Centre for Marketing

HEDONIC AND UTILITARIAN CONSUMER

BENEFITS OF SALES PROMOTIONS

Pierre Chandon

Brian Wansink

Gilles Laurent

Centre for Marketing Working Paper

No. 98-203

December 1998

Pierre Chandon is Assistant Professor at London Business School, e-mail:pchandon@lbs.ac.uk. Brian Wansink is Associate Professor of Marketing at theUniversity of Illinois at Urbana-Champaign, e-mail: wansink@uiuc.edu. Gilles Laurant isProfessor of Marketing at the Groupe HEC, e-mail: laurent@gwsmtp.hec.fr. This paperis based in part on Pierre Chandon’s dissertation and has benefited from the comments ofDonnie Lichenstein, Laurence Capron, James Hess, Stephen Hoch, Scott Neslin andTerrance Shrimp.

London Business School, Regent's Park, London NW1 4SA, U.K.Tel: +44 (0)171 262-5050 Fax: +44 (0)171 724-1145

pchandon@lbs.ac.uk http://www.lbs.ac.uk

Copyright © London Business School 1998

Hedonic and Utilitarian Consumer Benefits of Sales Promotions

Are monetary savings the only explanation for consumer response to a sales

promotion? To address this question, we develop and test a framework which shows that

sales promotions can provide consumers with three important hedonic benefits

(opportunities for self-expression, entertainment, and exploration) and three utilitarian

benefits (savings, higher product quality, and improved shopping convenience). We

further show in two choice experiments that a promotion’s effectiveness is determined by

the range of benefits it delivers and the congruency these benefits have with the promoted

product. By showing that sales promotions can have consumer benefits beyond monetary

savings, this research has implications for designing more effective promotions and for the

debate about the merits of everyday-low-price policies. It further suggests that behavioral

researchers should qualify consumer deal-proneness by the benefit sought, and that choice

modelers account for the differing effectiveness of monetary and nonmonetary promotions

across utilitarian and hedonic categories.

1

Hedonic and Utilitarian Consumer Benefits of Sales Promotions

A paradox of sales promotions is that their widespread growth is accompanied by

constant controversy. Marketers and academics often view the reliance on sales

promotions, especially monetary promotions, as a sub-optimal consequence of price

competition caused by myopic management (Buzzel, Quelch and Salmon 1990). In the

short-run, these critics argue that the proliferation of monetary promotions erodes their

capacity to “rent” market share and often renders the promotions unprofitable (Abraham

and Lodish 1990; Kahn and McAlister 1997). In the long run, they argue that these

promotions erode price sensitivity and brand equity—both with retailers and consumers

(Mela, Gupta and Lehman 1997; Papatla and Krishnamurthi 1996). As a result, while

some industry experts are calling for more effective and cost-efficient promotions that rely

less on price (Promotion Marketing Association of America 1994), others recommend

eliminating most promotions by switching to an every-day-low-price policy (Kahn and

McAlister 1997; Lal and Rao 1997).

Surprisingly, few arguments in this debate reflect the consumer’s perspective.

Most models simply assume that monetary savings are the only benefit of sales

promotions—thus making EDLP (every-day-low price) an efficient solution for passing

these savings to the consumer (e.g., Lal and Rao 1997). An often-ignored consideration

in this debate is whether sales promotions have consumer benefits beyond monetary

savings.

The absence of research on the consumer benefits of sales promotions is all the

more surprising since we know that purchase behavior is motivated by many utilitarian and

hedonic benefits (Babin, Darden and Griffin 1994). The existence of multiple consumer

benefits may help explain some puzzling findings, such as the surprisingly large response

to insignificant price reductions (Hoch, Drèze and Purk 1994; Inman, McAlister, and

2

Hoyer 1990), or the failure to redeem coupons responsible for the purchase decision

(Bawa and Shoemaker 1989; Dhar and Hoch 1996; Soman 1998). More generally, a

narrow focus on the monetary benefits of sales promotions may be partly to blame for the

lack of research on non-monetary sales promotions such as free gifts or sweepstakes,

which are used by 70% of grocery and health and beauty care manufacturers (Cox Direct

1997).

The objectives of this research are to 1) provide an integrative framework of the

hedonic and utilitarian consumer benefits of consumer sales promotions1 and 2) show how

these benefits explain the differences in the effectiveness of monetary and non-monetary

promotions across different types of products. We begin by showing how fragmented

explanations for consumer “deal-proneness” can be integrated into a multi-benefit

framework of sales promotions. We examine the implications of this framework by

developing a benefit congruency hypothesis, which predicts the types of products for

which monetary and non-monetary promotions are most effective. The empirical analyses

consist of a series of measurement studies validating the multi-benefit framework as well

as three choice experiments conducted across two countries. The discussion explores the

implications of the multi-benefit framework and the congruency hypothesis for managers,

consumer researchers, and modelers.

A Multi-Benefit Framework of Sales Promotions

Why do Consumers Respond to Sales Promotions?

1 Building on Blattberg and Neslin (1990), we define a consumer sales promotion as a temporary andtangible incentive intended to have a direct impact on consumer behavior. This definition encompassesmonetary techniques (e.g., coupons, temporary price cuts, rebates, bonus packs and “buy-one-get-one-free” offers, or BOGOs), as well as non-monetary techniques (e.g., free gifts, free samples, contests andsweepstakes).

3

Behavioral research on sales promotions has tended to focus on the demographics

(i.e., income and home ownership) of deal-prone consumers (Bawa and Shoemaker 1987;

Blattberg et al. 1978; Narasimhan 1984; Webster 1965) and on the identification of

personal traits such as coupon-proneness, value-consciousness or market mavenism (Feick

and Price 1987; Inman, McAlister and Hoyer 1990; Lichtenstein, Netemeyer and Burton

1990 and 1995; Mittal 1994). These studies offer a coherent portrait of the demographic

and psychographic characteristics of deal-prone consumers (for a review, see Blattberg

and Neslin 1990, p. 65-82). On the other hand, these studies have left the issue of the

existence and of the role of the benefits of sales promotions largely unexplored. As a

result, most models of consumer response to sales promotions assume that monetary

savings are the only benefit motivating consumers to respond to sales promotions (for a

review, see Blattberg and Neslin 1993).

Monetary savings cannot fully explain why and how consumers respond to sales

promotions. Why do consumers respond more to a coupon than to a similarly advertised

temporary price reduction offering the same monetary incentive (Dhar and Hoch 1996;

Schindler 1992)? Why do they respond to insignificant price reductions (Hoch, Drèze and

Purk 1994; Inman, McAlister, and Hoyer 1990), and why do they switch brands because

of a coupon, but then do not redeem it (Bawa and Shoemaker 1989; Dhar and Hoch 1996;

Soman 1998)? To account for these findings, researchers have advanced explanations

related to fairness (Thaler 1985), to achievement motives (Darke and Freedman 1995) or

to self-perception (Schindler 1992). However, the amount of support for these

explanations is also limited. For instance, the achievement and self-perception arguments

are contradicted by the finding that “lucky” bargains are enjoyed as much as those

acquired skillfully (Darke and Freedman 1995) and that consumers may feel embarrassed

to buy a promoted brand (Simonson, Carmon, and O’Curry. 1994). Similarly, if the

reference price is the “fair” price, fairness considerations should dissuade consumers from

4

responding to a promoted price that would be lower than the fair reference price, or would

not be fair to the seller. Finally, existing empirical analyses examine the consequences of

these non-monetary benefits without directly measuring these benefits or, when they do

(e.g., Shimp and Kavas 1984), they use single-item measures that preclude the study of

their construct validity.

In summary, the contributions of the personality studies, the parsimony of the

economic perspective, and the existing work on the non-monetary benefits of sales

promotions have greatly contributed to our understanding of sales promotion. An

integrated study of the benefits of sales promotions, however, would help reconcile the

fragmented nature, and the empirical and conceptual limitations of these seemingly

disparate studies.

Consumer Benefits of Sales Promotions

Drawing on Keller (1993), the benefits of sales promotion can be defined as the

perceived value attached to the sales promotion experience. This promotion experience

can include both promotion exposure (e.g., seeing a promotion on a product) and usage

(e.g., redeeming a coupon or buying a promoted product). This definition implies that

consumers respond to sales promotions because of the positive experiences they provide,

or, following Holbrook’s (1994) definition, because of their value to the customer.

To develop a framework of the different consumer benefits of sales promotions,

the literature on consumer response to sales promotions, customer value and hedonic

consumption was elaborated with a series of in-depth consumer interviews. Each

interview lasted 90 minutes and involved a variety of projective techniques designed to

prompt subjects to elicit the attributes, benefits and values provided by, or sought in, sales

promotions. These included free elicitation, Kelly’s (1955) repertory grid, means-end

chains, sentence completion, customer prototyping, retrospective protocols and role-

5

playing. The respondents were heterogeneous with respect to gender, residency, age,

education, and nationality. Transcripts from the interview were coded by each author

separately, and grouped into different consumer benefits. This list was then given to a

panel of experts from three sales promotion agencies for validation and further

suggestions.

INSERT TABLE 1 AND 2ABOUT HERE

The result of this inductive investigation is the six-benefit framework presented in

Table 1 and in Table 2. These tables provide a definition of each benefit, supporting

research studies, and interview excerpts. Table 1 indicates that one of the benefits of sales

promotions for the consumer is the monetary savings they provide (the “savings” benefit).

However, sales promotions may also enable consumers to upgrade to higher-quality

products by reducing the price of otherwise unaffordable products (the “quality” benefit),

which will often lead to higher price being paid. Because sales promotions signal the

availability of the brand at the point of sales and advertise its promotional status, they can

reduce consumer search and decision costs, and therefore improve shopping convenience

(the “convenience” benefit). Further, sales promotions can enhance consumers’ self-

perception of being a “smart” or a “good” shopper and provide an opportunity to reaffirm

their personal values (the “value-expression” benefit). Because sales promotions create an

ever-changing shopping environment, they also provide stimulation and fulfill consumers’

need for information and exploration (the “exploration” benefit). Finally, sales promotions

are often simply fun to see or to use (the “entertainment” benefit). It is worth noting that

the last five benefits can be achieved above and beyond any reduction in spending, and are

therefore theoretically independent of monetary savings.

These six benefits can be more parsimoniously classified. Most classifications of

the different types of customer benefits and value start with the distinction between

6

utilitarian (extrinsic) and hedonic (intrinsic) benefits (Holbrook 1994). Utilitarian benefits

are primarily instrumental, functional and cognitive; they provide customer value by being

a means to an end. Hedonic benefits are non-instrumental, experiential and affective; they

are appreciated for their own sake, without further regards for their practical purposes

(Hirschman and Holbrook 1982, p. 100). Babin, Darden and Griffin (1994) showed that

this distinction applies to shopping since shopping provides utilitarian benefits (by helping

consumers find and buy the best products efficiently) as well as hedonic benefits (by

creating entertainment and raising self-esteem). Similarly, the benefits of sales promotions

can be utilitarian when they help consumers maximize the utility, efficiency, and economy

of their purchase process, and hedonic when they provide intrinsic stimulation, fun and

self-esteem.

Using these definitions, the savings, quality, and convenience benefits of sales

promotions can be classified as utilitarian since they help consumers increase the

acquisition utility of their purchase and the efficiency of the shopping experience. On the

other hand, the entertainment and exploration benefits of sales promotions are hedonic

since they are intrinsically rewarding and related to experiential emotions, pleasure and

self-esteem. As Table 2 shows in more detail, we suggest that the value-expression benefit

of sales promotions is different, since it entails both utilitarian as well as hedonic

dimensions. On one hand, buying a promoted product can provide the intrinsic

satisfaction of behaving according to one’s principles (e.g., being a good shopper or a

smart shopper). On the other hand, by allowing an opportunity for one to express their

values, it can increase one’s self-esteem and prestige, and in turn achieve respect or group

affiliation.

In summary, a multi-benefit framework extends existing models, which focus on

monetary savings as the only consumer benefit of sales promotions. In contrast, this

framework proposes that sales promotions can provide six different benefits for the

7

consumer, and that these benefits can be classified according to their utilitarian or hedonic

dimension. To examine whether the added complexity of this framework is justified, we

now examine its ability to better predict the effectiveness of a sales promotion. We draw

on the framework to formulate a benefit-congruency hypothesis that predicts when a sales

promotion is more effective based on the type of benefits it offers. Specifically, we use the

multi-benefit framework to understand how monetary and non-monetary promotions

work, and for which type of product they are most effective.

Implications for the Effectiveness of Monetary and Non-Monetary Promotions

Monetary sales promotions (such as sales, coupons, rebates, or BOGOs) provide

monetary savings, enable consumers to buy a higher-quality product, and reduce search

and decision costs by serving as surrogate shopping lists, attention-getting devices or

value signals. These promotions typically offer few–if any–hedonic benefits because of

their pure monetary nature, the efforts they require to be redeemed, and, more generally,

their nature as a means to a rational end (savings, quality, efficiency). On the other hand,

non-monetary promotions such as free gifts, free samples, sweepstakes and contests

usually offer little savings, quality, or convenience because they do not reduce prices and

are time consuming. On the other hand, their probabilistic dimension can be entertaining,

and their free gifts and samples can provide stimulation and novelty. We therefore

hypothesize:

H1: Compared with monetary promotions, non-monetary promotions providemore hedonic benefits and less utilitarian benefits.

The Benefit Congruency Hypothesis

According to combinatorial models of attitude formation and to utility theory,

consumers will evaluate a promoted product (promotion and product combined) based on

8

its benefits weighted by the importance of these benefits. The weighting of the benefits

varies across individuals, time, and products (Eagly and Chaiken 1993; Meyer and Kahn

1991). As a result, the effectiveness of promotions providing different types of benefits

varies according to the weighting of these benefits. Simply stated, the benefit congruency

hypothesis proposes that sales promotions are more effective when they provide the

benefits that have the largest weight in the evaluation of purchase alternatives.

Many theories of attitude change formulate a similar “matching” hypothesis. For

instance, Fishbein and Ajzen (1975) argue that persuasion attempts are more effective

when they address the beliefs that are the most important antecedent of an attitude.

Functional theories contend that persuasion is enhanced when a persuasive message

emphasizes the utilitarian or hedonic function that provides the motivational basis of the

attitude to be modified (Katz 1960). Finally, the compatibility principle (Tversky, Sattath,

and Slovic 1988) suggests that consumers change the weighting of the attributes of an

alternative so as to give more weight to the type of attributes that are the most important

in a given evaluation mode.

We should expect, therefore, that the utilitarian benefits of a given choice

alternative are given more weight when consumers make a utilitarian decision, and that the

hedonic benefits are given more weight when making a hedonic decision. One way of

inferring the utilitarian or hedonic nature of the decision is to look at the product being

considered (Mano and Oliver 1993; Shavitt 1990). For instance, Shavitt (1990) showed

that the attitude toward a utilitarian product (an air-conditioner) was more influenced by

ads emphasizing utilitarian rather than hedonic benefits, and that the reverse was true with

a hedonic product (coffee).

We argue this congruency effect will also occur with promotions. That is, when

consumers evaluate a promotion when shopping for a utilitarian product (e.g., a battery),

they place a greater weight on its utilitarian benefits than on its hedonic benefits. As a

9

result, they will be more influenced by a promotion providing strong utilitarian benefits

(such as a monetary promotion) than by a promotion providing strong hedonic benefits

(such as a non-monetary promotion). Conversely, when evaluating a promotion for a

hedonic product (e.g., when choosing a wine or a dessert), consumers place a greater

emphasis on its hedonic benefits than on its utilitarian benefits, and will thus be more

receptive to a hedonic non-monetary promotion than to a utilitarian monetary promotion2.

More formally:

H2a: Monetary promotions are more effective for products purchased forutilitarian reasons than for hedonic ones.

H2b: Non-monetary promotions are more effective for products purchased forhedonic reasons than for utilitarian ones.

Measuring and Validating the Benefits of Sales PromotionsTo measure and validate the benefits of sales promotions, we conducted three-

measurement studies following Churchill’s (1979) scale development paradigm. Study 1

develops a scale measuring the six benefits of sales promotions. Study 2 validates the

construct validity of each benefit and the hedonic and utilitarian classification vis-à-vis the

prevailing unidimensional model. Study 3 examines the predictive validity of each benefit

and tests hypothesis H1 regarding the main benefits provided by monetary and non-

monetary promotions.

Study 1: Measuring the Benefits of Sales Promotions

Item generation and stimuli selection. To generate items that measured the ability

of a promotion to provide each benefit, ideas were obtained from the existing literature

2 Similarly, Strahilevitz and Myers (1998) found that donations to charities (a type of non-monetarypromotion) were more effective when offered for a hedonic product than for a utilitarian product. Theyattribute their finding to the affect-based (rather than use-based) complementarity of the product and thepromotion, and, in particular, to the fact that charity donations reduce the feelings of guilt caused by thepurchase of many hedonic products. Our rationale is similar except that it is not restricted to affect andextends beyond guilt-reducing promotions.

10

and from the interviews described earlier. The resulting 200 items were reduced to 45

through a discussion with experts from three promotion agencies. Twenty-four promotion

“exemplars” (real coupons, on-pack promotions, color pictures of advertised promotions,

sweepstakes, etc.) for frequently purchased consumer packaged goods were selected from

a database of 200 promotions maintained by one of the authors so as to encompass the full

array of promotion techniques commonly defined in the literature. This method was

preferred to the verbal descriptions used in many studies because they had higher external

validity and were able to describe some of the promotion techniques without implying

positive or negative associations. Another advantage of the picture-based method is that it

does not “sanitize” the deceptive devices (such as fine print), nor alter the aesthetics of

the promotion, which may trigger the hedonic benefits that should be measured.

Data collection. To develop and purify a scale of promotion benefits, two

convenience samples consisting of graduate students (n=37) and staff (n=28) were

recruited at a major French university through a lottery offering two $40 prizes. We asked

respondents to consider each promotion as an exemplar of an unspecified category of sales

promotions, then indicate their level of agreement with 45 statements such as, “With this

type of promotion, I can get new ideas of things to buy” on a five-point Likert scale

ranging from “strongly disagree” to “strongly agree”. The respondents could also

choose a “does not apply” category if they thought that the statement was irrelevant to

the promotion. To control for primacy and recency effects, the ordering of the items and

of the stimuli was counterbalanced according to a systematic sampling plan. Each subject

in the first and second sample evaluated, respectively, two or three promotions.

Analysis and results. The data was aggregated across consumers and promotions.

The resulting 45x45-correlation matrix was analyzed through a principal component

analysis with oblique rotation, using a “state” (vs. a “trait") analysis in a procedure

similar to the one used by Aaker (1997) to identify brand personality factors. In both

11

samples, a forced six-factor solution satisfactorily reproduced the six hypothesized

benefits. In both samples, however, the eigenvalue criterion suggested a four-dimension

solution (explaining 70% of the variance in the first study and 69% in the second study).

The items measuring the quality benefit were mixed with the savings benefit and those

measuring the value-expression benefit were scattered over the other dimensions. These

two dimensions were nevertheless retained for subsequent testing because the pattern of

results was consistent with our framework and with the literature (Blattberg and

Wisniewski 1989; Schindler 1992). Each subscale was then individually factor analyzed to

select a reliable set of three indicators for each benefit.

Study 2: Validating the Benefits of Sales Promotions

Procedure. Although principal component analyses are useful in the exploratory

phase of scale development and purification, confirmatory factor analyses are required to

validate the dimensionality and the higher-order structure of the model (Gerbing and

Anderson 1988). To accomplish this, Study 2 was conducted using the same procedure as

in the previous study. We asked 118 graduate students and staff at a different French

university to evaluate four promotions using the 18-items scale presented in Table 1 and 2.

The stimuli were randomly selected out of 21 new promotions, and 3 promotions used in

the prior study.

The covariance matrix was analyzed using AMOS 3.6 and maximum likelihood

estimation (Arbuckle 1997)3. Following the procedure recommended by Bollen (1989),

we started with the analysis of the six congeneric models. The coefficients of

determination of each scale ranged from .70 to .92; t-values of loadings were all highly

significant; and the error variance of 15 out of 18 indicators was smaller than the variance

3 We used MLE because it is robust to departures from multi-normality (Bollen 1989). A comparisonwith a bootstrap analysis and with WLS estimates on polychoric correlations found no significantdifferences.

12

extracted from the construct. These results show that each scale exhibits satisfactory levels

of internal consistency and reliability given the high number of new constructs tested, the

low number of items per construct, and the heterogeneity in the promotional stimuli used.

First-order factor analyses. The six scales also serve to test the convergent and

discriminant validity of each benefit. For this purpose, first-order confirmatory factor

analyses were used to compare a unidimensional model with the proposed six-dimension

model, in which each benefit is treated as a separate yet correlated construct. The results

show that the proposed six-benefit model has an adequate fit given the number of

parameters (51) and the sample size (461) (c2120=422; p<0.01, GFI=.906, AGFI=.867,

IFI=.918, RMSEA=.0744). The amount of variance extracted for each of the six benefits

is higher than the average they share with any other benefit, which demonstrates

discriminant validity (Fornell and Larcker 1981). The single-benefit model performs

significantly worse (c2135=1806; p<0.01, GFI=.631, AGFI=.532, IFI=.547,

RMSEA=.164), as evidenced by the significant improvement in chi-square with the multi-

factor model (c215=1384, p<0.01). The adjustment of the proposed model is also better

than any of the 15 models in which one pair of benefits is forced to be perfectly correlated

(the difference in c2 with 1 degree of freedom ranged from 50 to 513, with an average of

209, p<0.01). Overall, the data strongly support the existence of six correlated but distinct

benefits.

INSERT FIGURE 1 ABOUT

HERE

Second-order factor analyses. As predicted, the six benefit are not orthogonal

(correlation’s range from -.05 to .68 with an average of .44). In order to test the

4 The high value of the � VWDWLVWLF VKRXOG QRW EH JLYHQ WRR PXFK ZHLJKW JLYHQ WKDW WKH

K\SRWKHVLV RI SHUIHFW ILW LV TXLWH XQUHDOLVWLF LQ PRVW HPSLULFDO ZRUN ZLWK WHVW GDWD�

HVSHFLDOO\ ZLWK KLJK VDPSOH VL]HV �%ROOHQ ������ 7KH ILW LV ZLWKLQ WKH UDQJH XVXDOO\

UHJDUGHG DV DFFHSWDEOH Error! Bookmark not defined. .

13

utilitarian and hedonic higher-order structure of the six benefits, we estimated the model

with two second-order factors shown in Figure 1, and compared it with a single second-

order factor model (see Bollen 1989; Lichtenstein, Netemeyer, and Burton. 1995). All fit

indices support the two higher-order factor model over the unidimensional solution

(c2127=565; p<.01, GFI=.881, AGFI=.840, IFI=.882, RMSEA=.087 for the two-

construct model and c2128 = 830; p<.01, GFI=.851, AGFI=.801, IFI=.810,

RMSEA=.109 for the one-construct model). The difference in c2 between the two models

is highly significant (c2= 265, DF=1, p<0.01). Again, the proposed model provides the

best fit of all possible classifications of benefits into two higher-order constructs. The

correlation between the two factors is high (r=.67) but comparable to the one reported by

Batra and Ahtola (1990) (r=.55). However, the correlation doesn’t include the value of 1

in its confidence interval (estimated by bootstrap analyses) and the amount of variance

shared between the second-order factors is lower than the average variance extracted for

each factor. These two results support the discriminant validity of the two utilitarian and

hedonic factors.

Overall, the two measurement studies support the multi-benefit framework by

showing that consumers perceive the six benefits of sales promotions not only as

significantly different, but also as related to two different higher-order utilitarian and

hedonic dimensions. We now examine the predictive validity of the framework by

estimating (1) the extent to which the six benefits predict the overall evaluation of the

promotion and (2) testing Hypothesis H1 regarding the primary benefit of monetary and

non-monetary promotions.

Study 3: How Promotion Benefits Influence Overall Evaluation

Procedure. Three items measuring the overall evaluation of the promotion were

collected in the second study but not yet discussed (“I like this type of promotion a lot”,

14

“I wish there were more promotions like this,” and “With this type of promotion, I feel

like buying the product”). These items were selected based on a pre-test and the reliability

of the measure is conventionally acceptable (Cronbach’s α=.83). Since the influence of

hedonic benefits is often non-linear (Hirschman and Holbrook 1982), we checked for non-

linearities in the model. In their absence, we decided to use a linear structural equation

model using AMOS 3.6. We regressed the overall evaluation of the promotion (modeled

as a latent construct with the three indicators discussed above) on the six benefits

measured with the items described in Table 1 and 2.

In order to explore the differences between monetary and non-monetary

promotions, we estimated a multi-group model allowing for different regression

coefficients and indicator means for each sub-sample of promotions (see Bollen 1989, p.

306). The sub-sample of monetary promotions consists of five temporary price

reductions, four coupons, three rebates, and two multi-unit packs, for a total of 269

observations. The sub-sample of non-monetary promotions consists of two free gifts, two

free samples, and three sweepstakes, for a total of 192 observations. Because the

questionnaire asked respondents to use each specific promotion as a category exemplar,

we tentatively generalize the results to monetary and non-monetary promotions.

Predictive analyses. All models were estimated using the maximum-likelihood

estimator. A bootstrap analysis with 1000 replications showed that the ML estimates

were not subject to serious biases caused by multicolinearity or deviation from normality

assumptions. We found that a multi-group model in which the regression coefficients are

allowed to vary between monetary and non-monetary promotions exhibits a satisfactory fit

(c²354=931, IFI=.968, RMSEA=.060), and outperforms an aggregate model in which these

parameters are constrained to be equal across both groups (inc. χ2=29, df=4; p<.01).

Interestingly, allowing the correlations between the benefits to vary across the two

samples does not improve the fit of the model fit significantly. This shows that the multi-

15

benefit framework is robust and that the correlation between the benefits does not depend

on the selection of the promotions used to measure these benefits (inc. χ2=19, df=15;

p>.10).

Table 3 shows that monetary promotions are primarily evaluated on their utilitarian

benefits (savings and convenience) and on their value-expression benefits. On the other

hand, non-monetary promotions are evaluated on their hedonic (entertainment and

exploration) and value-expression benefits. Unexpectedly, the coefficient for quality is not

significant (and negative). This coefficient also exhibits a high instability in the bootstrap

analyses, probably because of multi-colinearity with the other utilitarian benefits. On the

other hand, value-expression is a very good predictor, and not only for monetary

promotions, as suggested by prior studies (Shimp and Kavas 1984; Schindler 1992). This

reinforces our conceptualization of value-expressive benefits as being both utilitarian and

hedonic. Overall, these results support our framework by showing that all benefits, except

savings, contribute to forming consumers overall evaluation of promotions.

INSERT TABLE 3 AND FIGURE 2 ABOUT

HERE

Primary benefits of monetary and non-monetary promotions. As predicted by H1,

we found that non-monetary promotions have a higher overall hedonic score than

monetary promotions (m=2.94 vs. 2.09) but a lower utilitarian score than monetary

promotions (m=3.57 vs. 2.09). H1 is formally supported by the fact that allowing the two

types of promotion to have different utilitarian and hedonic means results in a significant

improvement in model fit (inc. χ²=140; df =1; p<0.01). Surprisingly given the importance

of the convenience benefit for the evaluation of monetary promotions, in our study, both

types of promotions are indistinguishable on this benefit (cf. Table 2). Interestingly,

monetary promotions receive a higher score on the value-expression benefit. This

16

suggests that consumers gain more self-esteem from utilitarian benefits than from hedonic

benefits.

Figure 2 provides additional information by reporting the factor scores of the

different promotion techniques used in this study on the utilitarian and hedonic

dimensions5. It shows that non-monetary promotions are more heterogeneous than

monetary promotions and that their higher overall hedonic appeal is mainly due to free

gifts. On the other hand, sweepstakes are dominated by all other types of promotions;

they offer few utilitarian benefits and are also poorly rated in terms of hedonic benefits.

This finding may explain why sweepstakes are one of the few declining promotional

techniques (Cox Direct 1997).

Conclusion of Studies 1-3

The objective of the measurement studies was to develop a scale of sales

promotion benefits in order to validate the multi-benefit framework. The results show the

following:

1. Monetary savings are not the only consumer benefit of sales promotions.2. Consumers can distinguish between the six benefits hypothesized.3. These six benefits can be grouped according to their utilitarian or hedonic

nature.

While more refined or exhaustive classifications have yet to be examined, study 3

shows that the six benefits proposed in the framework provide a significant improvement

over the current emphasis on monetary savings alone. In fact, adding the five non-savings

benefits increases the amount of variance explained in the overall evaluation of non-

monetary promotion from .54 to .82, and even increases the evaluation of monetary

5 We computed second-order factor scores instead of estimating the utilitarian and hedonic means foreach technique, as we did when comparing monetary and non-monetary promotions, because of the smallnumber of observations available for each technique.

17

promotions from .71 to .79. Overall, the six hypothesized benefits are not only distinct

but also important in evaluating sales promotions.

These analyses also produce interesting side results. For instance, we found that

the correlation between the six benefits does not depend on the type of promotion being

evaluated. On the other hand, the factor score and the explanatory power of these benefits

vary significantly across the two types of promotions. As predicted, non-monetary

promotions provide more hedonic benefits and less utilitarian benefits than monetary

promotions and, with the exception of value expression which is a universal predictor,

each type of promotion tends to be evaluated on the basis of the benefit it provides the

most, utilitarian benefits for monetary promotions and hedonic benefits for non-monetary

promotions.

This section developed a scale of promotion benefit that can prove useful for

benchmarking or pre-testing purposes. It also showed that monetary and non-monetary

promotions are evaluated on different benefits. However, such a correlational study

cannot rigorously test the benefit congruency hypothesis because of the lack of control

over the stimuli and because of possible common-method biases. To examine the

effectiveness of monetary and non-monetary promotions for hedonic and utilitarian

products, we conducted two experimental studies. Study 4 tests the benefit congruency

for brands of comparable quality across two different populations. Study 5 examines the

effectiveness of monetary and non-monetary promotions on the competition between

brands with different levels of customer-based brand equity.

When are Monetary and Non-Monetary Promotions Most Effective?

Study 4: A Cross-national Test of the Benefit Congruency Hypothesis

18

Design and procedure. Following the procedure used by Simonson, Carmon and

O’Curry (1994) we used a 2 (monetary or non-monetary promotion) by 2 (utilitarian or

hedonic product) between-subject design with 4 within-subject replications6. Each

replication consisted of the use of a different pair of promotion for each of the 4 pairs of

products (see Table 4). The study was conducted in the US and in France using the same

design, procedure, promotions, and brands. To ensure consistency, the questionnaires

were separately translated and back translated. The two data collections were intended to

provide cross-cultural generalizations and no differences were hypothesized.

The subjects were 139 students aged 21-23 at a major French university and 51

students aged 20-22 at a major American university who were told that they were taking

part in a cross-cultural grocery shopping survey. In both countries, each subject had to

make a series of choices between pairs of hedonic or utilitarian brands, in which one brand

of the pair was offered with a monetary promotion and the other with a non-monetary

promotion. In order to strengthen the promotion manipulations, price information was

omitted. It was indicated that the two brands were priced similarly and were the only

brands available. We collected brand choice and purchase intentions for each pair of

products. Finally, subjects rated the hedonic and utilitarian nature of each product and

promotion.

INSERT TABLE 4 ABOUT

HERE

Stimuli. Based on pre-tests and prior research (Ratchford 1987; Laurent and

Kapferer 1985), ice cream and wine were selected for their hedonic benefits and 35mm

films and garbage bags were selected for their utilitarian benefits. In each category, the

choice set included two established brands: Ben & Jerry vs. Häagen Daaz ice-cream,

6 This study did not include a “no promotion” control condition since the objective was to compare theeffectiveness of monetary and non-monetary promotions across product categories rather than tocompare the impact of each of these promotions vis-à-vis no promotion at all.

19

Oakville Ranch vs. Clos du Bois white Californian wine, Fuji vs. Agfa film, and Glad vs.

Hefty garbage bags. Real brands were chosen to minimize consumers’ inferences about

the quality of promoted products. To maintain consistency with the measurement studies,

the monetary promotions consisted of two instant coupons and two rebates by mail, while

the non-monetary promotions consisted of two free gifts and two sweepstakes. The

stimuli were selected based on currently offered promotions in these four product

categories (see Table 4). The four pairs of promotions shown in Table 4 were rotated

between subject across the four product categories according to a Latin-square design.

Depending on the condition, the target brand of each category was promoted with either

the monetary or non-monetary promotion.

Results of Study 4

Manipulation checks showed that the monetary promotions chosen were all rated

as more utilitarian and less hedonic than their non-monetary counterparts (see Table 4).

Similarly, film and garbage bags were perceived as significantly more utilitarian and less

hedonic than either ice cream or wine (cf. Table 5). After testing for product category and

promotion differences, the choice and purchase intention data were aggregated at the

promotion type and products type level. Brand choices were then analyzed with a logit

regression and purchase intentions were analyzed with an analysis of variance. There were

three independent factors: product type (utilitarian or hedonic), promotion type (monetary

or non-monetary) and their interaction (see Table 5).

INSERT TABLE 5, TABLE 6 AND FIGURE 3 ABOUT

HERE

The interaction between promotion type and product type was significant in both

samples (B=.32, Wald=4.0, p<0.05 and B=.22, Wald=6.2, p<0.05 for, respectively, the

US and French data). Consistent with the benefit congruency hypothesis, brands

20

promoted with a non-monetary promotion were chosen more often in hedonic product

categories than in utilitarian product categories, and vice versa for monetary promotions

(see Table 6 and Figure 3). In other words, non-monetary promotions were more

effective for hedonic products than for utilitarian products and, conversely, monetary

promotions were more effective for utilitarian products than for hedonic products.

Remarkably, benefit congruency improves choice shares by the same amount (12 points) in

both countries.

Table 6 and Figure 3 also show a significant main effect for promotion type: in

both studies and across both product categories, monetary promotions were more

effective than non-monetary promotion. While this result depends entirely on the

magnitude of the promotions, the promotions were selected were typical of those offered

on these product categories. Finally, even though the stimuli were exactly the same, the

preference for monetary promotions is more pronounced for the US respondents than for

the French respondents (B=-.23, Wald=7.3, p<0.01). Switching from a non-monetary

promotion to a monetary promotion increases choice shares across both product

categories by 38 points for the US respondents (from 31% to 69%) but only by 16 points

for the French respondents (from 42% to 58%). This discrepancy echoes the higher

reliance of French marketers on non-monetary promotions (BIPP 1995).

Overall, study 4 provides strong evidence in favor of the benefit congruency

hypothesis across four executions of promotions in two countries. However, it is not

without limitations. The use of student respondents, the absence of price information, the

presence of two simultaneously promoted brands, and the use of a limited number of

promotions and products raise questions regarding the generalizability of the findings to

other settings and populations. Of particular interest is the question of the effectiveness of

sales promotions, and the generalizability of the benefit congruency hypothesis, to the

21

competition between brands with different levels of customer-based brand equity. Study 5

was conducted to address these limitations.

Study 5: Does Benefit Congruency Matter More For High Equity Brands?

The role of brand equity. Study 4 looked at the effectiveness of sales promotions

when competing brands have similar levels of consumer familiarity and perceived quality,

or customer-based brand equity (hereafter simply called “ brand equity”). Study 5 extends

the analysis to the more common case of competition between a high equity brand and a

low equity brand, say a national brand and a private label or a regional brand. Prior

studies suggest that national brands should randomly price discount in order to capture the

buyers of the national brand (Rao 1991) but the empirical evidence is weak. On one hand,

some studies found that sales promotions benefit higher-quality brands gain more than

lower quality brands (Blattberg and Wisniewski 1989, but see Bronnenberg and Wathieu

1997). On the other hand, other studies found that private labels are often as promotion

elastic as national brand (Dhar and Hoch 1997; Tourtoulou 1996). In addition, these

studies looked only at monetary promotions and none of them examined the effects of

benefit congruency.

Keller (1993) defines customer-based brand equity as the “differential effects of

brand knowledge on consumer response to the marketing of the brand.” We therefore

first expect that low equity brands will be less sensitive to any sales promotions than high

equity brands. We also expect benefit congruency to be less important for low equity

brands than for high equity brands. First, we expect monetary promotions to be little

effective for low equity and low price utilitarian brands—despite their benefit

congruency—because of their incapacity to attract the less price-elastic buyers of the high

equity utilitarian brand (cf. the cross-promotion asymmetry documented by Blattberg and

Wisniewski 1989). Second, we expect non-monetary promotions to be little effective for

22

low equity hedonic products because of their incapacity to attract the buyers of the high

equity hedonic brand who are, by definition, sensitive to the hedonic benefits of the

product. More generally, buyers of low equity products are, by definition, less involved

and interested in the benefits of the product and, therefore less sensitive to benefit

congruence than buyers of high equity products. If anything, buyers of low equity brands

should be interested in price. Formally:

H3a: Sales promotions are more effective for high equity brands than for lowequity brands.

H3b: Benefit congruency is more important for high equity brands than for lowequity brands. For low equity brands, monetary and non-monetarypromotions are equally effective for both utilitarian and hedonic brands.

Design and procedure. In study 5, subjects made five consecutive brand choices

for five pairs of products consisting of one well-known, high-quality brand and one lesser-

known brand of lower perceived quality priced at a 20% discount. We used a 2 (product

type) by 5 (promotion type) design with five within-subject replications. Unlike in Study

4, the promotion conditions included a control condition and only one brand of the pair

was promoted. The five promotions conditions are: (1) no promotion on high and low

equity brand, (2) monetary promotion on high equity brand only, (3) non-monetary

promotion on high equity brand only, (4) monetary promotion on low equity brand only

and (5) non-monetary promotion on low equity brand only. Depending on the design

treatment, there were two utilitarian products and three hedonic products or vice versa.

To evaluate the robustness of the results of study 4, we used a total of eight new

promotions: four monetary promotions (two coupons and two free product offers) and

four non-monetary promotions (two free gift and two sweepstakes). The order in which

the promotions, products, and type of target brand appeared was counterbalanced.

The procedure we used in study 5 is similar to the one used in study 4 except that

the questionnaires were mailed to 350 consumers in five states (CA, IA, IL, NH, and PA).

23

Of this sample 171 consumers responded in time to be included in the analysis. Sixty-one

percent of those surveyed were between the ages of 35-50; 77% were homeowners; 73%

had two or more children; 68% were female, and 58% were college graduates. Compared

to population norms, this sample was slightly more educated than what would be expected

from the basic age demographic. Each mailing consisted of the study survey, an unrelated

questionnaire, and a $6 check for participation.

INSERT TABLE 7 ABOUT HERE

In the first part of the questionnaire, respondents looked at the shelf labels of ten

brands grouped into five product categories. For each brand, a one-sentence product

description (e.g., “Planters Mixed Nuts, 10 oz, less than 50% peanuts”) was provided

along with its price and, when necessary, a shelf tag with the description of the promotions

presented in Table 7. The three utilitarian product categories were liquid laundry

detergent, AA batteries, and flour. The three hedonic product categories were assorted

chocolates, mixed nuts, and bubble bath. A pretest showed that Cheer, Duracell,

Pillsbury, Freeman Beautiful Bath, Whitman Assorted Chocolates and Planters were the

higher equity brands and Purex, Eveready, Robin Hood, Capri French Formula, Russell

Stover and Nutcracker were the lower equity brands. The prices used in the experiment

were current prices collected at a local supermarket. They ranged from $2.39 to $3.99,

equitably balanced between utilitarian and hedonic brands, with an average of $3.46.

Results of Study 5

Each utilitarian product scored higher on an utilitarian index inspired by Batra and

Ahtola (1990) and computed by subtracting the semantic differential score on (fun/not

fun) from the average SD score on (wise/foolish) and (useful/useless). The utilitarian

score was -0.01 for chocolate, 0.01 for nuts, 1.02 for flour and 1.73 for detergent

(F1,290=65.1, p<0.01, η=0.43 for a comparison of utilitarian and hedonic groups). With

24

two exceptions, all high-equity brands were more frequently purchased than any of the

low-equity brands (t=18.3; p<0.01), and preferred to low equity brands (F1,293=7.5,

p<0.01, η=0.16). The two exceptions are the two brands of bubble bath, which had

similarly low usage rates and brand evaluation, and the two brands of batteries, which had

similarly high usage rates and brand evaluations. Since this study examines the effects of

sales promotions for consumer choices between a high equity and a low equity brand, we

eliminated these two products from the subsequent analyses. Further manipulation checks

show that brand equity had no impact on the utilitarian index (F1,290=2.35, p=0.13, η

=0.09). Finally, monetary promotions were perceived as more utilitarian and less hedonic

than non-monetary promotions (see Table 7, group difference: F1,320=137.2, p<0.01, η

=0.55) and, as in study 5, were preferred to non-monetary promotions (F1,182=19.3,

p<0.01, η=0.31).

The effects of benefit congruency for high and low equity brands. The data were

analyzed following the same procedure as in study 4 with the choice of the target brand as

the dependent binary variable, and adding past usage, age and gender as individual

covariates. The independent variables included an indicator variable for product type, two

indicator variables for promotion type, one indicator variable for brand equity, and

indicator variables for all interactions7. The results of the logistic regression indicated a

significant three-way interaction between promotion type, product type and brand equity

(B=-1.5, Wald=4.8, p<0.03). This provides evidence that the effects of benefit

congruence vary depending on the equity of the target brand.

INSERT TABLE 8 AND TABLE 9 ABOUT HERE

7 The logistic regression (-2LL=702.4, χ²=105.4, df=13, p<0.01) show choices were not different betweenhedonic and utilitarian products (B=0.05, Wald=0.03, p=0.88), nor were they different for monetary vs. nopromotion (B=0.18, Wald=0.21, p=0.64), nor for non-monetary vs. no promotion (B= -0.07, Wald=0.03,p=0.86). Brand equity has a negative effect on brand choice because of the lower prices of low equitybrands (B=-2.14, Wald=41.3, p<0.01). Past usage is the only significant covariate (B=1.04, Wald=29.2,p<0.01) and no two-way interactions was significant.

25

Consider the differences between high equity brands with low equity brands (see

Table 8). As predicted by H3a and H3b, the market share of low equity brands was not

influenced by sales promotions, and monetary promotions had the same effect than non-

monetary promotions (the interaction between promotion type and product type is not

significant Wald=1.94, df=2, p=0.38). For high equity brands in contrast, monetary

promotions increased choice shares compared to no promotion (even though non-

monetary promotions did not), and the interaction between promotion type and product

type is significant (Wald=5.8, df=2, p<0.05). As Tables 9 shows, this result is essentially

due to the poor performance of monetary promotions for hedonic products. As expected,

monetary promotions were more effective for utilitarian products (+ 17 points) than for

hedonic products (+1 point). Unexpectedly however, non-monetary promotions are only

marginally more effective for hedonic products (+13 points) than for utilitarian products

(+11 points).

Discussion of Experimental Studies

Overall, the two choice experiments show that it is critical that managers take into

account the types of consumer benefits provided by their promotions, because it helps

them to predict how effective a particular promotion will be for a particular product.

Specifically, studies 4 and 5 support the benefit congruency hypothesis by showing that

sales promotions are more effective when they provide benefits that are congruent with

one’s need for the product. The robustness of the findings is demonstrated across

countries, across promotions, across ages, and across products. It is important to

understand, however, that benefit congruency holds only for familiar and high quality

brands. Buyers of low equity brands are less interested in obtaining the full extent of the

utilitarian or hedonic benefits possible in each product category, and are therefore less

sensitive to the congruency of the benefits provided by the promotion. This points to the

26

fact that the utilitarian or hedonic nature of a product is not the only antecedent of benefit

importance, and therefore of the effectiveness of a promotion. It is critical that marketers

not only understand the generic benefits of the product category but directly study the

type of benefits sought by the buyers for the targeted brand, purchase situation and

geographical market.

Our results also show that free gifts and free product offers may represent viable

alternatives to commonly used coupons and sweepstakes. Coupons, in particular, had a

negative effect on the market share of high equity hedonic brands and their performance

was the least consistent across high equity brands. In contrast, the effectiveness of free

product offers was more stable across product categories, probably because offering more

of the same product provides utilitarian benefits for utilitarian products, and hedonic

products for hedonic products. Our findings about the positive effects of free gifts

suggests that the negative effects documented in Simonson et al.’s (1994) study may be

due to the relevance of the gift. These authors used what they described as “unneeded”

free gifts targeted at a different segment that their respondents (collector’s plate, golf

umbrella and magazine subscription), whereas the studies reported here used gifts that

appealed directly to the buyers (free red rose and free movie rental). The effect in their

study would have been further weakened by the presence of multiple utilitarian products

(35mm film, CD player, wristwatch, calculator, VCR and dental plan) and only one

hedonic product (brownie mix). The findings reported here underscore that there may be

value to revisiting some of these studies and reinterpreting their results in light of the type

of promotions that were used.

Summary and DiscussionPerhaps because coupons and temporary price reductions are the most typical form

of sales promotions, most research on sales promotions has assumed that monetary

27

savings are the primary consumer benefit of sales promotions. As a result, many studies

have examined the costs of promotion usage or the effects of sales promotions on

purchase behavior, and comparatively fewer studies have examined their benefits to the

consumer. The purpose of this research was to provide an integrative framework of the

utilitarian and hedonic consumer benefits of sales promotions and to apply it to study the

effectiveness of monetary and non-monetary promotions across different product types.

These are the key conclusions:

1. Consumers do not evaluate sales promotions solely in terms of monetary savings. Theyconsider an array of hedonic and utilitarian benefits including the ability of theseproducts to improve product quality, shopping convenience, and to provideopportunities for value-expression, entertainment, and exploration.

2. In our study, non-monetary promotions provided more hedonic benefits and lessutilitarian benefits than monetary promotions.

3. Consistent with the benefit congruency hypothesis, non-monetary promotions weremore effective for hedonic products than for utilitarian products, and monetarypromotions were more effective for utilitarian products than for hedonic products.

4. The effects of benefit congruency were especially caused by the decreasedeffectiveness that monetary promotions (especially coupons) have for hedonic productswith high customer-based brand equity.

Limitations and Implications for ResearchersThese findings have implications for sales promotion theory as well as for

manufacturers' and retailers' promotional strategies. Prior research showed that consumer

response to coupons could be explained by two personal traits: coupon proneness and

value consciousness (Lichtenstein et al. 1990). The “promotion-based” multi-benefit

framework presents an alternative to this “person-based” perspective by specifying the

utilitarian or hedonic benefits that may motivate each type of "deal proneness.” For

instance, the framework suggests that one reason why “sale-prone” consumers buy more

products on end-of-aisle displays than “rebate-prone” consumers (Lichtenstein et al. 1995)

may be because they more heavily weigh the convenience benefit offered by sales

28

promotions. Similarly, this study cautions against referring to sales promotions in a

generic manner, and suggests to use a new typology of sales promotions based on the

benefits they deliver instead of on the technique (such as coupons or rebates) that delivers

them.

Examining this heterogeneity in sales promotion benefits opens the opportunity to

study how different segments of consumers differentially respond to each benefit. It is

probable that experts and novices, buyers and end-users, heavy and light users, would all

differ in their evaluations of the benefits of sales promotions. Accounting for these

different segments is critical in insightfully analyzing what might otherwise appear to be

inconclusive data (Wansink and Ray 1992). Segmenting consumers by the type of benefits

sought may prove valuable for direct marketing. For instance, most direct marketing

programs segment consumers by brand usage, loyalty or demographic variables.

However, demographic variables are poor predictor of promotion sensitivity (Hoch, Drèze

and Purk 1994), and it is unclear whether heavy users are more sensitive to monetary

savings or to symbolic recognition. A benefit-based segmentation would help target

promotions only to those segments that are most responsive to them.

Future research could also study the effects of benefit congruency beyond the

utilitarian or hedonic nature of the product. It would be interesting to study benefit

importance in the different phases of the product lifecycle, the purchase situation (e.g.,

agent/end user), the consumption context (e.g., gift giving) and general personality traits

(e.g., self-monitoring). For instance, we might expect that savings are more important

than value expression for mature products, for agents and for low self-monitoring

consumers. Finally, there are many opportunities to explore other consumer benefits of

sales promotions, or to refine the hedonic/utilitarian classification by drawing on prior

29

research on customer value or of product attributes and benefits. Prior examples can be

seen in Holbrook’s (1994) self-oriented/other-oriented and active/reactive dimensions, and

Keller’s (1993) distinction between utilitarian, hedonic and symbolic benefits.

Implications for Managers

The existence of multiple consumer benefits of sales promotions has implications

for the current concerns about the declining effectiveness of price promotions, their long-

term effects on price sensitivity, and the EDLP debate. Our findings suggest that value-

added promotions can provide an efficient alternative to price promotions by combining

utilitarian and hedonic benefits. That is, marketers could combine an in-pack coupon with

an on-pack contest, or they could combine a multi-pack refund with an in-store display

that emphasizes new product uses. From a long-term equity standpoint, promotions

offering more than savings are less likely to increase price sensitivity and more likely to

create unique brand associations that can reinforce brand image than monetary

promotions. We gathered preliminary evidence on this issue by asking a subsample of

respondents in Study 5 to rate brand image on the five personality dimensions suggested

by Aaker (1997). We found that, compared to monetary promotion and to the control

condition, non-monetary promotions resulted in higher prestige for the promoted brand.

While only tentative, these results suggest that some sales promotions should be seen as

brand-building activity rather than as a short-term sales incentive only.

Without minimizing the importance of supply-side arguments, these findings also

suggest that the EDLP debate should take into consideration the key demand-side issue—

the consumer. Many studies recommending EDLP characterize consumers on a

convenience-to-price continuum, assuming that deal-prone consumers are willing to

30

forego convenience for lower prices (see Kahn and McAlister 1997; Lal and Rao 1997;

Narasimhan 1984). Under these assumptions may not hold for all consumers. Indeed, our

results show that consumers actually find that sales promotions can provide savings and

improve shopping convenience by reducing search and decision costs. Similarly, EDLP

policies also run the risk of alienating hedonic deal-prone shoppers. More generally, the

multi-benefit framework suggests that sales promotions may be appropriate under

conditions that would not call for promotions if one follows the purely monetary

framework.

ConclusionPromotions provide both hedonic and utilitarian benefits to consumers.

Regrettably, savings are often assumed to be the sole consumer benefit of these

promotions. When managers overlook the importance of these utilitarian and hedonic

benefits, they may be reducing the effectiveness of their promotions and draw inaccurate

conclusions on their usefulness vis-à-vis an everyday-low-price strategy. When

researchers overlook these benefits, they risk overlooking the intrinsic functions

promotions have for consumers or mis-specifying a model of consumer response to sales

promotions. The benefit-congruency framework presented here opens the opportunity to

study how different segments of consumers and different retail situations influence

promotional effectiveness. By studying the reasons behind why consumers respond to

sales promotions, we better understand how and when promotions work, both in theory

and in the field. In this way we increase the effectiveness of sales promotions as they

increase their presence in the marketing mix.

31

References

Aaker, Jennifer L. (1997), “Dimensions of Brand Personality,” Journal of Marketing Research, 34(August), 347-356.

Abraham, Magid M. and Leonard M. Lodish (1990), “Getting the Most out of Advertising andPromotion,” Harvard Business Review, (May-June), 50-63.

Arbuckle, James L. (1997), Amos Users' Guide Version 3.6. Small Waters Corporation.

Ailawadi, Kusum, Karen Gedink, and Scott Neslin (1998), "The Cross-Elasticities of Price PromotedProducts and Store Brands," Working Paper, Dartmouth College, Hanover, NH.

Babin, Barry J., William R. Darden and Mitch Griffin (1994), “Work and/or Fun: MeasuringHedonic and Utilitarian Shopping Value,” Journal of Consumer Research, 20 (March), 644-656.

Bagozzi, Richard P., Hans Baumgartner and Youjae Yi (1992), “State versus Action Orientation andthe Theory of Reasoned Action: An Application to Coupon Usage,” Journal of ConsumerResearch, 18 (March), 505-518.

Batra, Rajeev and Olli T. Ahtola (1990), “Measuring the Hedonic and Utilitarian Sources ofConsumer Attitudes,” Marketing Letters, 2 (2), 159-170.

Baumgartner, Hans and Jan-Benedict E.M. Steenkamp (1996), “Exploratory Consumer BuyingBehavior: Conceptualization and Measurement,” International Journal of Research in Marketing,13 121-137.

Bawa, Kapil and Robert W. Shoemaker (1987), “The Coupon Prone-Consumer: Some FindingsBased on Purchase Behavior Across Product Classes,” Journal of Marketing, 51 (4), 99-110.

__________ and __________ (1989), “Analysing Incremental Sales From a Direct Mail CouponPromotion,” Journal of Marketing, 53 (3), 66-76.

BIPP, Banque Information Progress Promotion (1995), Statistiques Generales de la Promotion desVentes. Paris.

Blattberg, Robert C., Thomas Buesing, Peter Peacock and Subrata K. Sen (1978), “Identifying theDeal Prone Segment,” Journal of Marketing Research, 15 (August), 369-377.

__________ and Scott A. Neslin (1990), Sales Promotion: Concepts, Methods, and Strategies.Englewood Cliffs, New Jersey: Prentice Hall.

__________ and ___________ (1993). Sales Promotion Models in Marketing, J. Eliashberg and G.L. Lilien (eds.), North Holland, 553-610.

__________ and Kenneth J. Wisniewski (1989), “Price-Induced Patterns of Competition,”Marketing Science, 8 (4), 81-100.

Bollen, Kenneth A. (1989), Structural Equations with Latent Variables. New York: John Wiley &Sons.

Bronnenberg, Bart J. and Luc Wathieu (1997), “Asymmetric Promotion Effects and BrandPositioning,” Marketing Science, Vol. 15 (4), 379-394.

Browne, Michael W. and Robert Cudeck (1993), “Alternative Ways of Assessing Model Fit,” inTesting Structural Equation Models, Bollen Kenneth A. and S. Scott Long, eds., Newbury Park:Sage.

32

Buzzel, Robert, John Quelch and Walter Salmon (1990), “The Costly Bargain of Trade Promotion,”Harvard Business Review, March-April 141-149.

Churchill, G. A. Jr. (1979), “A Paradigm for Developing Better Measures of Marketing Constructs,”Journal of Marketing Research, 16 (February), 64-73.

Cox Direct (1997), 19th Annual Survey of Promotional Practices, Largo, FL.

Darke, Peter R. and Jonathan L. Freedman (1995), “Nonfinancial Motives and Bargain Hunting,”Journal of Applied Social Psychology, 25 (18), 1597-1610.

Dhar, Sanjay K. and Stephen J. Hoch (1996), “Price Discrimination Using In-Store Merchandising,”Journal of Marketing, 60 (1), 17-30.

________, and ________ (1997), "Why Store Brand Penetration Varies by Retailers", MarketingScience, 16 (23), 208-227.

Dickson, Peter R. and Alan G. Sawyer (1990), “The Price Knowledge and Search of SupermarketShoppers,” Journal of Marketing, 54 (3), 42-53.

Eagly, Alice H. and Shelly Chaiken (1993), The Psychology of Attitudes. Orlando: Harcourt Brace &Company.

Feick, Lawrence and Linda Price (1987), “The Market Maven: a Diffuser of MarketplaceInformation,” Journal of Marketing, 51 (1), 83-97.

Fishbein, Martin and Icek Ajzen (1975), Beliefs, Attitude, Intention, and Behavior: An Introductionto Theory and Research. Reading, Massachusetts: Addison-Wesley Publishing Co., Inc.

Fornell, Claes and David F. Larcker (1981), “Evaluating Structural Equation Models withUnobservable Variables and Measurement Error,” Journal of Marketing Research, 18(February), 39-50.

Gerbing, David W. and James C. Anderson (1988), “An Updated Paradigm for Scale DevelopmentIncorporating Unidimensionality and Its Assessment,” Journal of Marketing Research, 25 (May),186-192.

Hirschman, Elizabeth C. and Morris E. Holbrook (1982), “Hedonic Consumption: EmergingConcepts, Methods and Propositions,” Journal of Marketing, 46 (3), 92-101.

Hoch, Stephen J., Xavier Drèze and Mary Purk (1994), “EDLP, Hi-Lo, and Margin Arithmetic,”Journal of Marketing, 58 (4), 16-27.

Holbrook, Morris B. (1994), "The Nature of Customer Value". In Service Quality: New Directionsin Theory and Practice, R. T. Rust and R. L. Oliver (eds.). Thousand Oaks, CA, SagePublications.

Hoyer, Wayne D. (1984), “An Examination of Consumer Decision Making for a Common RepeatPurchase Product,” Journal of Consumer Research, 11 (December), 822-829.

Inman, J. Jeffrey, Leigh McAlister and Wayne D. Hoyer (1990), “Promotion Signal: Proxy for aPrice Cut?” Journal of Consumer Research, 17 (June), 74-81.

Jöreskog, Karl G. and Dag Sörbom (1989), LISREL 7 User's Reference Guide. Chicago, IL:Scientific Software International, Inc.

Kahn, Barbara E. (1995), “Consumer Variety-Seeking Among Goods and Services: An IntegrativeReview,” Journal of Retailing and Consumer Services, 2 (3), 139-148.

33

__________ and Therese A. Louie (1990), “Effects of Retraction of Price Promotions on BrandChoice Behavior for variety Seeking and Last Purchase Loyal Consumers,” Journal of MarketingResearch, 27 (3), 279-289.

Kahn, Barbara E. and Leigh McAlister (1997), Grocery Revolution, The New Focus on theConsumer. Addison-Wesley.

Katz, Daniel (1960), “The Functional Approach to the Study of Attitudes,” Public OpinionQuarterly, 24 163-204.

Keller, Kevin Lane (1993), “Conceptualizing, Measuring, and Managing Customer-based BrandEquity,” Journal of Marketing, 57 (1), 1-22.

Kelly, George Alexander (1955), The psychology of personal constructs. New York: Norton.

Laurent, Gilles and Jean-Noël Kapferer (1985), “Measuring Consumer Involvement Profiles,”Journal of Marketing Research, 22 (February), 41-53.

Lal, Rajiv and Ram Rao (1997), “Supermarket Competition: The Case of Every Day Low Pricing,”Marketing Science, 16 (1), 60-80.

Lichtenstein, Donald R., Richard G. Netemeyer and Scot Burton (1990), “Distinguishing CouponProneness From Value Consciousness: An Acquisition-Transaction Utility Theory Perspective,”Journal of Marketing, 54 (3), 54-67.

_____________, _____________ and _____________ (1995), “Assessing the Domain Specificityof Deal Proneness: A Field Study,” Journal of Consumer Research, 22 (December), 314-326.

_____________, _____________ and _____________ (1997), “Psychological Correlates of aProneness to Deals; A Domain-Specific Analysis,” in Advances in Consumer Research, Vol.XXIV, Merrie Brucks and Deborah J. MacInnis, eds.

___________, Nancy M. Ridgway and Richard G. Netemeyer (1993), “Price Perceptions andConsumer Shopping Behavior: A Field Study,” Journal of Marketing Research, 30 (May), 234-245.

Mano, Haim and Richard L. Oliver (1993) "Assessing the Dimensionality and Structure of theConsumption Experience: Evaluation, Feeling, and Satisfaction", Journal of Consumer Research,20 (December) 451-466.

Mela, Carl F., Sunil Gupta and Donald R. Lehman (1997), “The Long-Term Impact of Promotionand Advertising on Consumer Brand Choice,” Journal of Marketing Research, 34 (May), 248-261.

Meyer, Robert J., and Barbara E. Kahn (1991), “Probabilistic Models of Consumer ChoiceBehavior,” In Handbook of Consumer Behavior, Thomas S. Robertson and Harold K. Kassarjian,eds., Englewood Cliffs, N.J: Prentice Hall.

Mittal, Banwari (1994), “An Integrated Framework for Relating Diverse Consumer Characteristicsto Supermarket Coupon Redemption,” Journal of Marketing Research, 31 (November), 533-544.

Murray, Henri A. (1938), Explorations in Personality. New York: Oxford University Press.

Narasimhan, Chakravarthi (1984), “A Price Discrimination Theory of Coupons,” MarketingScience, 3 (2), 128-146.

Papatla, Purushottam and Lakshman Krishnamurthi (1996), “Measuring the Dynamic Effects ofPromotions on Brand Choice,” Journal of Marketing Research, 33 (February), 20-35.

34

Promotion Marketing Association of America, Inc (1994), Winning with Promotion Power: TheReggie Awards Winners. Ravenswood, IL: Dartnell Corporation.

Ratchford, Brian T. (1987), “New Insights about the FCB Grid,” Journal of Advertising Research,27 (August/September), 24-38.

Rao, Ram C. (1991), "Pricing and Promotions in Asymmetric Duopolies," Marketing Science, 10(2),131-144.

Schindler, Robert M. (1992), “A Coupon is More than a Low Price: Evidence from a Shopping-Simulation Study,” Psychology & Marketing, 9 (6), 431-451.

Shavitt, Sharon (1990), “The Role of Attitude Objects in Attitude Functions,” Journal ofExperimental Social Psychology, 26, 124-148.

Shimp, Terence A. and Alican Kavas (1984), “The Theory of Reasoned Action Applied to CouponUsage,” Journal of Consumer Research, 11 (December), 795-809.

Simonson, Itamar, Ziv Carmon and Suzanne O'Curry (1994), “Experimental Evidence on theNegative Effect of Product Features and Sales Promotions on Brand Choice,” Marketing Science,13 (1), 23-39.

Soman, Dilip (1998) “The Illusion of Delayed Incentives: Evaluating Future Effort-MoneyTransactions,” Journal of Marketing Research 25 (November): Forthcoming.

Strahilevitz, Michal and John G. Myers (1998), “Donations to Charity as Purchase Incentives: HowWell They Work May Depend on What You Are Trying to Sell,” Journal of Consumer Research,24 (March), 434-446.

Thaler, Richard (1985), “Mental Accounting and Consumer Choice,” Marketing Science, 4 (3), 199-214.

Tourtoulou, Anne-Sophie (1996), “Marques nationales, marques de distributeurs et premiers prix:effets de leurs mises en avant sur les ventes et la structure concurrentielle de la catégorie deproduits, ” Unpublished PhD. dissertation, Groupe HEC, Jouy en Josas, France.

Tversky, Amos, Shmuel Sattath and Paul Slovic (1988), “Contingent Weighting in Judgement andChoice,” Psychological Review, 95 (July), 204-217.

Wansink, Brian and Michael L. Ray (1992), “Estimating an Ad's Impact on One's Consumption of aBrand,” Journal of Advertising Research, 60 (1), 3-14.

Webster, Frederick E., Jr. (1965), “The 'Deal-Prone' Consumer,” Journal of Marketing Research, 2(2), 186-189.

Zeithaml, Valarie A. (1988), “Consumer Perceptions of Price, Quality and Value: a Means-EndModel and Synthesis of Evidence,” Journal of Marketing 52 (July), 2-22.

35

TABLE 1

Utilitarian Benefits of Sales Promotions

Benefit Existing Support Excerpts from Interviews Measuresa

(Disagree/Agree)

Savings

(Monetary

savings)

Sales promotions can provide perceptions of monetary savings by lowering the

unit price of the promoted product, by offering more of the same product for

free or by providing refunds or rebates on subsequent purchases of the same or

other products. Both the amount of the price reduction and the percent

deviation from a reference price can create perceptions of monetary savings

(Blattberg and Neslin 1990).

A promotion is a price cut or

a larger package size for the

same price.

A promotion is like new

money in your pocket that

you can use to buy

something else.

1. I really save money

2. I feel that I am getting a good

deal

3. I really spend less

Quality

(Increase in

the quality of

the product

bought)

By reducing the price of the product, or by offering a smaller package size,

sales promotions can relax budget constraints and enable consumers to

upgrade to a better product (Blattberg and Wisniewski 1989). Like the savings

benefit, the quality benefit boils down to increasing the value for the money

but, unlike the former, it usually involves spending more money. It can

therefore be linked to the “excellence” type of customer value discussed by

Holbrook (1994).

I normally don’t buy

packaged salads because

they are too expensive. But

I buy them when they cut

the price.

1. I can have a higher-quality

product at the same price

2. I can afford a better-than-usual

product

3. I can upgrade to a better brand

Convenience

(Reduction in

search and

decision costs)

Sales promotions can improve shopping efficiency by reducing search costs by

helping consumers find a product or by reminding them of a product that they

needed to buy. This “advertising” effect is supported by field experiments

(Inman, McAlister, and Hoyer. 1990; Bawa and Shoemaker 1989; Kahn and

McAlister 1997) and in-store surveys (Dickson and Sawyer 1990). Sales

promotions can also improve shopping efficiency by reducing decision costs by

providing consumers with an easy decision heuristic or rationale for buying, or

not buying, the promoted product (Hoyer 1984; Simonson, Carmon, and

O’Curry 1994). See Holbrook (1994) and Zeithaml (1988) for related

definitions.

Sometimes, I remember that

I needed a product when I

see it on sale

I buy the brand on deal

because I don’t know which

to buy.

I like promotional packs

because they are fast and

easy.

1. These promotions remind me

that I need the product

2. These promotions make my life

easy

3. I can remember what I need

a All measures beginning with: “(With) this type of promotion” and translated from French.

36

TABLE 2

Hedonic Benefits of Sales Promotions

Benefit Existing Support Excerpts from Interviews Measures

(Disagree/Agree)

Value-expressiona

(Expression and

enhancement of

self-concept and

personal values)

Some consumers respond to sales promotions to meet personal or moral

values such as being a “good shopper” (Mittal 1994). The value-expression

benefit can thus be linked to the “morality” value defined by Holbrook (1994).

This hedonic type of customer value describes the gratification earned from

fulfilling one’s duty. Other consumers respond to a promotion to express and

enhance their sense of themselves as smart shopper, and earn social

recognition or affiliation (Bagozzi, Baumgartner and Yi 1992; Feick and

Price 1987; Shimp and Kavas 1984; Schindler 1992). This dimension of the

value-expression benefit can be linked to Holbrook’s (1994) utilitarian

“politics” and “esteem” values since it describes how consumers respond to

sales promotions to earn status and control over others.

When my husband comes

back from his shopping trip,

he is always very proud to tell

me about the bargains he

found.

I sometimes feel guilty when

I could have used a coupon

but didn’t.

1. I feel good about myself

2. I can be proud of my

purchase

3. I feel like I am a smart

shopper

Exploration

(Stimulation and

variety)

Because sales promotions are constantly changing, and because theyattract consumer’s attention, promotions can fulfill intrinsic needs forexploration, variety and information (Baumgartner and Steenkamp1996; Kahn 1995; Kahn and Louie 1990). The exploration benefit hasbeen documented in the context of shopping (Babin et al. 1994), varietyseeking (Kahn 1995) and exploratory behavior (Baumgartner andSteenkamp 1996).

My husband likes to look atthe promotion in the paperseven though he never doesthe shopping!

When I buy, I look at salespromotions to get new ideasand to find variety.

1. I feel like trying new brands

2. I can avoid buying always

the same brands

3. I can get new ideas of things

to buy

Entertainment

(Amusement and

aesthetic value)

Many sales promotions such as sweepstakes, contests, and free gifts are

intrinsically fun to watch and to participate in. We therefore encompass

within this notion both the active “play” and reactive “aesthetic” values of

Holbrook’s (1994) typology. However, the entertainment benefit is distinct

from the overall positive affect, or enjoyment, resulting from buying a

promoted product, which is commonly used to measure “deal-proneness”

I read the contests on thecereal boxes every morning;they are fun.

Sweepstakes in the storecreate a nice and excitingatmosphere.

1. These promotions are fun

2. These promotions are

entertaining

3. These promotions are

enjoyable

37

(e.g. Lichtenstein, Netemeyer, and Burton. 1990).

a As discussed in the text, the value expression benefit has both a utilitarian and hedonic component.

38

TABLE 3

Study 3: How Utilitarian and Hedonic Benefits Influence Promotion Evaluations

Monetary promotions evaluation Non-monetary promotions evaluation

Benefit Mean scorec B CR Mean score B CR

Savings 3.57 .55 5.25a 2.09a .30 1.63

Quality 2.90 -.11 -1.76 2.17a .36 1.16

Convenience 2.25 .27 2.60a 2.17 -.84 -1.82

Value-expression 3.29 .26 2.18a 2.87a .35a 2.08a

Entertainment 2.09 .11 1.47 2.94a .43a 2.12a

Exploration 2.96 -.07 -.98 3.16b .79a 3.51a

a Critical Ratio of regression coefficient, or difference between monetary and non-monetary means, isstatistically significant at p<0.01 b Statistically significant at p<0.05.

c Mean scores of monetary and non-monetary promotions and regression coefficients estimatedsimultaneously in the structural equation model using Amos 3.6 (Bollen 1989).

TABLE 4

Promotion Stimuli used in Study 4

Type Technique Promotiona Pair Utilitarian scoreb

US FR

Monetary Price

discounts

Weekly Special! 75 cents off the regularprice!

A ___ c 0.29

Instant Coupon! Save 75c now! Redeem atcheckout

B 3.62 -0.15

Mail-in

Rebate

Special Buy! Your next purchase free! (Refund

by mail)

C 2.42 -0.12

Save $1! Refund by mail. D 2.24 -0.23

Non-Monetary Free gift Receive the Official Watch of the 2000 Olympic

Games in Sydney! Send a proof of purchase +

$7.95 for S&H

A -1.74 -1.33

Receive a 22” Duffel Bag! Send bar code + $9.95

for S&H.

C 0.74 -0.40

Sweepstakes

& contest

Scratch and win instantly! 50 winners will go to

Club Med. in Cancun!

B -3.76 -2.12

Finish this puzzle and win 2 movie tickets!Send completed crossword puzzle.

D -1.82 -0.86

a These descriptions were translated into French and the monetary amounts were converted intoFrancs.b Ranges between –8 to 8. Computed as the difference between the Likert score on “____ is useful”and “____ is fun” for the US study, and “____ est utile” and “____ est sympa” for the French study.Measures taken from Batra and Ahtola (1990). c No data collected for this promotion.

39

TABLE 5

Results of Study 4 for US and French Respondentsc

Choice Purchase Intentions

Respondents US French US French

Factor B Wald B Wald F1,204 F1,556

Product type -.18 1.3 .04 0.3 .75 .08

Promotion type .83 27.7a .33 14.7a 44.8a 15.0a

Interaction .32 4.0a .22 6.2a 2.8b 2.9b

a p<0.05, b p<.10.

c The coding is the following. For choice, 1=chosen and 0=not chosen. For purchase intentions,

1=will not buy and 9=will buy. For product type, 1=utilitarian and –1=hedonic. For promotion type,

1=monetary and –1=non monetary.

TABLE 6

Study 4: Effectiveness of Non-Monetary Promotions for Utilitarian and HedonicProducts

Choice shares of brand with non-

monetary promotiona (%)

Intent to purchase the brand with

non-monetary promotioncUtilitarian indexd

Studyb US French US French US French

All products 31 42 3.83 4.54 1.55 .67

Hedonic products 37 48 4.10 4.75 -1.60 -2.56

Wine 42 51 4.16 4.92 -.70 -1.59

Ice cream 33 45 4.04 4.57 -2.50 -3.54

Utilitarian products 25 36 3.57 4.33 4.71 3.91

Film 25 37 3.64 4.27 2.76 2.15

Garbage bags 25 33 3.50 4.40 6.66 5.66a The choice share of the brand offered with a monetary promotion is the complement to 100%b The number of respondents associated with the figures is 139 for the French study and 51 for the US

study.c 9= will definitely buy, 1=will definitely not buy. The purchase intention of the brand offered with a

monetary promotion is the complement to 10.d Ranges between –8 to 8. Computed as the difference between the score on “____ is useful” and“____ is fun” for the US study and “____ est utile” and “____ est sympa” for the French study.

40

TABLE 7

Promotion Stimuli used in Study 5

Type Technique Promotion Utilitarian scorea

Monetary Price offer Smart Saver! Save 35¢ with this coupon Redeem at checkout. 1.11

Weekly Special! Take an additional 10% off the markedprice.

1.64

Free product Value pack! 15% more product free. 1.41

Special Offer! Buy one, get another at half price! 0.31

Non-

Monetary

Free gift Buy this product and get one red rose free! Simply select a

rose from the flower department and show this pack with the

offer to the cashier.

-1.75

Free videotape rental! Go to the nearest Blockbuster Video

and show one proof of purchase

-1.10

Sweepstakes Want to have some fun? Fill this crossword puzzle and win a

week in Hawaii and many other prizes.

-2.31

Receive 2 Free tickets to go see the Major League Baseball

Team of your choice! Just look inside the pack to see if you

are one of the 50 winners

-1.42

a Ranges between –8 to 8. Computed by subtracting the semantic differential score on (fun/Not fun)

from the average SD score on (wise/foolish) and (useful/useless).

TABLE 8

Results of Study 5 for High and Low Equity Brands

High equity brand Low equity brand

B Wald df B Wald df

Product type .19 0.20 1 -.10 .05 1

Promotion type (total effect) 4.05 2 .58 2

Monetary .84 4.05a 1 .17 .15 1

Non monetary .49 1.34 1 -.15 .11 1

Product type x promotion type (total effect) 5.76a 2 1.94 2

Monetary prom. x hedonic product -1.23 3.91a 1 .48 .49 1

Non-monetary prom. x hedonic product .15 0.06 1 .94 1.93 1

Past usage -.84 11.92a 1 1.37 17.29a 1

Sex -.15 0.22 1 -.48 1.95 1

Age .01 1.37 1 -.03 6.45a 1a p<.05. Coding as follows: Product type (0 if utilitarian, 1 if hedonic); Monetary promotion: (1 if the

target brand offers a monetary promotion and zero otherwise); Non-monetary promotion (1 if target

brand offers a non-monetary promotion and zero otherwise).

41

TABLE 9

Study 5: Effects of Monetary and Non Monetary Promotions on the ChoiceShares of High Equity and Low Equity Utilitarian and Hedonic Brandsa

High equity brands Low equity brands

Product Type Utilitarian Hedonic Utilitarian Hedonic

No promotion 30 28 70 72

Monetary promotions 47 29 70 78

Coupons 44 23 75 80

Free product offers 50 37 67 76

Non monetary promotions 41 41 67 69

Free gifts 52 52 63 71

Sweepstakes 26 28 74 65

a The number of observations associated with each choice share varies between 11 and 19 for

individual brands by promotion cells and between 24 and 36 for each product type by promotion cell.

SAVINGS

s3

zs3

s2

zs2

s1

zs1

.84.80.88

QUALITY

q3

zq3

q2

zq2

q1

zq1

.61.78.66

CONVENIENCE

c3

zc3

c2

zc2

c1

zc1

.56.66.53

VALUEEXPRESSION

ve3

zve3

ve2

zve2

ve1

zve1

.76.69.76

ENTERTAINMENT

e3

ze3

e2

ze2

e1

ze1

.65.86.85

EXPLORATION

x3

zx3

x2

zx2

x1

zx1

.72.86.80

zve ze zx

UTILITARIAN HEDONIC

.67

.83 .74 .78 .46 .45 .75

zs zq zc

.41

42

FIGURE 1Validating the Multi-Benefit Framework of Sales Promotions

(Second-order Confirmatory Factor Analysis)

43

FIGURE 2

Perceived Hedonic and Utilitarian Benefits of Selected Promotion Techniques

FIGURE 3

Low HedonicBenefits

High HedonicBenefits

HighUtilitarianBenefits

LowUtilitarianBenefits

Sweepstakes

Free gifts

Free productoffers

Rebates

Coupons

Pricereductions

44

Results of Study 4 for US and French Respondents

US Respondents French Respondents

37%25%

48%36%

63%75%

52%64%

0%

25%

50%

75%

100%

Hedonic products Utilitarian products Hedonic products Utilitarian products

Non-monetary promotions Monetary promotions