Free Gift with Purchase: Promoting FREE GIFTS WITH PURCHASE: PROMOTING OR DISCOUNTING THE BRAND? RAGHUBIR or Discounting the Brand?

Priya Raghubir
Haas School of Business
University of California, Berkeley
Two experiments examine the process by which free gift promotions serve as a source of information about the underlying value of the product offered as a free gift. The value-discounting
hypothesis argues that by virtue of being offered as a free gift, products will be valued less as
evinced by lower purchase intentions and a lower price that consumers are willing to pay for
them. Conditions that inhibit the value-discounting effect include the (a) presence of alternate
price information to make judgments about the value of the gift, and (b) contextual information
about the value of the promoted brand.
A rich literature in sales promotions has shown that
short-term sales are positively affected by offering promotions (for a review, see Blattberg & Neslin, 1990). However,
the economic model of consumer promotions does not explicitly account for complementary routes through which
promotions can affect consumer behavior. The notion that
consumer promotions are informative and affect sales
through more than offering a monetary incentive to purchase
is not new (Inman, McAlister, & Hoyer, 1990; Inman, Peter,
& Raghubir, 1997; Raghubir, 1998), but it has never been applied to the context of free gift promotions. This article explores the information content of free gift offers, focusing on
inferences drawn about the free gift, rather than the promoted
product.
“Free gift with purchase” offers appear to be inundating
the marketplace. Whether they are in the cosmetics industry,
in duty-free catalogs, or in your everyday supermarket
shelves, marketers appear to be enticing consumers to buy
their product through the offer of a free gift with purchase.
The offers vary: Some explicitly mention the value of the offer, whereas others do not. Although research in price promotions is useful to understand how such offers affect current
and future sales of the promoted brand offering the free gift,
an unanswered question is how they affect future sales of the
brand being offered as a free gift.
In this article, we examine the manner in which consumers’ process free gift with purchase offers. Although the cosmetics industry is by far the largest user of direct gift with
purchase offers, with over 60% of department store makeup
and treatment sales and 40% of prestige fragrance sales associated with offers such as these (Sexton, 1987), such offers
are gaining ground in other industries as well. The efficacy of
free gifts is, however, in doubt, with detractors claiming that
giveaways take away from future sales, and defenders arguing that they increase them (Sexton, 1987). Companies also
differ markedly in their beliefs about the efficacy of such promotions: whereas over 50% of sales at Esteè Lauder are associated with free gift with purchase promotions (Trager,
1984), Chanel does not offer any (Matthews 1995), and
Aramis is moving from gifts to sampling, with over a quarter
of its marketing budget devoted to free samples of newly introduced products (Nayyar, 1993).
We argue that although a free product enhances the
transactional value of the purchase, it brings with it a mixed
bag of inferences that consumers may draw on the basis of
the offer. These inferences may be both about the product
being offered as a gift and about the brand offering the freebie. Although a free offer may increase sales of the promoted product by increasing the value of the transaction,
this article examines how it may affect stand-alone sales of
the product offered as a free gift. Identifying the situations
in which it does can help managers decide whether or not
to use free offers to help sales and brand managers to decide whether they are willing to allow their product to be
offered for free. Examining the moderating effect of manner of communication of the free offer—specifically, the
value of the free gift, suggests how managers should communicate a free gift offer.
JOURNAL OF CONSUMER PSYCHOLOGY, 14(1&2), 181–185
Copyright © 2004, Lawrence Erlbaum Associates, Inc.
Requests for reprints should be sent to Priya Raghubir, Haas School of
Business, University of California, Berkeley, CA 94720–1900. E-mail:
raghubir@haas.berkeley.eduTwo experiments examine the process by which free gift
promotions serve as a source of information about the underlying value of the free gift. We propose the value-discounting
hypothesis: By virtue of being offered as a free gift, products
will be valued less as evinced by lower purchase intentions
and a lower price that consumers are willing to pay for them.
We also propose that the value-discounting effect will be attenuated when (a) alternate price information is present to
make judgments about the value of the gift, and (b) the value
of the promoted brand is contextually available.
Testing the strength of the value-discounting effect, in
Study 1, a strong brand, “Cross” pens, is used as the free gift.
Results show that people are willing to pay less for the pen
when it has been offered for free with a cheaper brand as
compared to a more expensive one. This result underlines the
importance of brand managers carefully choosing partners in
a promotion, such that their brand equity is not eroded. Study
2 extends the examination to assess whether (a) the effect
seen for a single product offered as a free gift can percolate to
the entire product category, and (b) the value-discounting inference is made at the time of encoding information. In a
strong and subtle test of the value-discounting hypothesis,
we manipulated the order of display of a free gift offer of
pearls along with purchase of a cognac before or after another offer for an unrelated pearl product. The actual information provided is constant across conditions. Despite information equivalence in the two conditions, purchase
intentions of a branded pearl product that belongs to the same
product category (pearls) as the free gift are adversely affected when the free gift offer precedes the unrelated branded
pearl offer.
Across the two studies, we find evidence for value discounting of a free gift: a product that has been offered as a
free gift may later find it difficult to be a stand alone product—consumers may be less willing to purchase it or willing
to pay less for it. The value-discounting effect is moderated
by the type of brand offering the free gift and by the presence
of value information about the free gift—both alternate
sources of information that consumers can use to make inferences about the value of the gift. When the price of the promoting product is known, but the value of the free gift offer is
ambiguous, people appear to use an expected discount rate to
impute the value of the gift. This leads to higher priced promoting brands being perceived as offering gifts of higher
value.
The following hypotheses are developed and are followed
by a description of the two studies.
THE VALUE-DISCOUNTING HYPOTHESIS
Recent models of the manner in which price promotions
work have suggested that consumers make inferences based
on promotional offers, particularly when they do not have access to alternate sources of information to make judgments
that serve as inputs into their purchase decisions. Specifically, prior work has shown that to the extent a promotion
is used as a source of information to infer the quality of a
product (e.g., Raghubir & Corfman, 1999), its price (Inman
et al., 1990; Raghubir, 1998), or consumer demand (Inman et
al., 1997), the overall positive economic effect of offering a
promotion on purchase intentions may be undercut—being
less positive than it would have been in the absence of unfavorable inferences. At the limit, the overall effect of offering
a price promotion on sales may be null or negative. Investigating “purchase with purchase” promotions, Simonson,
Carmon, and O’Curry (1994) found that such promotions decreased the probability of a brand being chosen. For example,
when Pillsbury cake mix buyers were allowed the option of
purchasing a Doughboy Collector’s plate for $6.19, as compared to not being offered such a promotion, they were less
likely to purchase the item if they did not want the plate. This
is despite the fact that the second purchase was a cost-free
option that did not have to be used. Simonson et al. proposed
that this reduction was because consumers who were not interested in purchasing the additional product used this as a
reason to justify not availing of the promotion.
We propose an additional route through which a similar
effect could manifest: via inferences regarding the value of
the product being given away at a discounted price or, in our
case, for free—the value-discounting hypothesis. Consumers
believe that marketers are in the market to make a profit.
However, promotional offers lead to the manufacturer sharing profit margins with consumers. Consumers can resolve
this apparent inconsistency in multiple ways. One way is to
attribute the promotional offer to competitive forces, making
it relatively uninformative about the traits of the brand
(Raghubir & Corfman, 1999). When external attributions are
less likely, then promotional offers may become informative
about aspects of the brand. The fact that a manufacturer is
providing a free gift along with purchase of their product,
could either imply that the product itself was overpriced, or
that the free gift was of low value—that is, free gift promotions could lead to inferences about the cost and margin
structure of the promoted product or the free gift or both.
This is because, whereas consumers may believe that a manufacturer is offering a promotion to increase sales, they also
realize that the promotion will erode profit margins. If it
does, then they would expect a manufacturer to offer a promotion up to the point that it remains profitable for them.
Consumers may, therefore, reason that if a product is available as a gift with another product, it is because the promoted
product has a high margin that covers the cost of the free gift,
because the gift is low cost, or both.
If consumers make the inference that the cost of the free
gift is low, this should lower the price they are willing to
pay for it when it is offered as a stand-alone product. Thus,
if consumers make this type of inference, termed the
value-discounting hypothesis, a free gift offer should lead
consumers to lower their reservation prices for a product
182 RAGHUBIRoffered as a free gift as compared to when it has not been
offered as a free gift.
MODERATORS OF THE
VALUE-DISCOUNTING HYPOTHESIS
In a model of how consumers infer product price based on the
value of the coupon offered, Raghubir (1998) suggested that
consumers’ have a broad range of expectations about discount rates, (e.g., 20%–40%). When they do not know the
price of the product on which a coupon promotions is being
offered, they compute this price based on their expectation of
the average discount rate and the value of the coupon. Mirroring this argument in the context of free gifts, if consumers
are unaware of the value of the promotional offer, they can
compute this value via the price of the product, given an expectation of discount rates in the industry. Thus, a free gift of
ambiguous value should be perceived to be of higher value
when it is offered by a higher priced brand as compared to a
lower priced brand.
Further, the higher the price of the brand offering a free
gift, the higher consumers should expect its margin to be.
Given that higher margin brands can better afford to offer
more costly gifts along with purchase to reward their consumers for purchase, the value inference about the free gift
should be impeded. The same argument can be made through
economic information theory. Recent evidence suggests that
brands of higher value may have more to lose than those of
lower value, and would be less likely to be associated with a
cheap product (Rao, Lu, & Ruekert, 1999).
Based on the theory that a free gift is a source of information, it should be used to the extent alternate sources of information are unavailable to consumers. In support of this,
Raghubir and Corfman (1999) showed that the inference of a
price discount leading to perceptions of poor quality was
moderated by the product category expertise of consumers—novices were more likely to make the unfavorable inference from the promotional offer. Although individual
sources of information about product quality are one source
of information that consumers can use to make judgments,
other sources include the purchase preconditions required to
avail of the free gift, and the contextual presence of value information of the promotional offer.
The presence of alternate information should attenuate the
value-discounting effect as it provides an alternate source of
information by which consumers can make price judgments.
In other words, it allows consumers to anchor on the stated
price of the free gift, even if the full claim of its value is not
accepted (Blair & Landon, 1981). Prior literature has shown
that the presence of contextually provided price information
moderates the effect of promotional cues on price inferences.
For example, Raghubir (1998) showed that when the price of
a product was given along with the amount of the coupon
promotion, the coupon was uninformative of product price
and was not used to make price inferences. Thus, when price
information about the free gift is absent, the value-discounting hypothesis should operate. However, if the price is present alongside the price of the promoted product, then consumers have access to multiple sources of information to
compute a price for the free gift. When the value of the free
gift is mentioned, even if consumers do not believe it is worth
the amount advertised, they can use this value as an anchor
for the price of the product. This is likely to inhibit the inference that the product being offered for free is of low value.
STUDY 1
A strong test of the value-discounting hypothesis is to examine whether a well-known brand name may be prone to the
value-discounting effect. In this study, we test the value-discounting effect by examining whether price perceptions of a
well-known, high-priced brand, Cross pens, are affected by
whether the pen is offered as a freebie by a more or less expensive brand.
Method
Participants. Participants were 74 undergraduate students of the Introductory Marketing course at the Hong Kong
University of Science and Technology who participated in
this study for partial course credit.
Design and procedure. Participants were given an ad
copied from a leading airlines duty-free catalog. The ad was
an offer of a free gift with purchase of a bottle of alcohol. The
design was 2 (brand name of free gift: unbranded/well
known) × 2 (brand name of promoting product: less/more expensive), mixed design. The brand name of the free gift was
manipulated by offering either an unbranded key chain or a
Cross pen, and the brand name of the promoting product was
manipulated via the free gift being offered by either Bombay
Sapphire Gin or Royal Salute whiskey. Although the former
sells for $16 in the duty-free catalog, the latter sells for $90.
Both are well-known brand names in their respective product
categories. Participants saw an ad where the whiskey offered
the Cross pen and the gin offered the key chain, or the reverse—an ad where the whiskey offered the key chain and
the gin offered the Cross pen.
Measures are described in the Results section.
Results
Manipulation checks. Deal evaluations for the two offers were elicited on 7-point scales (1 = not at all; 7 = a very
good deal) to assess the face validity that the Cross pen offer
was perceived to be better than the key chain offer. As expected, gin was perceived to offer a better deal when it offered the Cross pen versus the key chain, Ms = 4.06 vs. 2.21,
FREE GIFTS WITH PURCHASE: PROMOTING OR DISCOUNTING THE BRAND? 183F(1, 72) = 21.56, p < .001. The same pattern was true for the
whiskey, Ms = 4.71 vs. 2.14, F(1, 72) = 51.74, p < .001.
Price perceptions of the free gift. All participants
indicated the maximum price they would be willing to pay
for a Cross pen, using an open-ended scale. As predicted by
the value-discounting hypothesis, when the Cross pen had
been offered by Royal Salute whiskey, reservation prices
were one third higher (M = $28.78), as compared to when it
had been offered by Bombay Salute gin, M = $21.24, F(1, 63)
= 3.69, p < .05.
These results show that even when the free gift is a well
known brand name, in and of itself, the price consumers are
willing to pay for it is affected by the brand that offers it.
STUDY 2
This study examines whether the price judgment about a specific product offered for free may also carry over to the product category in general. This should be particularly true if
people do not have well-formed beliefs about the product category. We test whether offering a pearl bracelet for free with
purchase of a bottle of cognac in a duty-free catalog affects
the price people are willing to pay for a branded pearl product
that appears later in the catalog (the target ad). We test that to
the extent consumers have made an inference about the product being offered as a free gift, this should carry through to
related products and affect their purchase intentions, even if
they have not been offered for free. Thus, even controlling for
the information available to consumers, the order of exposure
to two different advertisements could affect the manner in
which they are perceived, if the value-discounting inference
is made at the time of encoding the free gift promotional offer. This is because, to the extent that an inference has been
formed at the time of exposure to the free gift offer, this is
likely to carry through while consumers process the pearls
offer they are later exposed to.
Based on the value-discounting hypothesis, we expect
that when consumers have been exposed to a free pearl bracelet gift offer, they are less likely to purchase pearls at full
price. Based on the value-discounting hypothesis, if consumers draw the “pearls are cheap” inference, then they should be
willing to pay less for branded pearls advertised in the target
ad, when they have been exposed to the free pearl bracelet
message as compared to when they have not. Further, if this
inference occurs at the time of encoding the offer, then the order of exposure of promotional information should affect the
price people are willing to pay. If the inference has been
made at the time the target ad is seen, the effects should obtain. On the other hand, the effects would be weaker if the
free gift ad was viewed and the value-discounting inference
made after the target ad was viewed.
The purchase price of the brand should also directly affect
inferences of the value of the free gift itself —the higher the
price of the promoted product offering the free gift, the
higher the reservation prices of pearls—both those offered as
a free gift and those offered as a stand-alone product. Further,
the value of the brand offering the free gift should moderate
the effect of exposure to the free gift on judgments about the
unrelated pearl necklace. However, when the price of the free
gift is provided, consumers should be less likely to undergo
these inferential processes, attenuating the value-discounting
effect.
Method
Study participants. Participants were 75 undergraduate students of the Introductory Marketing course at the Haas
School of Business at University of California, Berkeley,
who participated in this study for partial course credit.
Design and procedure. Participants were given a set
ofadscopiedfromaleadingairlinesduty-freecatalog.Thisset
included the target ad for a Majorica pearl necklace and bracelet, and the ad that served as a manipulation —a cognac ad offering a double-strand pearl bracelet as a free gift with purchase. Participants were asked to flip through the set of ads, as
though they were flipping through a duty-free in-flight magazine, and then respond to dependent measures at the end.
The design was a between-subject 2 (order: free gift ad
first/later) × 2 (brand: high price/low price) × 2 (price of free
gift: present/absent) full factorial. The order of the ads manipulated whether consumers saw the ad for the Majorica pearls
before forming an impression of pearls based on the free gift
ad, versus without this. Brand name was manipulated by using
two differently priced cognacs offering the free gift: 700 ml
NapoleonCourvoisier(US$52)and700mlHennessyXO(US
$90). The third variable was manipulated by including a blurb
“US Value $32.50” next to the pearl offer in the cognac ad in
the price present condition. This was missing in the price absent condition. The target ad was an ad for a Majorica convertible necklace/bracelet set, priced at $149.
The key measures were reservation prices for the free gift
bracelet and Majorica pearls, elicited using an open-ended
measure; and purchase intentions for Majorica pearls elicited
on a 7-point scale. These were followed by general questions
about travel in airplanes, time spent reading in-flight magazines, and so on, designed to increase the credibility of the
cover story.
Results
Means are provided in Table 1. Analyses are presented by dependent measure.
Judgments about the free pearl bracelet. A 2 × 2 ×
2 analysis of variance (ANOVA) on price perceptions yielded
a main effect of brand name, F(1, 60) = 4.69, p < .05, such
that consumers were willing to pay 50% more for the free
184 RAGHUBIRpearl bracelet when they had seen it associated with the
Hennessy XO cognac (M = $32.03), as compared to the
lower priced Napoleon Courvoisier (M = $21.09). This replicates the results found with Cross pens in Study 1.
Carryover effects to Majorica pearls. A 2 × 2 × 2
ANOVA on the reservation price measure showed a main effect of brand, F(1, 60) = 5.31, p < .05, such that consumers
were willing to pay a higher price for the Majorica pearls
when the free gift of pearls was given by the more expensive
Hennessy XO cognac (M = $64.00), as compared to the relatively less expensive Napoleon Courvoisier (M = $37.27).
This is consistent with the model that consumers use brand
costs to infer the value of the gift.
Evidence that value discounting occurs at encoding.
A similar analysis on the purchase intentions measure
showed a main effect of order, F(1, 67) = 3.06, p < .05, reflecting that intentions to purchase the Majorica pearls were
higher in the condition when the cognac ad followed (M =
2.54) rather than preceded the Majorica pearls offer (M =
1.92). No other effect was significant at p < .05.
We found that people were willing to pay higher prices for
boththefreebieandforthebrandedpearlproductwhenthegift
had been offered by a higher priced brand. We also found that
purchase intentions of Majorica pearls were unfavorably impacted when participants were first exposed to a cognac ad that
offered a free pearl bracelet as a free gift with purchase, a
strong yet subtle test of the value-discounting hypothesis. Surprisingly, the price of the freebie did not exert main or interaction effects. It is, however, premature to draw a conclusion
fromthesenullresultsgiventherelativelysmallsamplesize.
GENERAL DISCUSSION
Studies 1 and 2 support the informational role of price promotions in the context of free gift offers. Results show that
free gift with purchase promotions can lead to inferences
about the value of the gift. Thus, (a) consumers were found to
discount the value of the gift, (b) with these effects carrying
over to the general product category to which the free gift belonged; (c) consumers behaved as though they imputed gift
value using the price of the brand offering the gift, leading to
(d) consumers attaching higher value to the same gift if it was
given by a higher priced brand name.
This article adds to the growing body of evidence that
price promotions are more than just money off—they are a
source of information that consumers use to make judgments
about products and their prices. This has been demonstrated
in the context of free gift offers for the first time.
The article also sheds light on how consumers process
free gift offers. As Simonson et al. (1994) found, the presence of a promotional offer could actually backfire and lead
to reduced choice probabilities as compared to the absence of
the promotional offer. They explained their findings in terms
of the promotion providing a reason to disqualify the brand
from a choice set. In this article, we show another effect—being offered as a free gift can cheapen the brand itself.
Areas for Future Research
The value-discounting hypothesis offered in this article, argued that the apparent incongruity of a manufacturer giving
away something for free can be resolved by devaluing what is
given away. However, consumers have an alternate way to resolve this incongruity if (a) the free gift is part of the product
line of the main promoted product and is a repeat purchase
item,or(b)thepriceofthefreegiftisprovidedandisincongruently high. Instead of focusing on the manager’s need to make
aprofitabletransactionintheimmediateterm,consumersmay
think of the free gift offer as a trial-inducing device. In this
case, the attribution for the promotion is not so much to increase the sales of the promoted product, but to increase future
sales of the free gift item. Economists have proposed that price
promotion offers may signal high quality (for a recent review,
FREE GIFTS WITH PURCHASE: PROMOTING OR DISCOUNTING THE BRAND? 185
TABLE 1
Study 2 Results
Value of Pearls Absent Value of Pearls: $32.50
Means Napoleon Hennessy X.O. Napoleon ($52) Hennessy ($90)
Purchase intention fo r Majorica pearls
Cognac ad first 1.56 2.22 1.78 2.11
(n = 9) (n = 9) (n = 9) (n = 9)
Pearls ad first 3.11 2.50 2.70 1.90
(n = 9) (n = 10) (n = 10) (n = 10)
Reservation price: Free gift of pearls
Cognac ad first $12.17 $26.86 $25.78 $31.89
Pearls ad first $20.00 $42.50 $23.00 $25.30
Reservation price: Majorica pearls
Cognac ad first $34.29 $81.86 $38.56 $63.33
Pearls ad first $34.29 $73.50 $40.30 $40.22see Kirmani & Rao, 2000) as low-quality products cannot afford to subsidize trial. If this type of reasoning dominates, then
offering a product for free should increase its purchase likelihood postdeal retraction. To examine the scenarios where this
could occur is an area for future research.
ACKNOWLEDGMENTS
The author thanks Chau Ly for her research assistance in this
project. Comments from seminar participants at Columbia
University, Dartmouth College, Harvard, INSEAD, MIT, Simon Fraser University, University of Florida, University of
Toronto, and Wharton are gratefully acknowledged.
REFERENCES
Blair, E., & Landon, E. L. (1981). The effect of reference prices in retail advertisements. Journal of Marketing, 45, 61–69.
Blattberg, R. C., & Neslin, S. A. (1990). Sales promotion: Concepts, methods, and strategies. Englewood Cliffs, NJ: Prentice Hall.
Inman, J. J., McAlister, L., & Hoyer, W. D. (1990). Promotion signal: Proxy
for a price cut? Journal of Consumer Research, 17, 74–81.
Inman, J. J., Peter, A. C., & Raghubir, P. (1997). Framing the deal: The role
of restrictions in accentuating deal value. Journal of Consumer Research,
24, 68–79.
Kirmani, A., & Rao, A. R. (2000). No pain, no gain: A critical review of the
literature on signalling unobserved product quality. Journal of Marketing,
64(2), 66–79.
Matthews, I. (1995, October 20). Heaven scent. Marketing Week, 31, 67–71.
Nayyar, S. (1993, January 18). Aramis shifts from gifts to sampling.
Brandweek, 34, 5.
Raghubir, P. (1998). Coupon value: A signal for price. Journal of Marketing
Research, 35, 316–324.
Raghubir, P., & Corfman, K. P. (1999). When do price promotions affect
brand evaluations? Journal of Marketing Research, 36(2), XXX–XXX.
Rao, A. R., Qu, L., & Ruekert, R. W. (1999). Signalling unobservable
quality through a brand ally. Journal of Marketing Research, 36,
258–268.
Sexton, N. (1987, March 2). Toiletries and beauty aids: GWPs and PWPs
losing edge as promotional tools. Advertising Age, 58, S4–S5.
Simonson, I., Carmon, Z., & O’Curry, S. (1994). Experimental evidence on
the negative effect of product features and sales promotions on brand
choice. Marketing Science, 13(1), 23–40.
Trager, C. S. (1984, May 10). Promotions blemish cosmetics industry. Advertising Age, 55, M22–23, M26.
186 RAGHUBIR

Source: http://groups.haas.berkeley.edu/marketing/PAPERS/PRIYA/paper2.pdf