The Effects of Promotional Bundling on Consumers’ Evaluations of Product Quality and Risk of Purchase

Judy Harris, University of Houston
ABSTRACT - This paper presents the results of an experiment designed to assess the effects of promotional bundling on consumers' perceptions of product quality and perceived risk of purchase. The results indicate that for a new product that is not a brand extension of an established product, promotional bundling with the established product can increase perceptions of product quality and decrease perceptions of risk among buyers of the established product. However, for a new product that is a brand extension, promotional bundling can decrease the perceived quality of the new product and increase the perceived risk ssociated with the purchase. Perceptions of the new product by non-buyers of the established product are not affected by bundling.
[ to cite ]:
Judy Harris (1997) ,"The Effects of Promotional Bundling on Consumers’ Evaluations of Product Quality and Risk of Purchase", in NA - Advances in Consumer Research Volume 24, eds. Merrie Brucks and Deborah J. MacInnis, Provo, UT : Association for Consumer Research, Pages: 168-172.

Advances in Consumer Research Volume 24, 1997      Pages 168-172

THE EFFECTS OF PROMOTIONAL BUNDLING ON CONSUMERS’ EVALUATIONS OF PRODUCT QUALITY AND RISK OF PURCHASE

Judy Harris, University of Houston

ABSTRACT -

This paper presents the results of an experiment designed to assess the effects of promotional bundling on consumers' perceptions of product quality and perceived risk of purchase. The results indicate that for a new product that is not a brand extension of an established product, promotional bundling with the established product can increase perceptions of product quality and decrease perceptions of risk among buyers of the established product. However, for a new product that is a brand extension, promotional bundling can decrease the perceived quality of the new product and increase the perceived risk ssociated with the purchase. Perceptions of the new product by non-buyers of the established product are not affected by bundling.

Recently, researchers have shown increasing interest in consumer response to bundling (e.g., Mazumdar and Jun 1993; Yadav 1994; Harlam, Krishna, Lehmann and Mela 1995). Bundling generally involves combining of two or more products into a single offering, but there are many forms that a bundle may take. The marketing literature has defined bundling as narrowly as single-product combinations that are physically packaged together, and as broadly as products that are implicitly linked by complementary usage situations (e.g., Mulhern and Leone 1991). Most consumer research, however, has focused on explicit bundles made up of separate products sold for a single price (Simonin and Ruth 1995).

This research concerns promotional bundling, one specific type of bundling that has received scant empirical investigation. With this form of bundling, consumers are offered a discount or premium when they buy different products together, such as toothpaste and a toothbrush or cake mix and frosting. As examples, consider the following promotional offers drawn from recent Sunday newspaper FSIs (free standing inserts):

- Save $1.00 on the purchase of new Lay’s dip when you buy Baked Lay’s Potato Crisps;

- Save $1.50 on Aleve pain reliever with any Vick’s product;

- Save 25 cents on the purchase of both Austex Beef Stew and Austex Chili.

Despite a trend toward more widespread use of cooperative sales promotions (Varadarajan 1986), little is known about how consumers respond to offers such as the ones above. Research has demonstrated that evaluations of one product in the bundle are influenced by the quality of the other products in the bundle (Gaeth, Levin, Chakraborty and Levin 1990; Yadav 1994), such that perceptions of a product should be enhanced by associating it with a high quality product. However, it has been argued also that consumers may react negatively to the increased restrictions and lack of flexibility involved with the purchase of a bundle, and will decrease product evaluations accordingly (see Wilson, Weiss and John 1990; Kinberg and Sudit 1979). Moreover, in the case of a promotional bundle, consumers’ heuristic beliefs about price discounts and unnecessary promotions might lead to lower evaluations of the quality of the product (e.g., Sawyer and Dickson 1984; Simonson, Carmon and O’Curry 1994).

The purpose of the research reported here is to examine whether consumers’ evaluations of a product are, in fact, affected positively or negatively by promotional bundling. We use the context of a new product introduction and investigate how the promotional bundling of a new product with an established product affects consumers’ responses to the new product. First, we review relevant past literature. Second, we report the results of an experiment designed to assess the effect of promotional bundling on evaluations of a new product that either is or is not a brand extension of the bundling partnr.

CONCEPTUAL BACKGROUND

Consumers may evaluate the worth of a product differently when it is in a bundle than it is not. Explicitly linking two or more items together is likely to influence the context in which consumers evaluate those items because it will literally force the consumer to evaluate them in the context of one another. Past research has consistently demonstrated that consumers’ evaluations of and preferences for a product can be influenced by the context of the choice task, including the set of alternatives considered and the manner in which these alternatives are presented (e.g., Simonson and Tversky 1992). Consistent with this viewpoint, past research in bundling has demonstrated that both the framing of the bundle and the perceived quality of the items in the bundle can influence how consumers evaluate the bundle as a whole and the individual items it contains. This research is reviewed below.

Potential Positive Effects of Bundling on Product Evaluations

Past research has shown that how consumers perceive the quality of one item in the bundle can affect how they perceive the quality of the other items. Gaeth, et al. (1990) found that the perceived quality of a tie-in product offered with a VCR had an influence on subjects’ perceptions of the quality of the VCR that far outweighed the monetary contribution of the premium to the bundle as a whole. In addition, Simonin and Ruth (1995) found that attitudes toward the items included in a promotional bundle affect attitudes toward the bundle itself. These findings are consistent with arguments that one potential benefit of joint promotions (whether or not they involve bundling) is a reduction of the perceived risk involved with the purchase of an untried product (Varadarajan 1986). That is, bundling a new product with an established product can serve as an endorsement for the new product. These arguments lead to our first hypothesis:

H1A: Promotional bundling of a new product with an established product will:

a) increase the perceived quality of the new product; and

b) decrease the perceived risk associated with the purchase of the new product

Potential Negative Effects on Evaluations

Although it has been recognized that bundling a product with another product is unwise if the second product is perceived unfavorably by consumers (Gaeth, et al. 1990), scholars have paid much less attention to the potential negative effects that bundling itself may have on product evaluations. Wilson, Weiss and John (1990) and Kinberg and Sudit (1979) argued that some consumers will evaluate the total worth of a bundle less favorably than the sum of the items it contains because of the additional restrictions that the purchase of a bundle entails. Additionally, Diamond and Sanyal (1990) demonstrated that consumers will choose a straight coupon offer over a promotional bundle worth more. They argued that this was because consumers perceived the promotional offer as a reduced loss rather than a gain, but they also suggested that it may have been due to negative reactions toward the loss of freedom associated with the additional requirement of purchasing two products. The logic underlying the latter argument is that a promotional bundle, by offering the consumer a discount and then placing a restriction on it, may evoke psychological reactance and cause the offer to be perceived as less favorable than either a straight promotional offer or no promotional offer would have been (see also Clee and Wicklund 1980). Although these arguments pertain to consumers’ evaluations of the bundle itself, rather than evaluations of the individual items in the bundle, negative feelings toward the offering as a whole are likely to color perceptions of the items as well (Simonin and Ruth 1995).

It is also possible that consumers will view the quality of the items in the bundle less favorably when they are bundled. They may attribute the fact that the manufacturer or retailer is offering a promotion to low product quality and a high risk of not being satisfied with the purchase (see Sawyer and Dickson 1984), or react negatively to a promotion that seems unnecessary (Simonson, Carmon and O’Curry 1994). Although this argument applies to straight promotional offers (e.g., a cents-off coupon or a premium offer) as well as promotional bundling, it is possible that a promotional bundle may be more suspect either because it is, in a sense, two promotions in one, or because it is more unusual and may attract more attention than a traditional discount offer. These arguments lead to a hypothesis that is offered as an alternative to H1A.

H1B: Promotional bundling of a new product with an established product will:

a) decrease the perceived quality of the new product; and

b) increase the perceived risk associated with the purchase of the new product.

We offer H1A and H1B as co-equal, alternative hypotheses, because arguments for both effects can be found in the literature. Promotional bundling may have a favorable effect on perceived quality (and therefore perceived risk) of a brand because of the implied endorsement of the bundling partner (H1A). On the other hand, promoional bundling may have an unfavorable effect on perceived quality (and therefore perceived risk) because of psychological reactance caused by additional purchase requirements or suspicion about a relatively unfamiliar type of promotion (H2A). Of course, the relative strength of these opposite effects may be affected by moderating variables. The next section of the paper discusses the likely effects of one potential moderator: the brand name of the product.

Brand Name of the New Product

A brand extension should be more likely to suffer negative consequences due to bundling than a new name product. This is because a brand extension does not need the endorsement of the bundling partner to help reduce the perceived risk involved with the purchase. It already has that endorsement by virtue of its name, and therefore will suffer only the negative effects of bundling. On the other hand, a new name product has more to gain by bundling. This argument leads to our second hypothesis:

H2: The effects of promotional bundling of a new product with an established product will:

a) decrease the perceived quality associated with the purchase of the new product when it is a brand extension, but increase it when the product has a new name; and

b) increase the perceived risk associated with the purchase of the new product when it is a brand extension, but decrease it when the product has a new name.

EXPERIMENT

The study used a 2 (promotional bundle or straight discount) x 2 (brand extension versus new name) between-subjects design. The established product was Cheerios cereal and the new product was a round (log-shaped) cereal bar. The bar was either named Cheerios cereal bars or Fruit’n’Oat cereal bars. Additionally, subjects were classified as Cheerios cereal buyers or non-buyers based on whether they had chosen Cheerios out of all cereals available at a local supermarket in a previous task.

Seventy-five cents was chosen as the value of the coupon because an examination of FSI coupons in a local newspaper indicated that it is a common amount for discounts on cereal and cereal bars. Cheerios was chosen as the established product because a pretest (n=75) indicated that it was a popular product with a student sample (25% "market share" when subjects were asked to indicate a choice among all the cereals available in a local supermarket). Cereal bars were chosen as the new product because a second pretest (n= 52) indicated that student subjects purchased cereal bars (86% reported purchasing cereal bars recently). Moreover, the subjects in this pretest thought that the "fit" to Cheerios cereal was relatively high (mean= 6.46 on an 8-point scale ranging from "very poor fit" to "very good fit"). For comparison, the mean rating of "fit" for oatmeal (chosen by Broniarczyk and Alba [1994] as a catgory that is very relevant to Cheerios) in this pretest was 5.31 on the same 8-point scale. The fit of the product was important given that research on brand extensions has shown that product associations transfer most readily among products with a moderately good fit (Aaker and Keller 1990).

The name Fruit’n’Oat was chosen as the new name because it described the bar and was rated as appropriate by the 18 pretest subjects rating it (the other pretest subjects rated other options). The mean rating for Fruit’n’Oat was 5.82 (on an 8-point scale ranging from "very poor name for a cereal bar" to "very good name for a cereal bar"). The rating of the name "Cheerios Cereal Bars" was 6.28 on the same scale. These means were not significantly different (t1,33=0.74, n.s.).

Procedure

One-hundred and fifty three undergraduate marketing students participated in the study. The subjects were randomly assigned to one of the four treatment conditions.

The study was conducted in two parts. First, the subjects were asked to imagine that they were grocery shopping for their household and were given a booklet containing listings of all the brands in several product categories that were available on the shelves of a local supermarket. The brand name, size, flavors and prices of the brands in the categories of soft drinks, cold breakfast cereal, kitchen trash bags and facial tissues were provided. The subjects were asked to indicate which of the brands they would be most likely to purchase given that they were out of the product and the choices listed were what was available on the store’s shelves. For each category, the subjects were given the option of not choosing any of the brands listed.

After a distraction task (an unrelated questionnaire), the subjects were given a second booklet containing brands, sizes, flavors and prices of all the brands in several product categories (pasta sauce, cereal/ granola bars, cold breakfast cereal, kitchen trash bags and facial tissue). The list for cereal/granola bars included either the Cheerios or Fruit’n’Oat brand, as well as 10 additional brands of breakfast bars. The price, size and flavor information was based on comparable information for Nutri-Grain cereal bars.

In this task, the subjects again were asked to imagine that they were shopping, but that this time they had the option of using coupons. They were to assume that if they choose a product for which they had a coupon, the shelf price listed in the booklet would be reduced by that amount (with no doubling or tripling). The subjects were instructed that this questionnaire reflected a different shopping trip than the first questionnaire. They could, but did not have to choose the same brands as before. The subjects were then instructed that they should not choose to use the coupons just because they were in the booklet, but rather they should use a particular coupon only if they would be likely to use it in real life.

The coupons were included in advertisements designed to mimic FSI ads, and were bound into the second booklet opposite the appropriate product list. Coupons for Ragu pasta sauce, the appropriate brand of cereal bars (either Fruit’n’Oat or Cheerios) and Glad trash bags were included in the booklet. The coupons for the pasta sauce and trash bags were created from actual FSIs. The coupon for the cereal bars was created using words and pictures from a Cheerios cereal box and Nutri-grain ceeal bar box and FSI. The cereal bar FSI contained the picture of the box of cereal bars (which were called Cheerios or Fruit’n’Oat), a picture of the Cheerios cereal box (in the bundled conditions only) and a coupon (good for 75 cents off the purchase of the bar or 75 cents off the purchase of the bar when you also buy Cheerios cereal). After the choice task, the subjects were asked to evaluate several brands, including the appropriate brand of cereal bars.

Dependent Measures

The dependent measures were subjects’ evaluations of the quality of the cereal bars and the perceived risk associated with their purchase. The quality of the bars was measured by averaging subjects’ ratings of the bars on two seven-points scales ranging from "low quality" to "high quality" and "inferior brand" to "superior brand" (r=0.83). The perceived risk of purchasing the bars was measured by averaging subjects’ ratings of the bars on two seven-point scales ranging from "likely to be unsatisfied if purchased" to "likely to be satisfied if purchased" and "risky purchase" to "safe purchase" (r=0.86). Choice of the cereal bar was collected as a potential dependent variable, but was not used because the number of subjects choosing the bar over the other options available was too small (less than 10% in all conditions).

RESULTS AND DISCUSSION

Recall that H1A predicted that promotional bundling would have beneficial effects on perceptions of the cereal bar, H1B predicted detrimental effects, and H2 predicted an interaction between bundling and brand name such that bundling was more beneficial for products with a new name and more detrimental for brand extensions. To test these hypotheses, we conducted separate analyses of variance (ANOVAs) appropriate for unequal sample sizes on perceived quality and perceived risk. The test between H1A and H1B concerned the direction and significance of the main effect for bundling in each analysis, and the test of H2 concerned the significance of the interaction effect between bundling and brand name.

The analyses of the effects of brand name and bundling on perceived quality revealed a significant main effect for brand name (F1,149=57.94, p<.0001; with Cheerios bars evaluated more favorably than Fruit’n’Oat bars), a significant main effect for bundling (F1,149=6.60, p<.01; with the bundled conditions evaluated less favorably than the non-bundled conditions), and a significant interaction effect (F1,149=9.50, p<.01). Individual planned contrasts indicated that the Cheerios bars were perceived less favorably when they were bundled than when they were not bundled (F1,149=56.68, p<.0001), but the Fruit’n’Oat bars were not (F1,149= 0.16, n.s.). The presence of a significant main effect for bundling was consistent with Hypothesis 1B, and indicated that promotional bundling had a negative effect on evaluations of quality. The presence of a significant interaction effect was consistent with Hypothesis 2, and indicaed that this negative effect occurred for the brand extension, but not for the new name product. Means and standard deviations are given in Table 1.

Similarly, the analysis of the effects of brand name and bundling on perceived risk revealed a significant main effect for brand name (F1,149=15.94, p<.0001; with Cheerios bars evaluated as less risky than Fruit’n’Oat bars), and a marginally significant interaction effect (F1,149=3.80, p<.06). Individual planned contrasts indicated that the Cheerios bars were perceived as more risky when they were bundled than when they were not bundled (F1,149=3.81, p<.06), but Fruit’n’Oat bars were not (F1,149=0.66, n.s.). These results were consistent with Hypothesis 2, and indicated that promotional bundling of a new product with an established product increases the perceived risk of a new product that is a brand extension, but has no effect for a new product that is not a brand extension. The means and standard deviations are given in Table 2.

From these results, it appears that the promotional bundling of a brand extension with the established product harms evaluations of the extension’s quality and risk. The promotional bundling of a new name product with an established product is neither helpful nor harmful to the new name product.

Differences in Evaluations by Cheerios Buyers and Non-buyers

The analyses discussed in the previous section were conducted on the sample as a whole, with no distinction between those subjects who were Cheerios buyers and those who were not. We repeated the analyses of variance on the perceived quality and perceived risk associated with the purchase of the cereal bars, adding whether or not the subject had chosen Cheerios cereal in the first shopping task as a second potential moderator of the effect of bundling (along with brand name). The analyses of the effects of bundling, brand name and previous purchase on the perceived quality of the cereal bar indicated a significant three-way interaction (F1,145=6.72, p<.01). Individual planned contrasts revealed that among the non-Cheerios buyers, the pattern was similar to that found in the total population, with bundling significantly decreasing evaluations of quality for the brand extension (F1,145=4.53, p<.05), but not for the new name product (F1,145=0.26, n.s.). However, among Cheerios customers, bundling significantly decreased evaluations of quality for the brand extension (F1,145=11.49, p<.001), but marginally increased evaluations for the new name product (F1,145=3.56, p<.06). In fact, among Cheerios customers, there was no significant difference between the Cheerios brand name and the Fruit’n’Oat brand name within the promotional bundling conditions (F1,145=0.43, n.s.). The means and standard deviations are given in Table 1.

TABLE 1

EFFECTS OF BRAND NAME, PROMOTIONAL BUNDLING AND PRIOR PURCHASE ON EVALUATIONS OF THE QUALITY OF THE CEREAL BARS

Similar effects were revealed in the analysis on perceived risk. Again, there was a significant three-way interaction (F1,145=7.78, p<.001). Individual planned contrasts revealed that among the non-Cheerios buyers, there were no significant differences between the bundled and non-bundled conditions for either the brand extension (F1,145= 0.21, n.s.) or the new name product (F1,145=0.03, n.s.). However, among the Cheerios buyers, bundling significantly increased perceptions of risk for the brand extension (F1,145=6.31, p<.05), but significantly decreased perceptions of risk for the new name product (F1,145=5.04, p<.05). Again, among Cheerios customers, there was no significant difference between the Cheerios brand name and the Fruit’n’Oat brand name within the promotional bundling conditions (F1,145=0.12, n.s.). The means and standard deviations are given in Table 2.

Promotional bundling appears to be a bad idea for a product introduced as a brand extension under all of the circumstances examined in this study. However, the results indicate that promotional bundling may be a viable strategy for introducing a new product that is not a brand extension. Promotional bundling increased perceptions of quality and decreased perceptions of risk for a new name product among customers of the bundling partner, and had no effect (either positive or negative) among non-customers of the bundling partner.

CONCLUSIONS AND LIMITATIONS

The usual limitations of studies using only student subjects apply to this study. However, pretests were used to ensure that the products used in the sample were relevant to student consumers and the stimulus materials were designed to be as realistic as possible for a pencil and paper choice task.

A second potential limitation is the use of only one of many potential variations on the bundling situation. For example, pretests indicated that the new product and bundling partner "fit" together well, but there are several potential ways that two products may fit (Aaker and Keller 1990; Broniarczyk and Alba 1994). Our product pair had a strong conceptual, or "image" fit, but did not have a high usage fit. Since many promotional bundles involve product complements, future research should examine if promotional bundling is a more viable marketing technique with such a product pair.

Similarly, previous research indicates that the framing of a bundling offer may affect perceptions of value (e.g., Yadav 1995; Harlam, et al. 1995). Our research used only one of many potential frames (i.e., "buy both and save"). Future studies may want to assess the extent to which framing the promotional offer in different ways affects how the offer is perceived.

Despite these limitations, this research has implications for how promotional bundling offers should be used in practice. As indicated by the examples of promotional bundles listed earlier in this paper, many such offers involve brand extensions. The results of this study indicate that this practice is likely to have negative consequences on how consumers perceive the quality of the products and the risk associated with purchasing the product.

Our results also indicate that promotional bundling may be a useful promotional technique for a new product introduced with a new brand name. Although consumers’ evaluations and perceptions of risk were most favorable in the condition in which the brand extension was introduced without bundling, there may be situations in which the manufacturer chooses not to use a brand extension. For example, the manufacturer may not want to risk the brand name of the established product on a new product introduction or may not want to forfeit the opportunity to create a new name (Aaker 1990). The results of this study indicate that under such circumstances, bundling may be a way to enhance consumers’ perceptions of the new product. Given the risks involved in new product introduction and the large number of new product failures that occur every year (Aaker 1990), this may be a situation in which marketers can use any help they can get.

TABLE 2

EFFECTS OF BRAND NAME, PROMOTIONAL BUNDLING AND PRIOR PURCHASE ON EVALUATIONS OF PERCEIVED RISK OF THE BARS

REFERENCES

Aaker, David (1990), "Brand Extensions: The Good, the Bad, and the Ugly," Sloan Management Review, Summer, 47-56.

Aaker, David and Kevin Keller (1990), "Consumer Response to Brand Extensions," Journal of Marketing, 54 (January), 27-41.

Broniarczyk, Susan M. and Joseph W. Alba (1994), "The Importance of the Brand in Brand Extension," Journal of Marketing Research, 31 (May), 214-228.

Clee, Mona A. and Robert A. Wicklund (1980), "Consumer Behavior and Psychological Reactance," Journal of Consumer Research, 6 (March), 389-405.

Diamond, William D. and Abhijit Sanyal (1990), "The Effect of Framing on the Choice of Supermarket Coupons," in Advances in Consumer Research, 17, Marvin Goldberg and Gerald Gorn, eds., Provo, UT: Association for Consumer Research, 488-493.

Gaeth, Gary J., Irwin P. Levin, Goutam Chakraborty and Aron M. Levin (1990), "Consumer Evaluations of Multi-Product Bundles: An Information Integration Perspective," Marketing Letters, 2 (December), 47-57.

Harlam, Bari A., Aradhna Krishna, Donald R. Lehmann and Carl Mela (1995), "Impact of Bundle Type, Price Framing and Familiarity on Purchase Intention for the Bundle," Journal of Business Research, 33, 57-66.

Kinberg, Yoram and Ephraim F. Sudit (1979), "Country/Service Bundling in International Tourism," Journal of International Business Studies, 10 (Fall), 51-63.

Mazumdar, Tridib and Sung Youl Jun (1993), "Consumer Evaluations of Multiple versus Single Price Changes," Journal of Consumer Research, 20 (December), 441-450.

Mulhern, Francis J. and Robert P. Leone (1991), "Implicit Price Bundling of Retail Products: A Multi-Product Approach to Maximizing Store Profitability," Journal of Marketing, 55 (October), 63-76.

Sawyer, Alan G. and Peter R. Dickson (1984), "Psychological Perspectives on Consumer Response to Sales Promotion," in Research on Sales Promotions: Collected Papers, Katherine E. Joss, ed. Cambridge, MA, Marketing Science Institute.

Simonin, Bernard L. and Julie A. Ruth (1995), "Bundling as a Strategy for New Product Introduction: Effects on Consumers’ Reservation Prices for the Bundle, the New Product, and its Tie-in," Journal of Business Research, 33, 219-230.

Simonson, Itamar and Amos Tversky (1992), "Choice in Context: Tradeoff Contrast and Extremeness Aversion," Journal of Marketing Research, 29 (Aug), 281-295.

Simonson, Itamar, Ziv Carmon, and Suzanne O’Curry (1994), "Experimental Evidence on the Negative Effect of Product Features and Sales Promotions on Brand Choice," Marketing Science, 13 (Winter), 23-40.

Varadarajan, P. Rajan (1986), "Horizontal Cooperative Sales Promotion: A Framework for Classification and Additional Perspectives," Journal of Marketing, 50 (April), 61-73.

Wilson, Lynn O., Allen M. Weiss and George John (1990), "Unbundling of Industrial Systems," Journal of Marketing Research, 27 (May), 123-138.

Yadav, Manjit S. (1995), "Bundle Evaluation in Different Market Segments: The Effects of Discount Framing and Buyers’ Preference Heterogeneity," Journal of the Academy of Marketing Science, 23 (3), 206-215.

Yadav, Manjit S. (1994), "How Buyers Evaluate Product Bundles: A Model of Anchoring and Adjustment," Journal of Consumer Research, 21 (Septeber), 342-353.

----------------------------------------