Market Pioneering, Learning, and Preference

Gregory S. Carpenter, Columbia University
Kent Nakamoto, University of Arizona
ABSTRACT - A number of recent studies suggest that product knowledge and experience affect the organization of a consumer's product perceptions, and consequently, his or her preferences. In this paper, we consider the order of these experiences and the biases that this order may induce in consumer decision processes. We then examine these decision making biases as a potential explanation for the persistent market share advantages that often accrue to successful early market entrants. This paper extends earlier work in this area (e.g., Carpenter and Nakamoto 1987) by more fully developing the behavioral foundation for this type of advantage. We suggest that, in many cases, early entrants frame the consumer's perceptions of the product category, thereby defining the rules of competition. As a result, these brands can distort brand preferences in their favor. Because these advantages are independent of production efficiency or outstanding marketing expertise, they have profound implications for the nature of competition in the market
[ to cite ]:
Gregory S. Carpenter and Kent Nakamoto (1988) ,"Market Pioneering, Learning, and Preference", in NA - Advances in Consumer Research Volume 15, eds. Micheal J. Houston, Provo, UT : Association for Consumer Research, Pages: 275-279.

Advances in Consumer Research Volume 15, 1988      Pages 275-279

MARKET PIONEERING, LEARNING, AND PREFERENCE

Gregory S. Carpenter, Columbia University

Kent Nakamoto, University of Arizona

ABSTRACT -

A number of recent studies suggest that product knowledge and experience affect the organization of a consumer's product perceptions, and consequently, his or her preferences. In this paper, we consider the order of these experiences and the biases that this order may induce in consumer decision processes. We then examine these decision making biases as a potential explanation for the persistent market share advantages that often accrue to successful early market entrants. This paper extends earlier work in this area (e.g., Carpenter and Nakamoto 1987) by more fully developing the behavioral foundation for this type of advantage. We suggest that, in many cases, early entrants frame the consumer's perceptions of the product category, thereby defining the rules of competition. As a result, these brands can distort brand preferences in their favor. Because these advantages are independent of production efficiency or outstanding marketing expertise, they have profound implications for the nature of competition in the market

INTRODUCTION

Memory for product information or past usage experience has been shown in a number of recent studies to have a profound impact on consumer decision behavior. For example, knowledgeable consumers appear to rely more heavily on brands as foci for organizing product information, and on categories and subcategories as bases for grouping brands (Bettman 1986; Johnson and Russo 1984; Sujan 1985). In addition, choice experience results in selective retention of brand information favoring chosen brands (Biehal and Chakravarti 1982). This selective retention continues to favor previously chosen brands even if a previously inferior brand is improved through addition of a new attribute (Biehal and Chakravarti 1983). Thus, choice affects memory, and memory, in turn, affects choice processes.

Because of this ongoing interaction of experience and preference, the order of a consumer's product experiences can affect his or her choice behavior over time. Changing the order of experiences, even if they are the same in aggregate, might well change product perceptions and subsequent choices.

In this paper, we consider the potential impact of the order of a consumer's initial experiences with a product class from the perspective of category learning. We limit ourselves to the consideration of learning situations where the consumer initially has poorly formed expectations regarding the product category. In addition, we focus on products or services where relative performance is either difficult to assess (e.g., vitamins) or dependent on personal taste (e.g., soft drinks). Of particular interest to us is the situation where consumers learn about a product class through experience with one brand a market pioneer.

These brands appear to command significant competitive advantages that often persist for decades. These advantages translate into higher market shares for pioneering brands (Robinson and Fornell 1985; Urban et al. 1986). Moreover, contrary to economic theories for this phenomenon, these advantages arise in the absence of any cost or awareness advantages for the pioneering brand. Carpenter and Nakamoto (1987) develop a consumer-based explanation for these advantages and demonstrate that the process of buyer preference formation can result in a significant pioneer advantage, even in the case where cost and awareness are equalized.

In the present paper, we consider-more fully the behavioral mechanism driving these findings, and extend the theory to explain the persistence of market share advantages accruing to a dominant brand in a category, however that dominance was achieved. In essence, we suggest that the consumer's initial experiences with a product class will frame the consumer's perceptions of all the brands in the product category. More specifically, we propose three ways in which the order of the consumer's product experience can affect consumer perceptions of brands in the category and preferences for the brands. The initial experiences will define the relationship of the product category to related ones. They will define the product attributes that are relevant to decision making (Johnson and Russo 1984). Finally, they will define the attribute configuration that is prototypical of the category. Because of the task environment in which the category structure is developed, we suggest that prototypicality will be closely related to preference.

PRODUCT EXPERIENCE AND CATEGORY LEARNING

In most studies of perceptual category learning, several examples of the category are presented simultaneously or in close succession. The distinguishing features of the category are immediately available and the subject can induce category structure through direct observation (Holland, Holyoak, Nisbett, and Thagard 1986).

Learning about products presents a very different problem for the consumer. Typically, product exposure is sequential with an appreciable period between exposures. It is rare that a broad spectrum of brand information is presented at one time. Thus, category structure must be induced from information in memory. In addition, a consumer's early exposure to a product is often limited to one brand, or a small set of brands, so that the information used to derive category perceptions is biased.

As a result, inferences regarding category membership based on these category perceptions will be affected. In a recent study, Elio and Anderson (1984) investigated the impact of order of presentation of category examples on people's ability to distinguish between members of two related categories. The examples were described in terms of a set of attributes, and categorization was based on a fairly complex rule that was not revealed to the subjects. Rather, subjects learned about the categories by examining sequentially a set of examples of each. For some subjects, highly typical members of each category were presented first, and the full range of attribute variation in each category was introduced later. Others were exposed immediately to the full range of variation. Subjects then classified a novel set of examples into the two categories. They were best able to do this when they had learned about the category by seeing highly typical members first. Elio and Anderson argue that the consistency in the narrower grouping promotes formation of strong category generalizations.

This result suggests that learning about a product category on the basis of a limited set of similar brands leads to strong perceptions of the nature of the category, including product features critical to the definition of the category, and features that would tend to place an item outside the category. Barsalou (1982) refers to these critical features as context-independent. Other attributes are context-dependent and are made salient by a particular task. These attributes might characterize variation among brands within a product category.

Category Learning and Brand Preference

The consumer's view of variation within a product category, like the category definition, would be expected to depend on early product experiences. In other words, early experience would set up an implicit context for product judgments.

Context has been shown to influence judgment in a variety of settings. For example, Tversky (1977) showed that the composition of a set of items affects perceptions of differences between items. Huber, Payne, and Puto (1982) extend these set composition effects to preference judgments, showing that changing one brand in a choice set can change the relative preference of one unaffected brand over another. Their explanation of the effect suggests that the change can make one brand more central to the category by changing perceptions of brand differences.

Barsalou (1985) presents an extended notion of typicality that lends credence to this linkage between perceptions of brands as members of a category and brand preference. For natural descriptive categories (e.g., the category of birds or of shapes), the "goodness" of an item as a member of a category appears to be related to its proximity to the "center" of the category. Thus, as the value of one of its attributes approaches a central value (e.g., the median or mode) of that attribute amongst all members of the category, the item's typicality increases.

Barsalou contrasts this case with that of goal-derived categories, such as "foods to eat on a diet" or "things to give as birthday I resents." Barsalou found that measures of central tendency are poor predictors of typicality in these categories. Indeed, a goal-derived category often has little natural organization so that central tendency has little rr caning. In addition, one is often interested in items that have extreme values of key attributes (e.g., diet foods have few rather than an average number of calories).

Consistent with these differences, Barsalou found that two good predictors of the goodness of an item as a member of a goal-derived category were 1) the relative proximity of an item's features to category ideals, and 2) the frequency with which an item had been experienced as a member of the category. In the absence of any natural structure, one must rely on one's own experiences to guide categorization judgments.

A product class might be viewed as a goal-derived category, the goal being the satisfaction of some need. The preceding analysis then suggests that the proximity of a brand's attributes to category ideals and the level of a consumers experience with a brand should predict the "goodness" of a brand as a category member. However, this category judgment is tantamount to one of relative preference for brands in a given usage situation. Thus, brand preference and category structure are closely linked.

Once again, the genesis of category structure lies in the consumer's initial product experiences. Because exposure is often limited to a small set of brands, the frequency with which they are encountered as category members will be relatively high. In addition, when consumer perceptions of product quality are poorly formed, category ideals may be defined through product use.

Deighton (1984) and Hoch and Ha (1986) provide data suggestive of such a process. In their studies, advertising messages were conceptualized as hypotheses, which were tested by consumers through product use. The advertising served to interpret consumer experience with the product, which in turn validated the hypotheses. Deighton found that-quality perceptions for automobiles were enhanced far more by advertising messages and objective data than by either type of information alone. Advertising alone was evidently untrustworthy; evidence alone was too difficult to interpret. By offering an interpretation of the evidence, the advertising had a major influence on brand evaluation. Hoch and Ha used direct inspection of the product as evidence, and also manipulated the timing of advertising presentation. Advertising had an effect only when presented before the experience, and had an effect only when there was ambiguity associated with the experience.

These studies considered categories for which global perceptions were already defined. What, then, of a category novel to the consumer?

When Vaseline petroleum jelly was introduced, it stressed the importance of purity. Satisfactory use of the product as a dressing for wounds might be perceived as vindication of that claim, making purity an important ideal of the product category. Thus, we suggest that initial product learning will generalize to the category, giving rise to preference-related product expectations.

Because of these expectations, product classes typically develop a "natural" structure or stereotype as well as a purely goal-oriented functional organization. Certain product features, both physical and functional, come to be associated with the category. For example, a vast array of features might distinguish soft drinks. However, common to most people's vision of a soft drink would be features like carbonated, sweet, flavored, and to be drunk cold. Indeed, a firm violating these expectations might have difficulty persuading consumers to categorize its product in this class.

Thus, it seems likely that product categories include a great deal of redundant information. While a variety of physical products might serve a given function, only certain ones will be viewed as legitimate. A mylar weather balloon might make an excellent insulator for a quilt, but fits the image of a quilt rather poorly. Thus, not only function but physical form restricts the types of allowed variation within the category.

Moreover, once this category stereotype is formed, research in social cognition suggests that it is extremely resistant to change (Fiske and Taylor 1984). When encountering product information incongruent with the stereotype, effort appears to focus on finding an interpretation of the data that reduces the incongruity. Later recall appears to be dominated by information congruent with the stereotype. As a result, the category learning process suggests that a consumer will develop a persistent image of the product category built around early product experiences.

In sum, the dynamics of category learning suggest that category ideals as well as defining features of the category are inferred through early product experiences. Therefore, the brands first experienced by the consumer will be perceived as highly representative of the category. And, because of the goal-derived nature of a product category, this representativeness will be closely linked to brand preference. Finally, because of the persistence of these category and brand perceptions, preferences will be long-lived.

CATEGORY LEARNING AND THE DYNAMICS OF COMPETITION

From a competitive standpoint, two extreme situations can be considered as archetypal sources of biased consumer learning. First, consider a market composed of a static set of brands with consumers entering and leaving, and assume one of the brands is dominant (i.e., has a large market share). In this case, an entering consumer's initial product experiences are likely to include the dominant brand. As a result, the consumer's category perceptions will reflect the characteristics of that brand.

Alternatively, consider an emerging market, into which a small set of brands enters. By default, consumer experience would be limited to those brands. For consumers who use the product, category perceptions will be based on the characteristics of that small set of brands. Thus, category structure and the rules of competition will reflect these brands. This inductive learning process makes later entrants, to the extent they differ from the pioneer, not only different but deviant and less preferred.

In either of these cases, we would expect to see consumer preferences biased in favor of the brands on which category perceptions are based. Thus, in the first case, we would expect consumer learning processes to help perpetuate the leading brand's market dominance. In the second case, decision processes would favor the early market entrants.

It is important to note, however, the limits of these effects. First, learning takes place at an individual level, so that if an early entrant is to realize its advantages, it must achieve a high penetration in the market so that the product stereotype is firmly implanted for a sizable proportion of consumers. Second, the advantage will have a limited life unless the linkage between specific product attributes and product performance is difficult to observe. Urban et al. (1986) have shown that product positioning can have an impact on the fortunes of later entrants, so that if a position can be shown to be objectively superior, the effects of entry order can be overcome.

To this point, we have considered a fairly general consumer-based explanation for the persistence of market dominance, however it was originally achieved. In the remainder of this paper, we will focus on a special case of the second scenarioCthat of a pioneering brand, where early consumer experiences are limited to a single brand.

Pioneers, Copycats, and Asymmetries in Preference Formation

In many markets, one brand achieves dominance early in the development of the product class. Seemingly immune to the vagaries of the marketplace, this brand then consistently outsells all competitors, an advantage that sometimes lasts for decades. This market share advantage has been dubbed a pioneer advantage.

Traditional explanations for this phenomenon focus OD some particular expertise of the firm. Some cite lower unit costs due to production experience or economies of scale. However, Robinson and Fornell (1985) found that the costs of pioneers are not lower than those of later entrants. Others cite superior foresight in positioning (Prescott and Visscher 1977) that allows the pioneer to preempt later entrants. However, pioneer advantages have often survived significant changes in consumer tastes, as well as repositioning by numerous brands.

The consumer learning process described earlier provides a natural explanation for long-lived pioneer advantage, particularly with respect to later entrants adopting a distinctive position in the market. If the product is too different, it would fall outside of the category. Inside the category, the later entrant faces the entrenched consumer category perceptions and preferences.

An alternative entry strategy for a later entrant would be to mimic the pioneer insofar as possible. These copycat or me-too brands are rarely successful in overtaking the pioneer. In this case, we must consider two cases quality that is observed vs. quality that is inferred. Schmalensee (1982) provides a formal analysis of the case of observable quality. The case of inferred quality represents a special case of biased category learning.

Learning Through Informative Trial

Schmalensee considers the case of a commodity whose quality is observable through a single trial or use. A brand either works or fails, and its performance is assumed to be unchanging over trials. Quality may vary across brands, which means that before use, there is some risk of failure associated with a brand. No other form of product differentiation is allowed.

In this situation, the pioneer has a significant advantage. Consumers who have tried the pioneer know it works. However, this is not the case for later entrants that have not been tried. Because of the risk of failure, the expected value of the later entrants is lowered. This forces the later entrants to charge a lower price in order to induce trial (and brand switching). As a result, the pioneer can charge a higher price than its competitors. If the later brand is successful, the pioneer advantage disappears. Thus, the life of the pioneer advantage in terms of market share can be limited (although this is not so for lifetime profits).

Empirically, however, one can cite numerous cases in which a pioneer advantage in terms of market share has persisted for decades, long after later entrants become well-known. Prime examples are Coca-Cola, Levi's jeans, Dial soap, and Maxwell House coffee. In all of these categories, most consumers are aware of competing brands, and many, if not the majority, have used them. Moreover, in all of these categories, some competitors are functionally indistinguishable from the pioneering brand. The risk of failure associated with later entering brands, therefore, is small. Yet, their market shares do not approach those of the category pioneers.

Learning By Interpreting Experience

Once again, as Carpenter and Nakamoto (1987) suggest, a potential explanation for this phenomenon lies in the pioneering brand's role in the category. In the examples just cited, the link between product attributes and quality would be initially poorly defined and difficult to observe. In this situation, we have argued that early entrants will define the relationship of the new category to previously established ones, specifying critical attributes of the category. In addition, they will frame competition among brands within the category by influencing relative brand perceptions, becoming highly representative of the category. In the case of the pioneering brand, the pioneer is uniquely representative or prototypical.

In terms of the previous analysis, the pioneer is close to the category ideals because the ideals are derived from experience with the pioneer. In addition, by default, the most frequently experienced brand would be the pioneer. As such, it provides a cognitive referent, and in extreme cases, pioneering brands become synonymous with the product category as in the cases of Levi's, Kleenex, and Jello.

Rosch (1977) notes that this prototypicality has a variety of effects, including faster identification of the object as a member of the category, and faster and more reliable recall of the object as a member of a category. In other words, the pioneer would be expected to enjoy a distinctiveness among brands in the category that would distinguish it from copycats. In a sense, simply because it has a different brand name, the copycat lacks this distinctiveness, is therefore less typical of the category, and less preferred. Consistent with this suggestion, Carpenter and Nakamoto (1987) found that the pioneer's advantage with respect to a copycat brand increased as the perceived similarity of the pioneer and the copycat increased. In addition, distinctively different second entrants fared better than copycats. In this study, the position adopted by the pioneer was manipulated, but the same pattern of results emerged regardless of the pioneer's location.

From a competitive standpoint, this result demonstrates that consumer preferences can be affected simply by entry order. As a result, it appears that pioneers, rather than selecting the most desirable position in the market, create that position. They define theirs as the superior product. The essence of the pioneer's advantage, then, lies in its effects on the framing of the consumer's perceptions of a product category and on the implicit context in which brand comparisons take place. The resulting perceptual and preference structures favor the pioneer, giving rise to its competitive advantage.

IMPLICATIONS

Our analysis of the bases for pioneer advantage has two central implications. First, consideration of the prerequisites for and limits of a pioneer advantage are suggestive of a theory of second entry. In a product category susceptible to framing by the pioneer, the problem for the second entrant will be to increase its own relative typicality and prominence. One approach would be to develop a distinctive product position, encouraging the consumer to form two subcategories, or in an aggregate sense, segmenting the market. Consistent with this, Carpenter and Nakamoto (1987) found that the more similar a group of later entrant brands taking a distinctive position were perceived, the weaker the pioneer advantage.

On the other hand, if the category is one in which product use will allow judgments of relative product superiority (e.g., the observably superior performance of Sony's TV sets), then direct competition with the leading brand may be feasible. Alternatively, if the pioneer has failed to obtain high market penetration by the time the second brand enters, the later entrant may usurp the pioneer advantage with a large marketing budget. This is a strategy employed by IBM with great success.

A second and more general issue raised by our analysis is an important dimension of marketing strategy beyond that embodied in the marketing concept, in which successful marketing strategies respond to consumer demands. In our view, consumer preferences are often malleable when product experience is very limited. As such, framing consumer perceptions and preferences may be an important objective of marketing strategy. The great sums of money devoted to advertising soft drinks to young teenagers and giving computers to schools may well prove effective in achieving a lasting competitive advantage for this reason.

CONCLUSION

We have argued that the learning environment faced by consumers often biases the order of product experience, and leads to biased perceptions of a product category and the brands in the category. As a result, brands that can attract early use by the consumer stand to gain a significant and long-lasting advantage, because those experiences will play an important role in the formation of preferences for all brands. In the special case of a market pioneer, early success produces a situation where the pioneer occupies a position that is difficult to imitate and where it becomes costly to attract consumers to a distinctive brand. Such a process has far-reaching implications for brand strategy and the analysis of market competition.

REFERENCES

Barsalou, Lawrence W. (1982), "Context-Independent and Context-Dependent Information in Concepts," Memory and Cognition, 10 (January), 82-93.

Barsalou, Lawrence W. (1985), "Ideals, Central Tendency, and Frequency of Instantiation as Determinants of Graded Structure in Categories," Journal of Experimental Psychology: Learning, Memory, and Cognition, 11 (October), 629-655.

Bettman, James R. (1986), "Consumer Psychology," Annual Review of Psychology, 37, 257-289.

Biehal, Gabriel and Dipankar Chakravarti (1982), "Information Presentation Format and Learning Goals as Determinants of Consumers' Memory-Retrieval and Choice Processes," Journal of Consumer Research, 8 (March), 431-441.

Biehal, Gabriel and Dipankar Chakravarti (1983), "Information Accessibility as a Moderator of Consumer Choice," Journal of Consumer Research, 10 (June), 1-14.

Carpenter, Gregory S. and Kent Nakamoto (1987), "Consumer Preference Formation and Pioneering Advantage," Working Paper, Columbia University.

Deighton, John (1984), "The Interaction of Advertising and Evidence," Journal of Consumer Research, 11 (December), 763-770.

Elio, Renee and John R. Anderson (1984), "The Effects of Information Order and Learning Mode on Schema Abstraction," Memory and Cognition, 12 (January), 20-30.

Fiske, Susan and Taylor, Shelley E. (1984), Social Cognition, Reading: MA, Addison-Wesley Publishing Co.

Hoch, Stephen J. and Young-Won Ha (1986), "Consumer Learning: Advertising and the Ambiguity of Product Experience," Journal of Consumer Research, 13 (September), 221-233.

Holland, John H., Keith J. Holyoak, Richard E. Nisbett, and Paul R. Thagard (1986), Induction, Cambridge, MA: MIT Press.

Huber, Joel, John W. Payne, and Christopher Puto (1982), "Adding Asymmetrically Dominated Alternatives: Violations of Regularity and the Similarity Hypothesis," Journal of Consumer Research, 9 (June), 90-98.

Johnson, Eric J. and J. Edward Russo (1984), "Product Familiarity and Learning New Information," Journal of Consumer Research, 11 (June), 542-550.

Prescott, Edward C. and Michael Visscher (1977), "Sequential Location Among Firms with Foresight," Bell Journal of Economics, 8 (Autumn), 378-393.

Rips, L.J., E.J. Shoben, and E.E. Smith (1973), "Semantic Distance and the Verification of Semantic Relations," Journal of Verbal Learning and Verbal Memory, 12, 1-20.

Robinson, William T. and Claes Fornell (1985), "Sources of Market Pioneer Advantages in Consumer Goods Industries," Journal of Marketing Research, 22 (August), 305-318.

Rosch, Eleanor (1977), "Principles of Categorization," in Cognition and Categorization, E. Rosch and B.B. Lloyd, eds., Hillsdale, NJ: Lawrence Erlbaum Assoc.

Schmalensee, Richard (1982), "Product Differentiation Advantages of Pioneering Brands," American Economic Review, 27, 349-365.

Sujan, Mita (1985), "Consumer Knowledge: Effects on Evaluation Strategies Mediating Consumer Judgments," Journal of Consumer Research, 12 (June), 31-46.

Tversky, Amos (1977), "Features of Similarity," Psychological Review, 84 (July), 327-352.

Urban, Glen, Theresa Carter, Steve Gaskin, and Zofia Mucha (1986), "Market Share Rewards to Pioneering Brands: An Empirical Analysis and Strategic Implications," Management Science, 32 (June), 645-659.

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