An Investigation of the Nature of Word of Mouth Communication Across Adoption Categories For a Food Innovation

Citation:

Russell W. Belk and Ivan Ross (1971) ,"An Investigation of the Nature of Word of Mouth Communication Across Adoption Categories For a Food Innovation", in SV - Proceedings of the Second Annual Conference of the Association for Consumer Research, eds. David M. Gardner, College Park, MD : Association for Consumer Research, Pages: 470-475.

Proceedings of the Second Annual Conference of the Association for Consumer Research, 1971     Pages 470-475

AN INVESTIGATION OF THE NATURE OF WORD OF MOUTH COMMUNICATION ACROSS ADOPTION CATEGORIES FOR A FOOD INNOVATION

Russell W. Belk, University of Minnesota

Ivan Ross, University of Minnesota

The broad objectives of this study centered on a longitudinal view of new product acceptance and coincident informal communications. At the level of the individual, comparisons were planned between the earliest triers of the product, later adopters, and non-adopters. At an aggregate level, points in time one month and two years after product introduction were selected to develop descriptions of product usage and word of mouth activity. And at a level combining individual and aggregate observations, it was hoped that time of adoption, brand loyalty, brand attitude, and word of mouth behaviors could be integrated into a model of the role of informal product-related conversations in the diffusion of a new food product.

To gather data for these perspectives, a two wave field study was designed with four separate samples: (1) a sample of those housewives who were aware of an innovative food product one month after its introduction, (2) a sample of the same respondents two years later, (3) a sample of those not initially aware of the product, conducted two years after their rejection from the first sample, and (4) a general sample of those aware of the product two years after its introduction. This paper will deal primarily with findings from samples one and four.

METHOD

In early 1961, General Foods' "Maxim" coffee became the first freeze dried coffee in the Minneapolis-St. Paul market. By early 1971, there were five brands of freeze dried coffee in this market and area sales had leveled off for Maxim. These two periods were selected to investigate innovator and non-innovator behaviors for the new food product. The first sample obtained 134 telephone interviews from a randomly selected sample of the area telephone directories during a 10 day period in February, 1969. A second sample of 28 of these respondents was re-interviewed during February of 1971. Procedures paralleled the original sample and questionnaires were extended to gather data on the period since the first interview. The fourth sample of 128 interviews was collected by similar methods during February, 1971, from a sampling population constructed on the same basis as in sample one. Questions were directed toward behaviors in the past one and one-half months in order to create floor and recall effects comparable to conditions in the original interviews. Additional data were obtained on exposure to and usage of Maxim since its introduction.

RESULTS AND DISCUSSION

While all results of the study cannot be accommodated in this paper, results on six hypotheses present a fairly concise description of the role of word of mouth in adoPtion of the new product.

Hypothesis 1:

Word of mouth about Maxim is more likely to be generated by recipients of word of mouth about Maxim than by non-recipients.

TABLE 1

RECEIVERS WHO SEND

TABLE 2

RECEIVERS WHO SEND

One means of legitimizing discussion of a product or brand may be hearing others using the same topic in conversation. However, the relationship between sending and receiving appears only for the general sample two years after product introduction. In part the lack of sender-receiver overlap in sample one may reflect the fact that there were relatively more potential receivers and relatively fewer potential senders (i.e. persons aware of Maxim) at this point soon after the product had been introduced. This explanation could not account for all of the difference between samples however, and argues for a different type of word of mouth during the two time periods. While there appears to be at least a de facto opinion leadership in the introductory stage, the high overlap of senders and receivers in the second period suggests a more generalized type of word of mouth.

It may be noted also that there is a lower incidence of word of mouth present in the fourth sample. It is likely that conversational relevance wanes somewhat as the product becomes accepted and as competing brands enter the market. It is also possible that former "opinion leaders" have moved on to other conversational topics, leaving Maxim to general discussion. A low number of recent word of mouth occurrences in sample two (1 of 28 persons) also tends to support this view.

Hypothesis 2:

Those who have either a very favorable or a very unfavorable reaction when they try Maxim will report more word of mouth than those whose reaction was less extreme.

TABLE 3

TRIER WOM BY SATISFACTION RATING

TABLE 4

TRIER WOM BY SATISFACTION RATING

While word of mouth in the first sample occurred most frequently among triers with extreme evaluations, later adopters showed more frequent word of mouth when their evaluations of the product were low. Though interpretation is not entirely clear, perhaps the novelty aspect of an extremely good or bad product make it an enticing topic of conversation only for the more self-confident innovators. Results would also support an interpretation that only among later adopters does word of mouth play a role in information acquisition and problem solving, and therefore further input is sought where evaluations are non-definitive. A greater utilization of word of mouth as an initial source of information for later adopters as opposed to innovators (14% versus 6%) would add to the explanation of increasingly functional word of mouth, but the greater possible exposure time confounds these results.

Hypothesis 3:

Those who have purchased Maxim will report earlier word of mouth conversation than those who have not yet purchased the brand.

TABLE 5

MEAN NUMBER OF WEEKS SINCE:

These data were not obtained from sample four. This hypothesis speculated that the critical period for both frequency and effectiveness of word of mouth centers around the purchase decision. One difficulty with this formulation is that the post-purchase discussion of one group of buyers may coincide with the pre-purchase discussion of another. Another difficulty is that this study does not record a purchase decision not to buy. All together 44% of receptions occurred before purchase (17% during, 39% after), and 50% of transmissions were delivered before recipient purchase (12.5% during, 37.5% after).

Hypothesis 4:

Those who have engaged in word of mouth about Maxim will have purchased Maxim more frequently than those not sending or receiving word of mouth.

TABLE 6

MEAN NUMBER OF MAXIM PURCHASES:

TABLE 7

MEAN NUMBER OF MAXIM PURCHASES:

While the direction of causality is not apparent in these data, the timing of purchase and word of mouth occurrences suggests that receiving word of mouth encourages trial and sending facilitates retrial.

Hypothesis 5:

Those consumers who were formerly brand loyal to another brand of coffee (before using Maxim) are more likely to report the initiation of word of mouth about Maxim than are those who were not formerly brand loyal to another coffee.

TABLE 8

WOM BY BRAND LOYALTY (TRIERS ONLY)

Hypothesis 6:

Among purchasers of Maxim, those who were formerly brand loyal to another coffee will have first purchased Maxim at a later date than those who were not formerly brand loyal.

TABLE 9

MEAN NUMBER OF WEEKS SINCE FIRST MAXIM PURCHASE

All measures of brand loyalty defined the purchase of only one brand in the last three purchases of coffee as brand loyal and a greater number of brands purchased on these occasions as not brand loyal. The sample one data show a slight but non-significant tendency for brand loyal persons to engage in greater word of mouth behavior. Sample four data show a tendency for those formerly brand loyal to be later adopters of Maxim than those who were not. Thus, ties of brand loyalty appear to hamper adoption only among later adopters. There was no significant difference in adopter times by brand loyalty in the first sample innovators.

Although sample two data are based on only 28 observations, some striking differences are found between these Maxim innovators two years later and the sample four members who had only recently tried the product.

While nearly all innovators now rated Maxim as "average" or below, the majority of later adopters rated the product as "good" or "excellent."

Over one-half of the innovator resample said they would not purchase Maxim in the future, as opposed to only 13% of the new sample.

Somewhat surprisingly, fewer innovators said that they had used other new freeze-dried coffees than did later adoPters.

Thus while the appeal of Maxim appears to have worn off for innovators, it is not because they have also become innovators for newer brands of the product. In fact, most innovators are now brand loyal to regular or instant coffees, while few in the new sample are currently brand loyal.

Although purchase behavior seems to have stopped for the innovator sample, these persons continued to engage in word of mouth about the product.

About 1/4 of the resample had engaged in conversations about Maxim since they were originally interviewed, as opposed to a much lower percentage among the new sample for the same time period.

Very few persons in the resampled group ever used Maxim for a long enough period to have been called brand loyal to the product.

Early triers gave nearly all positive comments at the time of the original sample and largely negative word of mouth about the product by the time of resample.

One-half of the early triers now feel that their close friends would not like the product, whereas only one-sixth of the new sample feel that this would be the case with their friends.

Both samples agree that the majority of people now like Maxim, but the resampled group feels more strongly that the product's popularity will decline in the future.

These findings suggest that innovators seem to perceive themselves as different from most and precursers of opinion change, while later adopters view themselves as more typical consumers. It may be that a continued desire for originality sparks the potentially damaging negative word of mouth found now in the innovator group.

CONCLUSIONS

The picture of early triers of this food innovation is very different from that of later adopters of the product. The concept of adoption may not even be appropriate for this frequently purchased item, since the adoption by early triers appears to have only been temporary.

It has been suggested that the nature and functions of word of mouth behavior is markedly different between the two groups. While the early group seem close to the classic description of opinion leaders, the later adopting group seem to be engaging in a more conversational form of word of mouth, probably devoid of opinion leadership. Future research may seek to further clarify differences in the nature and functions of word of mouth between adopter categories and investigate the amount of interaction and influence across category memberships. It would be well to examine these relationships in the context of greater and lesser innovations. The present study recommends primarily that it is an oversimplification to speak of a single type and function of product-related conversations in the adoption process.

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