The Role of Brand Familiarity on the Impact of Word-Of-Mouth Communication on Brand Evaluations

ABSTRACT - The findings of this study indicate that the impact of word-of-mouth (WOM) on brand evaluations (purchase intentions and brand attitudes) is moderated by brand familiarity. While consumer evaluations become less favorable for both familiar and unfamiliar brands as a result of negative WOM, the negative messages have a detrimental effect on unfamiliar brands than familiar brands. Conversely, while positive WOM benefits both familiar and unfamiliar brands, the unfamiliar brand has more to gain from positive WOM. Overall, unfamiliar brand evaluations are more susceptible to change as a result of WOM communication than familiar brand evaluations.

Citation:

D.S. Sundaram and Cynthia Webster (1999) ,"The Role of Brand Familiarity on the Impact of Word-Of-Mouth Communication on Brand Evaluations", in NA - Advances in Consumer Research Volume 26, eds. Eric J. Arnould and Linda M. Scott, Provo, UT : Association for Consumer Research, Pages: 664-670.

Advances in Consumer Research Volume 26, 1999      Pages 664-670

THE ROLE OF BRAND FAMILIARITY ON THE IMPACT OF WORD-OF-MOUTH COMMUNICATION ON BRAND EVALUATIONS

D.S. Sundaram, Indiana University East

Cynthia Webster, Mississippi State University

ABSTRACT -

The findings of this study indicate that the impact of word-of-mouth (WOM) on brand evaluations (purchase intentions and brand attitudes) is moderated by brand familiarity. While consumer evaluations become less favorable for both familiar and unfamiliar brands as a result of negative WOM, the negative messages have a detrimental effect on unfamiliar brands than familiar brands. Conversely, while positive WOM benefits both familiar and unfamiliar brands, the unfamiliar brand has more to gain from positive WOM. Overall, unfamiliar brand evaluations are more susceptible to change as a result of WOM communication than familiar brand evaluations.

Consumers frequently rely on word-of-mouth (WOM) communications to make purchase decisions. The importance of WOM in the marketplace is well recognized because of the definitive role it plays in shaping consumers’ attitudes and purchase behavior. Research on WOM communication has examined the extent to which the effectiveness of the WOM communication process is affected by several predictors, such as the valence (positive or negative) of WOM communication, characteristics of WOM receivers and providers, and various situational factors (Arndt 1967). Thugh such studies have undoubtedly expanded our understanding of the WOM phenomena, there are other salient aspects of WOM communication that remain virtually unexplored. One related, unexplored issue pertains to how the characteristics of the target of WOM (which could be a product, a company, or a brand) affect consumers’ reactions to WOM communication.

Past studies provide evidence attesting to the notion that characteristics of the product, company, or brand (i.e., the target of WOM communications) may affect how consumers process the WOM message. For example, the parameters of the relationship between marketplace WOM communications and consumers decision-making behavior were found to vary according to whether the WOM target was a good or a service (Weinberger and Dillon 1980). The results of this study revealed that both favorable and unfavorable product information produced greater change in consumers’ purchase intentions for services than for goods. Another study revealed that the characteristics of the WOM target, i.e, services’ heterogeneity and intangibility, were cited by consumers as the primary reasons why they perceive greater risk when purchasing a service and subsequently place more importance on WOM when making purchase decisions (Murray 1991). Similarly, in an investigation of how consumers process unfavorable information pertaining to recall of defective products, Mowen (1980) found that consumers’ reactions to unfavorable communication varied by their familiarity with the company. Mowen’s study showed that an unfamiliar company was perceived to be more responsible for a defect than a familiar company.

Brand name is one particular variable that is likely to moderate the processing of WOM messages. Brand name is an inherent product characteristic and is considered a significant factor influencing consumers’ perceptions (Low and Fullerton 1994). The salience of brand name is well acknowledged because consumers often use brand name as a major cueCsometimes, the only cueC in making purchase decisions (Low and Fullerton 1994).

Brand familiarity, one specific component of the brand name concept, affects various facets of consumer decision making. Brand familiarity has been found to affect information search process (Biswas 1992), product evaluation and choice heuristics (Raju 1977), advertising message processing (Kent and Allen 1994), and ultimate brand choice (Hoyer and Brown 1990). Given that consumers exhibit purchase behavior variations depending on brand familiarity and that the impact of WOM varies according to the WOM target familiarity, it is only logical to suspect that brand name familiarity may play an important role in determining how consumers process and respond to WOM communications. Therefore, we conducted a study to meet three objectives. First, we examine the effect of brand familiarity on brand evaluations. Second, we investigate the impact of the nature of the WOM communication (positive and negative) on brand evaluations. And finally, we examine the interaction effect of brand familiarity with WOM communication on brand evaluations.

BRAND FAMILIARITY

Brand familiarity reflects the brand-related experiences accumulated by the consumer (Alba and Hutchinson 1987). Increased brand familiarity may be due to exposure to the brand in advertisements or in a store, recognition of the brand name, and prior purchase and/or usage of the brand. Past research provides some insight into how brand familiarity affects consumer purchase decisions. For instance, Hoyer and Brown (1990) found that when consumers were asked to select a brand from a given choice set, those who were familiar with a brand tended to select the known brand although it was relatively lower in quality, while those who were unfamiliar with the brands in the given choice set sampled more brands and selected the higher-quality brand. Based on the findings of their study, the authors suggest that brand familiarity ay influence consumers’ information processing style and their ultimate brand choice. Another study found that brand familiarity moderates the recall of advertising message (Kent and Allen 1994). Lane and Jacobson (1995) found that brand familiarity also influences brand’s performance in the stock market. An experimental study found that greater brand name familiarity enhanced purchase intentions of both automobile insurance and photocopying services (Arora and Stoner 1996). Research evidence also indicates that brand familiarity moderates information search. For example, Biswas’s (1992) study revealed that consumers intend to spend less time shopping for a familiar brand than they do for an unfamiliar brand.

Viewed collectively, these studies suggest that consumers react more favorably toward a familiar brand than they do toward an unfamiliar brand. Possibly, consumers will behave in a similar pattern when processing WOM messages about a product or a brand. Thus, we expect that consumers’ attitude toward a brand following exposure to WOM communication will vary depending on their brand familiarity, with a familiar brand being evaluated more favorably than an unfamiliar brand.

H1: Brand familiarity will moderate brand evaluations (purchase intentions and brand attitudes), with familiar brand receiving more favorable evaluations than an unfamiliar brand.

Both the positive and negative WOM communications (PWOM and NWOM, respectively) exert strong influence on consumers’ behavior. PWOM is likely to enhance consumers’ brand image and ensuing purchase intentions (Arndt 1967). Conversely, NWOM is likely to dissuade potential buyers from considering a particular brand and damage the reputation of the brand. In other words, PWOM is likely to have a facilitating and NWOM an inhibiting effect on consumers’ brand evaluations.

H2: The nature of WOM communications will impact brand evaluations (purchase intentions and brand attitudes); the PWOM will strengthen and the NWOM will weaken brand evaluations.

Consumers familiar with a brand possess a much more detailed, rich cognitive structure with respect to that brand than a less familiar brand. Higher levels of brand familiarity, generated through direct or indirect brand-related experiences, are associated with a well developed knowledge structure about the brand and its attributes (Alba and Hutchinson 1987). Often, the brand-related experiences may lead to development of brand evaluations that are less susceptible to change. Given that consumers are less likely to change their attitude toward a brand especially when it pertains to familiar brands (Hoyer and MacInnis 1997), it is expected that exposure of familiar brand to WOM communications, either positive or negative, may not produce significant change in its pre-existing brand evaluations. On the other hand, less familiar brands that are usually associated with not so well developed knowledge structure are more susceptible to change in brand evaluations following exposure to any new brand-related communications. The reason is that consumers less familiar with a brand are more amenable to process the new brand-related information and change their brand evaluations based on the valence of the information.

More specifically, it is expected that unfamiliar brands will benefit more from PWOM messages than familiar brands because PWOM associated with unfamiliar brands may act as a risk reducer to the WOM recipients and thereby enhance their confidence and subsequent brand evaluations. On the other hand an unfamiliar brand, compared with a familiar brand, may sustain greater damage in brand evaluations due to NWOM because consumers’ prevailing uncertainty nd fear initiated by the brand unfamiliarity will be further magnified by the negative brand-related information.

Overall, though NWOM communications will lead to unfavorable brand evaluations for both familiar and unfamiliar brands and PWOM communications will strengthen brand evaluations for both familiar and unfamiliar brands, the magnitude of the impact of WOM messages will be different for familiar and unfamiliar brands. It is expected that an unfamiliar brand, compared with a familiar brand, may sustain greater change in brand evaluations due to WOM. Since the impact of WOM may vary by brand familiarity, an interaction effect between brand familiarity and WOM is anticipated.

H3: There will be a significant interaction effect of brand familiarity with WOM communication on brand evaluations (purchase intentions and brand attitudes).

RESEARCH METHOD

The study employed a 2 X 3 between subjects experimental design. The factors manipulated were brand name familiarity (familiar or unfamiliar), and WOM (PWOM, NWOM, or no WOM). The subjects for the study were drawn from a pool of undergraduate students enrolled in the business college of a midwestern university. To control for the confounding effect of subjects’ prior purchasing experience and to maintain simultaneously participants’ involvement, the product selected for this study had to meet the following two characteristics: (1) not previously purchased by respondents and (2) likely to be purchased in the future. A pretest participated by 30 students evaluated twenty-four product categories on the previously stated criteria. Based on their evaluation room-air conditioner was selected as the product for this study.

Another pretest, participated by 34 students, was used to identify a familiar and an unfamiliar brand for the study. The subjects were provided with twenty brand names (identified through a search of trade publications) and were then asked to indicate their familiarity with each brand on a 7-point (familiar/unfamiliar) scale. Based on their ratings, Whirlpool (familiar) and Electrohome (unfamiliar) were selected for brand familiarity manipulations.

The study scenarios required the participants to assume that they are in the process of buying a new air-conditioner and collecting information on several brands by visiting retail stores and scanning advertisements. As a part of the brand manipulations, the participants were provided with an advertisement for either a familiar or an unfamiliar brand; thus the particular brand was encountered during prepurchase search. The advertisements (which used real ads.) were identical except for the manipulated brand names. WOM communication was manipulated as follows. For the PWOM condition, the participants were provided with the following message:

At a recent social gathering, you mentioned your plan to buy a new air conditioner to your friend and also mentioned having seen (manipulated brand name). Your friend said that (manipulated brand name) is highly reliable, simple to install, durable, and has low maintenance and energy cost.

For the NWOM condition, the participants read the following scenario:

At a recent social gathering, you mentioned your plan to buy a new air conditioner to your friend and also mentioned having seen (manipulated brand name). Your friend said that (manipulated brand name) is not reliable, breaks down too often, not easy to install, and has high maintenance and energy cost.

The participants assigned to no-WOM condition were exposed only to manipulation of brand familiarity. These participants were treated as control groups.

The subjects, randomly assigned to one of the six experimental scenarios, read the descriptions and responded to the questions regarding brand evaluation, that is, purchase intentions and brand attitude. A total of 120 individuals, 20 per scenario, participated in the study. Forty percent of the respondents were male and the average age was 26.4 years. About 72.4 percent were employed full- or part-time.

Manipulation Checks. In the later part of the questionnaire, the subjects were asked to indicate their level of familiarity with the brand name to which they were exposed, using a 9-point scale where 1=not familiar and 9=very familiar. An analysis of variance test indicated significant differences (F=106.03, p=.0001) between unfamiliar (x=2.36) and familiar (x=6.20) treatments, suggesting brand familiarity was effectively manipulated.

To evaluate the perceived realism of the scenarios, subjects were asked to respond to the item "I believe the situations described in the scenario can actually happen in real life," using a 9-point scale (1=strongly disagree and 9=strongly agree). A resulting mean score of 6.28 (standard deviation=2.4) suggests that the subjects considered the scenarios to be very realistic. In addition, to determine the effectiveness of manipulation of WOM messages, subjects were required to rate the WOM referral messages on a 5-point scale. Subjects exposed to the NWOM message rated the referral using the anchors "Not Very Negative" and "Extremely Negative" and those exposed to PWOM messages rated using the anchors "Not Very Positive" and "Extremely Positive". Eighty percent of the subjects exposed to NWOM message rated the referral message as very or extremely negative, and 75 percent of those exposed to PWOM rated it as very or extremely positive.

TABLE 1

RESULTS OF FACTOR LOADINGS

Hypothesis Testing

The two dependent variables, purchase intentions and brand attitude, were measured using multiple items. Purchase intention was operationalized by asking the subjects to indicate their likelihood of purchase, overall product liking, perceived brand reputation, perception of after-sales service, and risk associated with the purchase of the exposed brand. The six items to measure brand attitude were adopted from Alpert and Kamins (1995). These items were measured using nine-point itemized rating scale.

A principal components factor analysis of the eleven items with varimax rotation resulted in a two-factor solution. The risk item that loaded on both the factors was removed and the factor analysis was re-run. The final factor solution, shown in Table 1, has an acceptable loading pattern and explains 83.5 percent of the construct variance. Since the reliability of these two scales, estimated using Cronbach’s alpha, is above the norm with 0.93 for purchase intentions and 0.95 for brand attitude, the items corresponding to each dependent variable were averaged to create measures of the dependent variables. Further, since the estimate of correlation coefficient between the two dependent variables was 0.85 (p=0.001), the data were analyzed using both MANOVA and ANOVA.

FINDINGS

Results of the MANOVA tests, presented in Table 2, indicate that the main effects of brand familiarity (Wilks’=0.68, p<0.001), type of WOM (Wilks’=0.29, p<0.001), and interaction of brand familiarity and type of WOM (Wilks’=0.88, p=0.01) are significant.

The ANOVA results indicate that the main effect of brand familiarity on purchase intention (F_1,114=42.8, p<0.001) and brand attitude (F_1,114=40.54, p<0.001) are significant. Further, examination of the means revealed that the ean value for purchase intentions was greater for the familiar brand (x=6.06, std. dev.=1.68) than for the unfamiliar brand (x=4.77, std. dev.=2.17). Similarly, the mean value for brand attitude was greater for the familiar brand (x=6.14, std. dev.=1.16) than for the unfamiliar brand (x=4.67, std. dev.=1.64). These findings indicates that brand familiarity has a significant effect on brand evaluations and a familiar brand is more likely to receive favorable evaluation than an unfamiliar brand. Thus, Hypothesis 1 was supported.

The main effect of type of WOM message on purchase intentions (F_1,114=124.9, p<0.001) and brand attitude (F_1,114=47.23, p<0.001) are significant. Examination of means reveals that the mean value for purchase intentions is greater in the PWOM message treatment (x=7.27, std. dev.=0.94) than in the NWOM message treatment (x=3.38, std. dev.=1.33), with the mean in no-WOM message treatment as 5.54 (std. dev.=1.19). Similarly, the mean value for brand attitude is greater in the PWOM treatment (x=6.60, std. dev.=0.92) than in the NWOM message treatment (x=4.26, std. dev.=1.51). The mean brand attitude rating in no-WOM treatment is 5.33, with a standard deviation of 1.25. These results confirm the hypothesis that type of WOM message has a significant impact on brand evaluations, with PWOM messages resulting in favorable evaluations and NWOM messages producing an opposite effect. Thus, H2 was supported. Further, examination of deviations of means for PWOM and NWOM treatments with respect to no-WOM treatment for both purchase intention and brand attitude revealed that NWOM was more prepotent than PWOM message on effecting brand evaluations.

The interaction effects of brand familiarity with type of WOM message on purchase intentions (F_2,114=4.89, p=0.009) and on brand attitude (F_2,114=4.74, p=0.01) are significant. These findings lend support to Hypothesis 3. A pictorial representation of these two two-way interactions presented in Figures 1 and 2 reveal that both PWOM and NWOM messages produce significantly different levels of purchase intentions and brand attitude depending on brand familiarity.

TABLE 2

SUMMARY OF MANOVA AND ANOVA RESULTS

FIGURE 1

INTERACTION EFFECT OF BRAND FAMILIARITY AND WOM ON PURCHASE INTENTION

The cell means presented in Table 3 can be examined to better understand the impact of PWOM and NWOM messages on brand evaluations for familiar and unfamiliar brands. Because the control group was not provided with any WOM messages, their mean purchase intentions and brand attitude ratings can be used as reference point to estimate the magnitude of the impact of PWOM and NWOM communication. As anticipated, NWOM communication was associated with lower purchase intentions and brand attitude for both familiar and unfamiliar brands. However, the unfamiliar brand, when exposed to NWOM communication, lost 2.3 and 1.55 units (computed as the difference between means for control and NWOM groupsCsee Table 3) on purchase intentions and brand attitude ratings, respectively. In contrast, for familiar brand, the purchase intentions rating dropped by only 2.11 units and the brand attitude rating dropped by 0.69 units due to NWOM communication. These findings suggest that an unfamiliar brand, compared with a familiar brand, experiences greater damage due to NWOM messages.

A completely different scenario emerges when the impact of PWOM communication on purchase intentions and brand attitude are examined. As expected, PWOM increased purchase intentions and brand attitude ratings for both familiar and unfamiliar brands. The familiar brand, when exposed to PWOM, gained 1.09 and 0.87 units (computed as the difference between means for control and PWOM groups) on purchase intentions and brand attitude ratings, respectively. In contrast, for unfamiliar brand the purchase intention rating increased by 2.33 units and brand attitude rating increased by 1.63 units due to PWOM. The findings suggest that gain in mean rating on purchase intention and brand attitude due to PWOM is greater for unfamiliar brand than for familiar brand.

FIGURE 2

INTERACTION EFFECT OF BRAND FAMILIARITY AND WOM ON BRAND ATTITUDE

TABLE 3

INTERACTION EFFECT OF WOM AND BRAND FAMILIARITY (MEANS AND STANDARD DEVIATIONS)

DISCUSSION

This study demonstrates that the effect of WOM on brand evaluations is moderated by brand familiarity. The results reported here, apart from confirming the earlier findings that WOM messages (Arndt 1967) and brand familiarity (Hoyer and Brown 1990) affect brand perceptions, provide insight into the interaction effect of WOM and brand familiarity on brand evaluations.

Support for the main effect of brand familiarity indicates that familiar brands receive more favorable evaluations than unfamiliar brands. Further, significantly greater mean ratings for purchase intentions and brand attitude associated with familiar brands, as compared to unfamiliar brand, for each experimental condition (no-WOM, PWOM, NWOM) is an indication of competitive advantage familiar brands enjoy in the marketplace. Further, our study indicates that if both the familiar and unfamiliar brands receive PWOM or NWOM communication to an equal degree in the marketplace, familiar brands will continue to maintain their advantage over unfamiliar brands.

The finding lending support for the main effect of WOM communication indicates that PWOM messages enhance brand evaluations and NWOM messages reduce brand evaluations for both familiar and unfamiliar brands. Albeit not novel, this study reaffirms the principle that brand managers should avoid NWOM and attempt to stimulate PWOM communication to ensure success in the marketplace.

The findings suggest that unfamiliar brand benefits more from PWOM messages than familiar brand. Further, unfamiliar brand experiences a greater reduction in brand evaluations due to NWOM than familiar brands. While the increase in mean purchase intentions and brand attitude ratings for familiar brand due to PWOM were 1.09 and 0.87 units, corresponding increases for unfamiliar brands were 2.33 and 1.63 units. Further, while the decrease in mean purchase intentions and brand attitude ratings for familiar brand due to NWOM were 2.11 and 0.69 units, corresponding decreases for unfamiliar brands were 2.3 and 1.55 units. The results lend support to our expectation that unfamiliar brands more than familiar brands are susceptible to change in brand evaluations due to WOM.

The results of this study indicate that though it is important for all brands to avoid NWOM communication, it is more important for unfamiliar brands to avoid negative messages because of the drastic impact such negativity has on brand evaluations. The likelihood of creating a success of unfamiliar brands can be increased by generating PWOM communication associated with those brands. As can be seen in Figures 1 and 2, an unfamiliar brand receiving PWOM is likely to be favored over a familiar brand that is experiencing NWOM or no WOM messages.

On the other hand, a familiar brand does not necessarily require PWOM (unless it is viewed negatively). Of course, PWOM messages will further strengthen its competitive position in the marketplace. Even the familiar brand should avoid NWOM as much as possible, because any NWOM messages pertaining to familiar brands will reduce purchase intentions significantly. While one NWOM message may not reduce favorable attitude towards familiar brand, repeated NWOM communication is likely to produce significant negative consequences. Further, an unfamiliar brand experiencing high levels of PWOM communication may threaten the survival of familiar brands.

Though WOM is not a marketing communication tool to be manipulated by marketers like advertising and sales promotion, it is an important marketplace communication tool to effect some consumer responses. Therefore, it is important for managers to have an understanding of the role of WOM communication in the marketplace. It is possible that many less familiar brands (or less known businesses) that can not afford to spend on advertising and sales promotion could rely on existing customers to generate positive referrals and thereby manage to survive among large and familiar brands.

An examination of the means reveals that the predictors affect these two dependent variables, purchase intentions and brand attitude, in the same manner. However, brand attitude ratings exhibt less fluctuation than purchase intention ratings. This may be due to long-term cognitive evaluations of brand attitudes and the relatively short-term nature of purchase intentions.

In this study, the experimental scenario provided information for only one brand factor at two levels (familiar or unfamiliar). Whereas, in reality a consumer is likely to be exposed to multiple brands during a typical prepurchase search and may receive PWOM for one brand and NWOM communication for another brand. Possibly, exposure to PWOM or NWOM messages for one brand may impact how consumers react to other brands in their choice set. Therefore, future research might employ a research design that parallels reality by providing the participants with WOM communication for more than one brand.

REFERENCES

Alba, Joseph W, and J. Wesley Hutchinson (1987), "Dimensions of Consumer Expertise," Journal of Consumer Research, 13 (March), 411-454.

Alpert, Frank H., and Michael A. Kamins (1995), "An Empirical Investigation of Consumer Memory, Attitude, and Perceptions Toward Pioneer and Follower Brands," Journal of Marketing, 59 (October), 34-45.

Arndt, Johan (1967), "Role of Product-Related Conversations in the Diffusion of a New Product," Journal of Marketing Research, 4 (August), 291-295.

Arora, Raj, and Charles Stoner (1996), "The Effect of Perceived Service Quality and Name Familiarity on the Service Selection Decision," Journal of Services Marketing, 10, 22-34.

Biswas, Abhijit (1992), "The Moderating Role of Brand Familiarity in Reference Price Perceptions," Journal of Business Research, 15, 251-262.

Hoyer, Wayne D., and Steven P. Brown (1990), "Effects of Brand Awareness on Choice for a Common, Repeat-Purchase Product," Journal of Consumer Research, 17 (September), 141-148.

Hoyer, Wayne D., and Deborah J. MacInnis (1997), Consumer Behavior, Houghton Mufflin Co: MA.

Kent, Robert J. and Chris T. Allen (1994), "Competitive Interference Effects in Consumer Memory for Advertising: The Role of Brand Familiarity," Journal of Marketing, 58 (July), 97-105.

Lane, Vicki, and Robert Jacobson (1995), "Stock Market Reactions to Brand extension Announcements: The Effects of Brand Attitude and Familiarity," Journal of Marketing, 59 (January), 63-77.

Low, George S., and Ronald A Fullerton (1994), "Brand, Brand Management, and the Brand Manager System: A Critical-Historical Evaluation," Journal of Marketing Research, 31 (May), 173-190.

Mowen, John C. (1980), "Further Information on Consumer Perceptions of Product Recall," in Advances in Consumer Research, Vol. VII, Jerry C. Olson (ed.), San Francisco, CA, 519-523.

Murray, Keith B. (1991), "A Test of Services Marketing Theory: Consumer Information Acquisition Activities," Journal of Marketing, 55 (January), 10-25.

Raju, P.S. (1977), "Product Familiarity, Brand Name, and Price Influences on Product Evaluations," in Advances in Consumer Research, 4, 64-71.

Weinberger, Marc G., and William R. Dillon (1980), "The Effects of Unfavorable Product Rating Information," in Proceedings of the Association for Consumer Research, Vol. 7, J. Olson, ed., Ann Arbor, MI.: Association for Consumer Research.

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