaining the utility of the otherwise bought product of service. Effort costs are the costs for consumers in search for the coupon and processing and redeeming them (Blattberg and Neslin, 1990). An example of the effect of
effort costs was found by Chakraborty and Cole (1991). They found that coupon redemption is influenced by the ease or difficulty with which consumers can redeem a coupon. When participants had to redeem the coupon within 24 hours and had to bring additional possessions, i.e. difficult condition, they redeemed the coupon less often. The
effort costs are particularly important when discussing the distance to the redemption location. When one has to drive or take the public transportation to another city, the effort to redeem the coupon will go up. Redeeming that same coupon in one’s own city will
cost less effort. Thus, when a coupon can be used in one’s own town it will cost less effort and is consequently more likely to be redeemed. Therefore, the following hypothesis is proposed:
H3: When the distance to the redemption location is short, it is more likely that the
coupon will be redeemed.
2.4 Type of product
The type of product is also a characteristic that can influence coupon redemption.
desire for sensual pleasure, fantasy and fun (Strahilevitz and Myers, 1998). In contrast, utilitarian or goal oriented consumption is more cognitively driven, instrumental, and goal oriented and accomplishes a functional or practical task (Strahilevitz and Myers,
1998). Consumes love to be given a gift they would not usually buy with their own money. Nunes and Dréze (2006) state that in loyalty programs; hedonic products are redeemed more than utilitarian products. For example, when a large loyalty program
merchant analyzed the data from their customers, they found that consumers preferred the latest electronics, i.e. hedonic, over household goods, i.e. utilitarian. Furthermore, when customers redeem points for hedonic products they subsequently buy more at the loyalty program because of pleasant associations. This shows that when consumers have the
opportunity to redeem points collected with a loyalty program, they are more likely to redeem hedonic products. A parallel can be drawn from loyalty programs to coupons because both require consumers to redeem. Consumers can also see a coupon as a little
extra for something they usually would not buy. Historically, coupons were redeemed by loyal, repeat customers instead of new customers (Shoemaker and Tibrewala, 1985). These customers used coupon to buy products that they normally would also buy, i.e. utilitarian products. With the rise of the availability of coupons in an online-setting this
has changed. Coupons on these websites are frequently for the little extras like dinners and temporary memberships for gyms, i.e. hedonic products. Therefore, the following hypothesis is proposed:
H4: Coupons for hedonic products will be redeemed more often those for utilitarian
2.5 Expiration date
The expiration date is also a characteristic that can influence coupon redemption. In 1987, 26% of all coupons were issued without an expiration date. By 1991, less than 1% were issued without an expiration date (Martin, 1991). Inman and McAlister (1994)
found that right after a coupon was issued, the redemption rate was highest. The redemption then slowly decreased and just prior to the expiration date that redemption rate increased again. Thus, right before the coupon expired there was another peak in
coupon redemption. Roehm & Roehm (2011) found that gift certificates with a short time frame are more evaluated on concrete features like face value and the frame. Gift certificates with a long time frame are more evaluated on abstract features like incentive goal congruity and the fit with personal values. Finally, Chakraborty and Cole (1991) use
an expiration duration to induce a high effort condition. Looking at equation (1), high effort leads to lower coupon redemption. These examples show that the expiration date can have an effect on the coupon redemption. It is expected that coupons which can be
redeemed over a longer period, i.e. have a longer expiration duration, induce lower subjective effort which results in higher coupon redemption. Furthermore, when the expiration duration is longer the likelihood that a coupon expires is lower and therefore there is a higher probability that a coupon is redeemed. Therefore, the following
hypothesis is proposed:
H5: A longer expiration duration leads to a higher coupon redemption rate.
2.6 Framing
is the use of decision-related information by a buyer to evaluate a product or service relative to a reference point (Kahneman and Tversky 1979; Puto 1987). Changes in how something is framed can reverse consumer decisions, turning strong preferences in one
direction into an equally strong preference in the opposite direction (Gleitman et al., 2003). Thus, framing of decision problems can influence individuals' judgments and preferences (Kahneman and Tversky 1979; Taversky and Kahneman 1981; 1991). For
example, Levin and Gaeth (1988) found that participants evaluated ground beef labeled 75% lean more positively than beef labeled 25% fat.
In a meta-analysis about the effect of price presentation of perceived savings, Krishna et al. (2002) found that in general, consumers tend to perceive percentage-off
promotions as larger than cents-off promotions. There are however several moderators that can influence this relation. First, Chen et al. (1998) found that the effect of framing of a discount depends on the regular price of the product. For example, a 1000 dollar
discount from a 20.000 dollar is car is seen as more attractive than a 5 percent discount from that same car. On the other hand, a 25 dollar cent discount from a 50 dollar cents chewing gum is less attractive than a 50 percent discount from that same gum. Thus, a percentage-off frame is more attractive than a cents-off frame when the regular price is
low and it is the other way around when the regular price is high. Second, when consumers are motivated to calculate the exact resulting price, they have a more difficult time when calculating a promotion when it is framed in percentage-off (DelVecchio et al.,
multiplication or division required when working with percentages is more difficult for consumers (Rhymer et al., 2002). This will result in uncertainty with regard to the resulting price. Consumers tend to prefer highly certain outcomes to outcomes that are
not precisely known (Camerer and Weber, 1992; Winkler, 1991). Thus, in this case cents-off should be the choice of customers. Thus, a percentage-cents-off frame is more attractive than a cents-off frame when consumers do not calculate the resulting price and it is the
other way around when they do calculate the resulting price. In the second study where the effect of framing is evaluated, the regular price is low and consumers are not expected to have any difficulties with calculating the resulting price. Therefore, the following hypothesis is proposed:
H6a: A coupon which is framed in percentage-off will be redeemed more often
than coupons which are framed in reduced price and cents-off.
Yin and Dubinsky (2004) argue that, for both lower and higher priced products, the reduced price is perceived as between the percentage-off and cents-off frame. Della Bitta et al. (1981) found that for high priced products the perceived savings and intent to purchase is highest for cents-off frames, followed by reduced price frames and finally
percentage-off frames. Based on Chen et al. (1998) finding that the attractiveness of different frames used on coupons is reversed when a low priced products is used compared to a high priced product, it is expected that a coupon which is framed in
H6b: A coupon which is framed in reduced price will be redeemed more often than
a coupon which is framed in cents-off but less than a coupon which is framed in
percentage-off.
2.7 Conceptual model
The effect of the six characteristics on the number of coupons redeemed can be
found in figure 1. The three control variables are all related to the website where the coupons are displayed on; the day of the week the coupon is displayed, the specific city under which the coupon is displayed and how many days the coupon is displayed.
Therefore, the control factors are discussed in the methodology section of the first study.
Figure 1. Conceptual model
H5 H4 H3 H2 H1 Redemption distance Face value Type of product Expiration date Regular price Coupons redeemed Framing H6 Control variables: Day of the week City
3. Actual data from coupon website
The first study is done in order to test H1 to H5. Actual data from the Groupon
website including face value, regular price, redemption location, type of product and expiration date of coupons were used.
3.1 Methodology
3.1.1 DataGroupon is a website that provides coupons for various products and services. It was launched in 2008 and has localized website for markets in the U.S., Europe, Asia and South America. It serves 35 million registered users and it is valued at $1.35 billion (www.groupon.com). Every coupon displays the same information, i.e. the face value
(framed in three different ways), the regular price, what can be done with the coupon, where this can be done and the expiration date, for an example see figure 2.
Figure 2. Example coupon on Groupon.com
The data was collected from the Groupon website (www.groupon.nl) from the
3.1.2 Operationalization of variables
The dependent variable is the number of coupons redeemed. Every day a new coupon is put online per city in the Netherlands. Per coupon one can see how many times
the coupon has been redeemed. At the end of each day the data was collected from the website.
There are five independent variables. The regular price and the face value are both mentioned on the website. Both are given in euros and are thus metric variables. The type
of the product was determined by the author for each coupon as either hedonic or utilitarian. The distance to the redemption location was determined by seeing if the redemption location was the same or not as the city under which the coupon could be purchased. The expiration date was stated on the website. This variable is the number of
days the coupon can be redeemed and is thus a metric variable.
There are three control variables which have to be accounted for. First, the specific day that the coupon is displayed on the website. Rutz and Bucklin (2011) found
that consumers perform less searching on the internet during the weekend than the weekdays. Thus, it might be that coupons redemption is higher in the weekdays than the weekends. Second, the specific city where the coupons is displayed on the website. A
coupon that is displayed under the city of Amsterdam which has almost 800.000 inhabitants will probably be redeemed more than a coupon that is displayed under the city of Assen which has little over 60.000 inhabitants. Cities that account for less than 2% of
for 24 hours or displayed for 48 hours. It is expected that a coupon that is displayed for a longer duration will be redeemed more often.
3.1.3 Methodology
The dependent variable is count data with a Poisson distribution (Cameron and
Trivedi, 2005; Hill et al., 2001). The models to be estimated can be expressed as:
i dwi i cw i cw i cw twi edi tpi rdi rpi fvi i ε β β β β β β β β β β α γ + + + + + + + + + + + = ) * 10 3 * 9 2 * 8 1 * 7 * 6 * 5 * 4 * 3 * 2 * 1 0 exp( (2) where
γ : Number of redeemed coupons
i : Each individual coupon
α0: Intercept
fv: Face value, 1β = 0.009, a higher face value leads to more coupons redeemed
rp: Regular price, 2β = -0.01, a lower regular price leads to more coupons redeemed
rd: Dummy variable for own city redemption location, base case: neighboring city
tp: Dummy variable for hedonic product, base case: utilitarian
ed: Expiration date, 5β = 0.003, a longer expiration duration leads to more coupons
redeemed
tw: Dummy variable for 24 hours on website, base case: 48 hours cw1: Dummy variable for large city on website, base case: very large city
cw2: Dummy variable for medium city on website, base case: very large city cw3: Dummy variable for small city on website, base case: very large city dw: Dummy variable for weekday on website, base case: weekends
3.2 Results
3.2.1 Descriptive statistics
The number of coupons redeemed ranged from 0 to 2105 with an average of 136. However, the median, 43 coupons, is a lot lower indicating that there are a couple of high
outliers. The regular price is between 5 and 2500 with a mean of 179 euro’s. The median however is 76.50 euro’s, again indicating that there are some high outliers. The face value in euros has a range of 3 till 1501 with a mean of 111. The median is 42 for the face
value in euros revealing again that that are high outliers. The face value in percentages has a minimum of 50 with a maximum of 95. The mean face value in percentages is 59 and the median is 54. The expiration date ranges from 1 day till 1 year, i.e. 360 days. The
mean is 117 days which is around 4 months. The above information is summarized in table 1.
Variables Mean Standard
deviation
Median Minimum Maximum
Coupons redeemed 135.84 266.98 43 0 2105
Regular price 179.41 326.63 77 5 2500
Face value (€) 111.45 207.69 42 3 1501
Face value (%) 59.14 10.29 55 50 95
Expiration date 116.94 67.81 129 1 360
Table 1. Descriptive statistics metric variables
There are a lot more coupons offering hedonic products and services (81.1%) than utilitarian products and services. The redemption location is more evenly distributed. 42.2% of the coupons can be redeemed in one’s own city. The above information is
Variable Level Percentage
Type of product Hedonic 81.1
Utilitarian 15.9
Redemption location Same city 42.2
Other city 57.8
Table 2. Descriptive statistics nominal variables
Around 84% of the coupons are displayed for 24 hours on the website and most of the coupons (76%) are displayed on the weekdays. Coupons that are displayed on the
website for small cities account for 20% of the coupons, medium cities for 37%, large cities for 25% and very large cities for 18%. The above information is summarized in table 3.
Variable Level Percentage
Time 24 hours 84.1
48 hours 15.9
Day of the week Weekdays 76.2
Weekends 23.8
City Small city 20.3
Medium city 36.5
Large city 24.9
Very large city 18.4
Table 3. Descriptive statistics control variables
3.2.2 Model formulation
The dependent variable “coupons redeemed” is the number of times a coupon is bought online. The number of coupons redeemed is not normally distributed. Even the logs are not normal thus GLIM has to be used, see figure 3 and 4. GLIM stands for
GLIM combines two generalizations over the regular linear model. First, the GLIM allows for the fact that the Y-variable doesn’t have to be normally distributed. For example, if the case of count data, a Poisson distribution is most suitable. Second, one
can linearizable the model by taking a link function. It is called a link function because it links the expectation of Y to a formula that is linear to X. There are several link functions, e.g. identity, log and logit. If for example the expectation of Y is positive, in the case of
count data, the log-link should be chosen. There are preferred combinations of distributions and link functions. In the case of count data, a model with a Poisson distribution would create values of the counts that are positive, which is what the log-link achieves. Concluding, the model that is used is a Poission loglinear GLIM.
Figure 3. Coupons redeemed Figure 4. Log of coupons redeemed
The five independent variables and three control variables are included in the model. The regular price, the face value and the expiration date are all metric and
The fit of the model stands for the likelihood that the parameters generate the observed data. Thus, a model with a higher likelihood has a better fit. The deviance is a measure of fit and measures the distance of the model to a perfect-fitting model. The
absolute magnitude of the deviance (147.46) can not be interpreted on its own. It can only be interpreted in relation to another model. The model with the lower deviance has a better fit. The Bayesian Information Criterion (54984.08) is also a measure of fit and is
suggested when dealing with large sample sizes. The BIC corrects for the amount of additional parameters coming from an increased number of included independent variables. Lower values of BIC identify a model with a better fit, but again the absolute value has no meaning on its own. It can only be interpreted in relation to another model.
Variables B Std. Error Significance
Intercept 5.758 0.02960 < 0.01
Regular price -0.010 0.00020 < 0.01
Face value 0.009 0.00030 < 0.01
Expiration date 0.003 0.00007 < 0.01
Type of product (hedonic) 0.357 0.01850 < 0.01
Redemption location (same city) 0.343 0.01040 < 0.01
Time (24 hours) -0.674 0.01490 < 0.01
Day of the week (weekdays) 0.379 0.01300 < 0.01
City (small) -2.225 0.02110 < 0.01
City (medium) -1.524 0.01280 < 0.01
City (large) -1.184 0.01320 < 0.01
Table 4. Parameter estimated model
regular price has a negative beta indicating that a lower price leads to a higher number of coupons redeemed. This gives an initial glimpse into the effect of the regular price on coupon redemption. A possible explanation as already indicated in the literature review is
that the perceived prepurchase risk is higher with high priced products. This finding rejects hypothesis 2a en confirms hypothesis 2b. Third, the redemption location has a positive beta indicating that coupons that can be redeemed in one’s own city are
redeemed more often. This confirms hypothesis 3. Fourth, the type of product has a positive beta indicating that hedonic products and services are redeemed more often. This confirms hypothesis 4. Fifth, the expiration date has a positive beta indicating a longer expiration duration leads to a higher number of coupons redeemed. This confirms
hypothesis 5.
All thee control variables are significant. First, coupons that are displayed on weekdays are redeemed more often than coupons that are displayed on weekends. Second,
the larger the city that a coupon is displayed on the more that coupon is redeemed. Third, coupons which are displayed on the website for 48 hours are redeemed more often than coupons which are displayed on the website for 24 hours.
3.2.3 Importance characteristics
The importance of each of the coupon characteristics was determined by varying
values from table 5, are entered into equation (2), it is expected that 306 coupons are redeemed.
Face value 11 euro
Regular price 179 euro
Redemption location: Own city
Type of product Hedonic
Expiration date 117 days
Time 24 hours
City Very large
Day of the week Weekdays
Table 5. Base case
The face value and regular price are combined in order to determine the percentage-off for each coupon. The overall average discount is 62.12%, calculated by
dividing the face value by the regular price. Thus, with a 62.12% discount, it is expected that 306 coupons are redeemed. When a 1% increase to 62.74% on the discount is offered the predicted number of coupons redeemed is 309, which is an increase of 0.98%. When
a 1% decrease to 61.50% on the discount is offered the predicted number of coupons redeemed is 303, which is a decrease of 0.98%. When a 5% increase to 65.23% on the discount is offered the predicted number of coupons redeemed is 322, which is an
increase of 5.23%. When a 5% decrease to 59.01% on the discount is offered the predicted number of coupons redeemed is 291, which is a decrease of 4.90%. When a 10% increase to 68.33% on the discount is offered the predicted number of coupons
0 50 100 150 200 250 300 350 400 10% less discount 5% less discount 1% less discount Base case (62.12% discount) 1% more discount 5% more discount 10% more discount N u m b e r o f c o u p o n s r e d e e m e d
Figure 5. Percentage-off discount importance
The average expiration date is 117 which results in 306 coupons redeemed. When
the expiration date is increased with 1% to 118.17 days the predicted number of coupons redeemed is 307, which is an increase of 0.33%. When the expiration date is decreased with 1% to 115.83 days the predicted number of coupons redeemed is 305, which is a
decrease of 0.33%. When the expiration date is increased with 5% to 122.85 days the predicted number of coupons redeemed is 312, which is an increase of 1.96%. When the expiration date is decreased with 5% to 111.15 days the predicted number of coupons
redeemed is 300, which is a decrease of 1.96%. When the expiration date is increased with 10% to 128.70 days the predicted number of coupons redeemed is 319, which is an increase of 4.25%. When the expiration date is decreased with 10% to 105.30 days the
0 50 100 150 200 250 300 350 400 10% less days 5% less days 1% less days Base case (117 days) 1% more days 5% more days 10% more days N u m b e r o f c o u p o n s r e d e e m e d
Figure 6. Expiration date importance
Concluding, the percentage-off discount of the regular price is almost two and a
half times more important than the expiration date.
The number of coupons redeemed when one can redeem a coupon in one’s own
city is 306. When the redemption location is changed from one’s own city to a neighbouring city, the predicted number of coupons redeemed is 217, which is a decrease of 29.08%. The number of coupons redeemed when the coupon is for a hedonic product
or service is 306. When the type of product is changed from hedonic to utilitarian, the predicted number of coupons redeemed is 214, which is a decrease of 30.07%.
96.08%. The number of coupons redeemed when the coupon is displayed on a weekday is 306. When the specific day that a coupon is displayed on the website is changed from a weekday to a day in the weekend, the predicted number of coupons redeemed is 210,
which is a decrease of 31.37%. The number of coupons redeemed when the coupon is displayed on a very large city is 306. When the specific city that a coupon is displayed on the website is changed from a very large city to a large city, the predicted number of
coupons redeemed is 94, which is a decrease of 69.28%. When it is changed to a medium city, the predicted number of coupons redeemed is 67, which is a decrease of 78.10%. When it is changed to a small city, the predicted number of coupons redeemed is 33, which is a decrease of 89.22%. The decrease from a large city to a medium city is
28.72% and the decrease from a large city to a small city is 64.89%. The decrease from a medium city to a small city is 50.75%.
Concluding, the time on the website and the specific city on the website seem to be the most important nominal variables with percentage change values around 90 percent. The specific day on the website, the type of product and the distance to the redemption location are less important and have percentage change values around the 30.
The total ranking for the importance of all variables can be found in table 6.
Variables Change in
coupons redeemed
1. Hours on website (change from 24 to 48 hours) 96.08%
2. City on website (average change from very large city to other cities) -78.87%
3. Day on website (change from weekday to weekend) -31.37%
1. Percentage-off discount (per additional percentage) 1.00%
2. Expiration date (per additional percentage) 0.41%
Table 6. Ranking importance variables
In order to compare the metric and nominal variables the following equation are computed. The importance function of the percentage-off discount is almost a linear. Per additional percentage discount an additional 1.00% coupons is redeemed. This means that
in order to achieve a 30% increase in coupons redeemed, the discount for a coupon should be increased with 30% above the average of 62.12% which results in a 81% discount. In order to achieve a 90% increase in coupons redeemed, the discount for a
coupon should be increased with 90% above the average of 62.12% which is not possible. The importance function of the expiration date is also almost a linear. Per additional percentage extra days an additional 0.41% coupons is redeemed. This means
that in order to achieve a 30% increase in coupons redeemed, the expiration date for a coupon should be increased with 73.04% above the average of 117 days which results in 85 additional days. In order to achieve a 90% increase in coupons redeemed, the expiration date for a coupon should be increased with 219.91% above the average of 117
3.3 Discussion and implications
The main purpose of this first study was to determine the effect of face value, regular price, distance to redemption location, type of product and expiration date on coupon redemption. The results showed that a higher face value and the longer the
expiration duration lead to a higher number of coupons redeemed. Furthermore, hedonic products and coupons that can be redeemed in one’s own city are redeemed more often. These findings are congruent with our expectations. For the regular price two conflicting
hypotheses were given. The results show that a lower regular price leads to a higher number of coupons redeemed. Coupons that are displayed on weekdays, coupons that are displayed on larger cities and coupons that are displayed on the website for 48 hours are redeemed more often.
Looking at the metric variables; the percentage-off discount of the regular price is almost two and a half times more important than the expiration date. The time on the website and the specific city on the website are the most important nominal variables.
The specific day on the website, the type of product and the distance to the redemption location are less important. Now follows a discussion of each of the variables in more in-depth.
First, in accordance with Leone and Srinivasan (1996) and Reibstein and Traver (1982), it was found that the higher the face value the more coupons are redeemed.
Second, the results show that a lower regular price leads to a higher redemption
service (Lovelock and Gummesson, 2004). Especially when dealing with coupons that a predominantly used for experience and credence services. Experience attributes can only be evaluated after the purchase; consumers have to experience it, e.g. dinners. Credence
attributes can never be evaluated, e.g. quality of repair and maintenance work (Zeithaml, 1981). Most coupons on the Groupon website are for dinners, gym memberships, wellness, workshops and one-day trips. These activities contain a lot of experience and
credence attributes. Therefore, there is considerable prepurchase risk. When the regular price is lower, the consumer can lose less, i.e. money. This will decrease the prepurchase risk and increase the probability that the consumer will buy the coupon.
Third, in accordance with Chakraborty and Cole (1991), it was found that
coupons which require less effort to redeem, i.e. in one’s own city, are redeemed more often.
Fourth, in the literature review it was already hypothesized that with the rise of
the availability of coupons in an online-setting, the percentage of hedonic product and services would be higher. On the Groupon website, 81.1% of all coupons were for hedonic product and services. Furthermore, in accordance with Nunes and Dréze (2006), the hedonic products and services are redeemed more often than utilitarian products and
services. .
Fifth, in accordance with Chakraborty and Cole (1991), it was found that the longer the expiration duration the more coupons are redeemed.
the website for a second day thus not result in lower results, in fact the number of coupons redeemed is the same as on the second day as it is on the first day.
Seventh, the specific city that a coupon is displayed on also has a large effect on
the number of coupons redeemed. The average number of citizens for a city in the very large city category is 549120, for the large city category 195510, for the medium city category 141653 and for the small city category 85449. The decrease in citizens from one
category to another closely resembles the decrease in coupons redeemed. For example the decrease in citizens from a very large city category to the large city category is 64.40%. The percentage changed in coupons redeemed when going from a very large city to a large city is 69.28%. The rest of the percentages are given in table 7.
Coupons redeemed change Number of citizens change
Very large city – large city 69.28% 64.40%
Large city – medium city 78.10% 74.20%
Medium city – small city 89.22% 84.44%
Table 7. Percentage change cities
Therefore, it seems that the change in number of coupons redeemed closely resembles to the change in number if citizens from one category to another.
Eight, in accordance with Rutz and Bucklin (2011) coupons are less redeemed in the weekend than in the weekdays.
It is therefore recommended to companies thinking of partnering with Groupon, to present your coupon on weekdays, for 48 hours, in one’s own city and in a very large city
3.4 Limitations
This first study had several limitations. First, only a limited number of coupon characteristics were taking into account. Although the ones that were included are very
important other important characteristics which can influence coupon redemption were excluded because of the design of this study. For example, consumer characteristics like coupon proneness were not included because no data from consumers could be collected.
Second, only data from the Groupon website was used from the Netherlands gathered in March and April 2011. Third, an assumption made in the design of this study is that a purchase from a coupon in Amsterdam is made by a consumer living in Amsterdam. For
example, it is assumed that a coupon displayed under Amsterdam on the website which can be redeemed in Haarlem (a small town next to Amsterdam) is purchased only by consumers living in Amsterdam. It is obviously possible that someone from Haarlem bought the coupon. This might have lead to some bias in coding the “distance to the
redemption location” variable. Where a consumer that was coded as someone who had to go to another city in order to redeem the coupon while in fact that consumer could just go to their own city. Fourth, coupons displayed on the website under the national heading
and coupons that could be delivered at home were excluded from the analysis in order to reduce the bias in the “distance to the redemption location” variable. Fifth, some of the coupons were so successful that they met their target, which is different for each coupon, and were therefore removed from the website before the duration of 24 hours. These
4. Experiment framing
On the website of Groupon, the coupons are stated in each of the three framing
methods, i.e. cents-off, percentage-off and reduced price. Therefore, it is not possible to determine the specific effect that each of these methods has on the redemption of a coupon. For that reason a distinct experiment will be conducted to analyze the effect of
framing on coupon redemption.
4.1 Methodology
4.1.1 DesignThe study is an experiment with an independent variable that has three different levels, i.e. cents-off, percentage-off and reduced price. This is operationalised by offering each participant one of three different coupons. Each coupon will display one of the three
frames. The main dependent variable is the number of coupons that are redeemed at a coffee bar on campus: Mi Caffè.
4.1.2 Subjects
90 undergraduate students from the University of Groningen participated in the
4.1.3 Dependent variables
The experiment contained a dependent variable and three mediators. These were all measured with different constructs. Semantic differential and seven point likert scales
were used, see appendix 1. First, the main dependent variable is discussed; coupon redemption. Second, three mediators are discussed which can mediate the expected causal link; printing out the coupon, perception of the value of the deal and buying intention.
Actual redemption behaviour
This dependent variable shows whether the participants actually redeemed the coupon at Mi Caffè. Participants had to go to the Mi Caffè, bring the coupon that they
printed out and show it when purchasing a coffee or tea. Therefore, actual behaviour is observed.
Printing out coupon
Participants were seated in front of a computer on which the coupon was displayed. Then they had the opportunity to print the coupon on a printer located next to
the computer. This is done to make it as similar as when using the Groupon website. Because when someone makes a purchase on the Groupon website they receive a digital coupon which they still need to print out.
Perception value deal
point scales anchored as follows: Bad deal/good deal, No-savings-at-all/extremely large savings, Low value for money/high value for money, less valuable/extremely worth and unattractive/attractive. This list determines how valuable the participants thought the deal
was. This scale is based on DelVecchio et al. (2009) and Darke and Chung (2005) and Lichtenstein et al. (1991).
Buying intention
This dependent variable was measured with a single question: “How likely is it that you are going to use the coupon?” Participant answered this question on a seven point likert scale from very unlikely to very likely.
Manipulation checks
A number of manipulation questions were asked. First, what was the regular price
and second what was the discount. These two questions were asked in order to determine if the participants had closely looked at the coupon. Finally, the purpose of the study was asked.
4.1.4 Procedure
Students on the Zernike campus were approached to participate in a short
experiment. Students who accepted were guided to a separate room where a laptop and a printer were located.
First, all participants signed a consent form. Complete anonymity was guaranteed.
end of the questionnaire participants saw a coupon with one of the three frames, see appendix 2. Participants were told that the coupon is a reward for participating with the experiment and are asked to closely look at the coupon. The coupon is for a coffee or tea
at Mi Café. A coffee or tea costs 1.20 euro and the face value was 50%. This precise percentage was chosen because it is the most common discount used by Groupon and the minimum discount as well. The three frames can be seen in table 8.
Frame Text on the coupon
Percentage-off Coffee or tea normally for 1.20 euro’s, now 50% discount! Cents-off Coffee or tea normally for 1.20 euro’s, now 0.60 cents discount! Reduced price Coffee or tea normally for 1.20 euro’s, now for 0.60 cents! Table 8. The three frames
The participants then had the opportunity to print the coupon if they wanted to. If
they printed out the coupon, the researcher put a stamp, signature and participant number on the coupon to make it official and student could not fraud with it. Also, the expiration was written on the coupon. The coupon could be used for one week after participating in the experiment. Participants were then asked to fill in the questionnaire. Finally, the
participants had the opportunity to write down their contact information if they wanted to be debriefed.
4.1.5 Analysis
The research data are analyzed with SPSS 16. All data are analyzed with
4.2 Results
Each of the four dependent variables (printing out the coupon, perception of deal value, buying intention and coupon redemption) is discussed next.
First, a logistic regression was done with “frame” as an independent variable and
the number of times the coupon was printed out as dependent variable. The reduced price frame was printed out 22 out 30 times, the cents-off frame was printed out 23 out of 30 times and the percentage-off frame was printed out 26 out of 30 times. However, these differences did not reach statistical significance. The above information is summarized in
figure 7.
Printing out coupons
0 5 10 15 20 25 30
Reduced price frame Cents-off frame Percentage-off frame
Printed Not printed
Figure 7. Printing out coupons
of the value of the deal variable as the dependent variable. There were not significant differences between the three frames (F (2,68)= 0.06, p = 0.94).
Third, an ANOVA was performed with frame as independent variable and the
buying intention as dependent variable. There were not significant differences between the three frames (F (2,68)= 0.55, p = 0.58).
Fourth, coupon redemption was analyzed with a logistic regression. Frame was
the independent variable and the number of times the coupon was redeemed was the dependent variable. The reduced price frame was redeemed 12 out 22 times (55%), the cents-off frame was redeemed 9 out of 23 times (39%) and the percentage-off frame was redeemed 10 out of 26 times (39%). However, these differences did not reach statistical
significance. Hypotheses 6a and 6b are thus not supported. The above information is summarized in figure 8. Redeeming coupons 0 2 4 6 8 10 12 14 16 18
Reduced price frame Cents-off frame Percentage-off frame
Redeemed Not redeemed
4.3 Discussion and implications
The main purpose of this second study was to determine the effect of framing on
coupon redemption. The results showed that there were no significant differences between the three frames on each of the four dependent variables. The dependent measures were also on different levels; attitudes, intention and actual behavior. Thus, an initial conclusion is that it does not matter which frame is used on a coupon. This is
counter to the findings of Chen et al. (1998) who found that when the regular price is low, the percentage-off frame is seen as more significant.
71 out of the 90 participants (78.89%) printed out the coupon and 31 out of 90
participants (34.44%) redeemed the coupon. These percentages are a lot higher than the 1% redemption rate found by Coupon Fast Facts (2009). This might indicate that participants liked the coupon or they were highly involved with the experiment.
Based on the results no specific frame can be recommended to Groupon. This
might also be the reason why Groupon has all three frames mentioned on the website per coupon. Camerer and Weber (1992) found that consumers tend to prefer highly certain outcomes to outcomes that are not precisely known. Having all three frames on the
website will result in highly certain outcomes because consumers get the information provided in all different ways. Hence, the coupon is evaluated more favorably.
4.4 Limitations
The experiment had several limitations. First, participants were only given the opportunity to order a coffee or tea with a discount of 0.60 euro cents. Therefore the participants had no choice in which coupon to purchase. In a normal circumstance,
consumers are exposed to lots of coupons and discounts from which to choose from. Second, only a low priced product was used, i.e. a coffee or tea for 1.20 euros. Chen et al. (1998) state that there are different effects of the different frames when using a low or high priced product. However, because actual behavior was measured and giving a
discount on a high priced product would lead to too high costs only a low priced product was used. Third, only a fixed face value was used, i.e. 50%. A deeper discount might have lead to different results. For example, DelVecchio et al. (2007) argue that a deeper
5. General discussion
The research question of this study was:
What is the effect of coupon characteristics on coupon redemption in an on-line setting?
The results show that a higher face value, the lower the regular price and the longer the expiration duration lead to a higher number of coupons redeemed. Furthermore, hedonic products and coupons that can be redeemed in one’s own city are redeemed more often. The results show that the way the discount on the coupon is framed has no effect
on coupon redemption.
The time on the website and the specific city on the website are the most important nominal variables. The specific day on the website, the type of product and the
distance to the redemption location are less important. For the metric variables, the percentage-off discount of the regular price is almost two and a half times more important than the expiration date.
An interesting finding is that the three control variables are the three most important nominal variables. The five main independent variables were all based on literature and were therefore included in the model. However, these variables were based
on the traditional way coupons were distributed. Because this study had a focus on an on-line setting where consumers buy coupons onon-line these control variables were added. Because selling coupons in an on-line setting is quite new, these variables have not been empirically investigated. The finding that these three variables are in fact the most
Important managerial implication can be given. First, recommendations are given for on-line coupon providers like Groupon. Second, recommendations are given for companies thinking partnering with on-line coupon providers.
One of the most important findings is that consumers tend to prefer lower priced, hedonic products. Therefore, Groupon should try to attract companies that offer these kinds of products. For example, restaurants and cheaper beauty products and services like
massages. Groupon could also demand a higher provision when dealing with these kinds of companies because it is expected that more coupons are sold by these companies. Based on the results no specific frame can be recommended to Groupon. Therefore, having all three frames on the website might be the right course of action. Finally, the
time on the website is one of the most important variables. When a coupon is displayed on the website for 48 hours instead of 24 hours the number of coupons redeemed is almost doubles. This shows that the coupon redemption does not decline over time. If this
trend continues after 48 hours is outside of the scope of this study. Therefore, making predictions what will happen after 48 hours is not appropriate. This study also shows that coupons displayed on weekdays are redeemed more often than coupons displayed on weekends. Therefore, Groupon should demand a higher provision when companies want
to have their coupon displayed on weekdays. Finally, when a company that provides a coupon is located in a smaller town near to the vicinity of a larger town, then the coupon should be displayed in both cities. This is recommended because the size of the specific
For companies thinking partnering with on-line coupon providers there are also several important recommendations. The coupon should be displayed on the website on weekdays, for 48 hours, in one’s own city and in a very large city if that is in the near
vicinity. Also, offering a high face value and a longer expiration duration is advised.
There are several areas for further research. Another study can be done in
collaboration with Groupon. This gives the opportunity to collect consumer characteristics, including coupon proneness and demographic information. With demographic information like where someone lives, the “distance to redemption location” variable can be calculated with less bias. Furthermore, data can be collected
from other website, in other countries and over longer periods. Finally, another experiment regarding framing can be done with a low and high priced product and with a steep and deep discount. Furthermore, an experiment can be set up where participants
References
Bawa, K., & Shoemaker, R.W. (1987). The coupon-prone consumer: some findings based on purchase behavior across product classes. Journal of Marketing, 51,
99-110.
Ben-Zion, U., Aharon, H., & Spiegel, U. (1999). The optimal face value of a discount
coupon. Journal of Economics and Business, 51, 159-174.
Blattberg, R.C, Eppen, G.D., & Lieberman, J. (1981). A theoretical and empirical evaluation of price deals for consumer nondurables. Journal of Marketing,
45,116-129.
Blattberg, R.C., & Neslin, S.A. (1990). Sales promotion: concepts, methods, and
strategies. Englewood Cliffs, NJ: Prentice Hall.
Blundo, C., Cimato, S., & De Bonis, A. (2005). Secure E-coupons. Electronic Commerce
Research, 5, 117-139.
Brock, T.C., & Green, M.C. (2003). Persuasion: psychological insights and perspectives. Sage publications: Thousands Oaks, CA.
Camerer, C.F., & Weber, M. (1992). Recent developments in modeling preferences: uncertainty and ambiguity. Journal of Risk and Uncertainty, 5 325–70.
Cameron, A. C., & Trivedi, P.K. (2005). Microeconometrics: Methods and Applications.
New York, USA, Cambridge University Press.
Chakraborty, G., & Cole, C. (1991). Coupon characteristics and brand choice.
Psychology and Marketing, 8, 145-159.
Chen, S.S., Monroe, K.B., & Lou, Y. (1998). The effects of framing price information messages on consumers' perceptions and purchase intentions. Journal of Retailing,
74, 353-372
Coupon fast facts. (2009). Coupon Fast Facts by inmar. Retrieved July 10, 2009, from www.couponinfonow.com/coupons/coupon-fast-facts.cfm
Coupon Information Corporation. (2008). What is a coupon? [Press release]. Retrieved June 5, 2008, from www.cents-off.com
Darke, P.R., & Chung, C.M.Y. (2005). Effects of pricing and promotion on consumer perceptions: it Depends on how you frame it. Journal of Retailing, 81, 35–47.
comparative price advertisements. Journal of Marketing Research, 18, 416-427.
DelVecchio, D., Krishnan, H.S., & Smith, D.C. (2007). Cents of percent? The effects of
promotion framing on price expectations and choice. Journal of Marketing, 71, 158-170.
DelVecchio, D., Lakshmanan, A., & Krishnan, H.S. (2009). The effects of discount location and frame on consumers’ price estimates. Journal of Retailing, 85, 336-346.
Dickson, P.R., & Sawyer, A.G. (1990). The price knowledge and search of supermarkets shoppers. Journal of Marketing, 54, 42-53.
Gleitman, H., Fridlund, A.J., & Reisberg, D. (2003). Psychology. W.W. Norton & Company, Inc: New York, NY.
Han, K., & Sung, Y. (2006). American Academy of Advertising Annual Proceedings.
Hill, R. C., Griffiths, W.E., & Judge, G.G. (2001). Undergraduate Econometrics. 2nd edition, USA, John Wiley & Sons, Inc.
Hutton, J., & Francis, M. (2002). Marketing communications: Integrated theory, strategy,
Inman, J.J., & McAlister, L. (1994). Do coupon expiration dates affect consumer behavior. Journal of Marketing Research, 31, 423-428.
Jones, J.M., & Zufryden, F.S. (1981). Relating deal purchases and consumer
characteristics to repeat purchase probability. Journal of the Market Research
Society, 23, 84-99.
Kahneman, D., & Tversky, A. (1979). Prospect theory: an analysis of decision under risk.
Econometrica, 47, 263-291.
Krishna, A.R.B., Lehman, D.R., & Yuan, H. (2002). A meta-analysis of the impact of price presentation on perceived savings. Journal of Retailing, 78, 101–118.
Kuehn, A.A., & Rohloff, A.C. (1967). Consumer response to promotions. In: Robinson, P.J. (Ed.), Promotional Decisions Using Mathematical Models, Boston, Allyn and Bacon Inc., 145-148.
Leone, R. P., & Srinivasan, S.S. (1996). Coupon face value: its impact on coupon redemptions, brand sales, and brand profitability. Journal of Retailing, 72,
273-289.
information before and after consuming the product. Journal of Consumer
Research, 15, 374-378.
Lichtenstein, D.R., Burton, S., & Karson, E.J. (1991). The effect of semantic cues on consumer perceptions of reference price ads. Journal of Consumer Research, 18, 380–91.
Lichtenstein, D.R., Netemeyer, R.G., & Burton, S. (1990). Distinguishing coupon proneness from value consciousness: an acquisition-transaction utility theory perspective. Journal of Marketing, 54, 54-67.
Lovelock, C., & Gummesson, E. (2004). Whither services marketing? In search of a new paradigm and fresh perspectives. Journal of Services Research, 7, 20-41.
Martin, P. (1991). Coupon expiration accelerates. Adweek’s Marketing Week, 32, 31.
NCH. (2008). NCH coupon redemption trends. Retrieved June 15, 2008, from https://
www.nchmarketing.com/ResourceCenter/CouponKnowledgeStream2.aspx?id=29 79
Nickels, W.G., & Wood, M.B. (1997). Marketing: relationships, quality, value, New York, NY, Worth Publishers.
Nunes, J.C., & Dréze, X. (2006). Your loyalty program is betraying you. Harvard
Business Review, 84,124-131.
Puto, C. P. (1987). The framing of buying decisions. Journal of Consumer Research, 14, 301-315.
Reibstein, D.J., & Traver, P.A. (1982). Factors affecting coupon redemption rates.
Journal of Marketing, 46, l02-113.
Rhymer, K.N., Skinner, C.H., Jackson, S., McNeill, S., Smith, T., & Jackson, B. (2002).
The 1-minute explicit timing intervention: the influence of mathematics problem difficulty. Journal of Instructional Psychology, 29, 305–311.
Roehm, M.L., & Roehm Jr., H.A. (2011). The influence of redemption time frame on
responses to incentives. Journal of the Academy of Marketing Science, 39, 363-375.
Shoemaker, R.W. (1979). An analysis of consumer reactions to product promotions. Proceedings of the American Marketing Association, Chicago, IL: AMA, 244-248.
Shoemaker, R.W., & Tibrewala, V. (1985). Relating coupon redemption rates to past purchasing of the brand. Journal of Retailing, 25, 40-47.
Strahilevitz, M.A., & Myers, J.G. (1998). Donations to charity as purchase incentives: How well they work may depend on what you are trying to sell. Journal of
Consumer Research, 24, 434-446.
Sullivan, J. (2008). Mobile coupon acceptance by age, for casual dining restaurants. Working paper.
Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of
choice. Science, 211, 453-458.
Tversky, A., & Kahneman, D. (1991). Loss aversion in riskless choice: a reference-
dependent model. Quarterly Journal of Economics, 106, 1039-1061.
Wang, X. (1998). The impact of coupon characteristics on redemption probability,
Masters Thesis, Purdue University, West Lafayette, IN.
Winkler, R.L. (1991). Ambiguity, probability, preference, and decision-analysis. Journal
Yin, W., & Dubinsky, A.J. (2004). Framing effects of coupon face value on coupon redemption: a literature review with propositions. Journal of Marketing
Management, 20, 877-896.
Zeithaml, V.A. (1981). How consumer evaluation processes differ between goods and
