Marketing Communications

David W Stewart & Michael A Kamins. Handbook of Marketing. Editor: Barton A Weitz & Robin Wensley. Sage Publication. 2002.

Introduction

Effective marketing and business strategies require obtaining the attention, informing and influencing relevant constituencies of the firm. Potential customers do not buy products or services that they are unaware of. Communication of the reasons a particular product should be purchased, e.g., the benefits associated with the product, and the ways in which a given brand within a product category is different or superior to competitive products, e.g., how the brand is differentiated, are fundamental to marketing success. Kotler (2000: 550) observes, ‘the question is not whether to communicate, but rather what to say, to whom and how often.’

Research on marketing communications has a long and rich history. It has roots in early research on verbal learning in psychology (Ebbinghaus, 1885). The first systematic treatment of the role and influence of marketing communications was published by psychologist Walter Dill Scott (1908). Subsequently, the work of social psychologists such as McGuire (1969), Hovland et al. (1953), Lewin (1948), Merton (1946), Katz and Lazarsfeld (1955), and Klapper (1960), among others, played an important role in the development of both empirical research and theory focused on mass communications in general, and marketing communications more specifically. By the 1950s a rich tradition of research and theory development on marketing communications had begun to develop within the marketing discipline, but there continues to be a strong link between work on marketing communications and research in psychology on social influence, attitude change, and communications processes. Contemporary theories of marketing communications rest on such psychological foundations as the functional theory of attitudes (Fazio, 1990; Katz, 1960), the theory of reasoned action (Fishbein & Ajzen, 1975), the elaboration likelihood model (Petty & Cacioppo, 1981), social cognition (Bandura, 1994; Fiske & Taylor, 1992) and theories of goal-derived categorization (Barsalou, 1991). At the same time, research in communications (e.g., Bryant & Zillmann, 1994; Rogers, 1983) and other fields have offered insights into the influence of marketing communication. Finally, important work by practitioners and academics working with commercial data (e.g., Haley, 1985; Jones, 1995; Lodish et al., 1995; McDonald, 1996; Schwerin & Newell, 1981) have provided useful and important evidence regarding the effects of marketing communications.

The purpose of the present chapter is to review the generally accepted body of knowledge about the influence and effects of marketing communications. In addition, the chapter will examine factors that may mediate or moderate the influence of marketing communications. Page constraints prevent a detailed discussion of the rich theoretical foundations on which the body of knowledge rests. The focus of the chapter is the influence of marketing communications at the market or market segment level. Thus, the considerable literature on individual differences in response to marketing communication and on individual-level processes associated with marketing communication will not receive detailed treatment.

The advent of the Internet, other interactive media, and sophisticated customer relationship management systems, clearly create new opportunities for one-to-one targeting of consumers (Pavlou & Stewart, 2000). However, these opportunities are still largely untapped and the literature on such targeting is still in its infancy. To a large degree, the relatively recent origin of these technologies confounds the unique effects of this type of communication with the effects of the adoption of new technology by both consumers and marketers. For these reasons this chapter will not provide a comprehensive review of the role of interactive media and one-to-one marketing. A comprehensive review of interactive media can be found in Stewart et al. (2002), and Internet marketing is discussed elsewhere in this volume. While it is likely various new media will open new areas of research, it is also likely that much of the considerable theory and research related to more traditional forms of marketing communication will ultimately be found to be generalizable to various new media once the effects of technology adoption are sorted out. Thus, the goal of the present chapter is to provide a summary of a vast literature on marketing communication that can serve as an introduction to the rich empirical research and theory in the field.

The Role of Marketing Communication

The role of marketing communication is to convey an appropriate message to one or more of the firm’s various constituencies. These constituencies include present and future customers, shareholders, competitors, employees, distributors, retailers, regulators, government officials, and the public at large. The messages, media, and objects of communication vary depending on the constituency involved. The reasons a firm may seek to convey a message also vary. It is not always the case that a firm uses marketing communications to sell its products or services. Firms may use marketing communications to create awareness of products, to provide specific information, or to establish or reinforce associations, attitudes, or preferences.

Lavidge and Steiner (1961) introduced a model termed the ‘Hierarchy of Effects’ to describe the steps or stages through which consumers typically pass when contemplating the purchase of a product. This model, which has its foundations in the earlier work on the social psychology of communication, posits that consumers move from being unaware to aware, to having knowledge, then to liking and preference, and finally to conviction and purchase. Although numerous variations of the hierarchy have subsequently been suggested (Moriarity, 1983) and there has been debate about the order in which various stages may occur (Ray, 1973), the general model has withstood the test of time as a means for conceptualizing the roles of marketing communication. The hierarchy of effects suggests that the goal of marketing communication is to move consumers through these various stages toward an ultimate purchase. Looked at more broadly, however, the model suggests that marketing communication may play a role in three processes: cognition (awareness and knowledge); affect (liking, preference, and conviction); conation or behavior (shopping, purchase). These underlying processes are linked to such marketing objectives as building product awareness and identity, brand building, and product choice, among others.

The relative importance of the various potential goals of marketing communication vary as a function of different moderating factors, including the product’s stage in the life cycle, the competitive position of the firm, the intensity of competitive activity, and the degree of loyalty or brand switching in the product category. For example, creating awareness is much more important early in the life cycle of a product than it is for a mature product with a long and well-known history. Similarly, providing information about the positive dimensions of a product, to differentiate it from competitors’ products, will be more important in the face of competition than in the absence of significant competition. These mediating factors will be considered in greater detail later in this chapter.

Types of Marketing Communication

Although there are many variations of marketing communication, there are five well-recognized types of communication over which marketers exercise some control: advertising, sales promotion, publicity and public relations, personal selling, and interactive or direct response marketing. To these five types of marketing communication under the control of the marketer must be added word-of-mouth communication—a powerful form of communication in markets, but one over which marketers have decidedly less control.

Advertising involves non-personal forms of communication offered through paid media under clear sponsorship. Major goals of advertising include building primary demand (product category sales) and selective demand (individual brand sales). Advertising may take a variety of forms, and different forms of advertising may produce different effects in the market and be differentially effective depending on various characteristics of the market, the advertised brand, and competitors’ brands. Advertising has evolved over time as new media have been developed. Thus, advertising may be found in a variety of different media ranging from signs and outdoor billboards, to newspapers and magazines, to broadcast media. Each of these types of media impose different constraints on the kind of advertising message that can be effectively conveyed and the types of responses and outcomes that can be realistically expected in the market. Historically, much of advertising has tended to involve one-way interaction between the advertiser and the customer. The advent of the Internet and other interactive media has recently created the opportunity for interactive advertising, which has added to the types of response and outcomes that might be expected of advertising (Pavlou & Stewart, 2000).

Although advertising has frequently been depicted in the popular press as a powerful force (cf Galbraith, 1967; Packard, 1957), it has generally been shown to have relatively modest effects on customers (Lodish et al., 1995; Stewart, 1992). Such modest effects do not mean that advertising is unimportant, but they do suggest that not all advertising is effective. Even advertising that is effective, in the sense that it achieves its objectives, has less effect than other factors such as customers’ underlying needs and preferences, the quality of the product or service offering, pricing, and availability.

Sales promotion represents an eclectic collection of various promotional incentives designed to stimulate volume or speed of purchase (Blattberg & Neslin, 1990). Such incentives include sampling, coupons, premiums or gifts, and price deals, among others. As a percentage of the total promotional budget, sales promotion expenditures exceed advertising (Kotler, 2000). Because of the frequency of sales promotion techniques it could be argued that consumers have become promotion elastic and that larger or more lucrative incentives will be required to encourage the goal of new trial or increased usage among current users. It is certainly the case that the primary message communicated by most sales promotions is related to the importance of price as a determinant of purchase. A study by Krishna, Currim and Shoemaker (1991) regarding the deal price component of sales promotions revealed that consumers are generally knowledgeable regarding when a particular brand has been on sale. In addition, the authors observed that the expected frequency of dealing is positively correlated with perceived past-deal frequency. Such knowledge may impact the consumer’s image of the brand and result in the consumer only buying the brand when it is on sale. Sales promotion can play an important and complementary role in generating trial for new products and brands, however. Although sales promotion clearly plays a role in marketing communication, this chapter will not specifically deal with sales promotion.

Publicity and public relations involves the use of non-paid media to reach relevant audiences. Firms use press releases, media events, press kits, product announcements and a variety of other tools that are directed at the media (journalists, reporters, etc.) rather than customers. The intent of public relations activities is to obtain coverage of the firm and its products in stories and editorials that appear in the media. Since such stories are not paid for (at least not directly) and are attributed to a neutral party, they tend to be much lower in cost and have greater credibility than advertising. On the other hand, the nature of the message and the size and character of the audience reached through publicity is not under the control of the marketer.

Personal selling involves direct interpersonal communication between a representative of the firm and the potential customer. Whereas advertising has traditionally carried on a monolog with the target market, personal selling is dyadic in nature and hence inherently offers many advantages as a communication device (Williams et al., 1990). These advantages are critical since a personal selling approach is often required to complete the sale. One especially important advantage of personal selling is that the marketer’s message can be specifically tailored to the consumer. Weitz et al. (1986) note that salespeople may engage in adaptive selling based on their knowledge of customer types and sales strategies. A second advantage of personal selling is that salespeople can develop relationships with customers that may increase purchase probability and brand loyalty (Farber & Wycoff, 1992). Personal selling is generally more expensive than advertising, and the number of individuals with which a salesperson can have contact within a given period is severely constrained. Although many of the principles and concepts associated with communication in general apply in a personal selling context, the interpersonal and interactive nature of personal selling adds dimensions to the communication process that are beyond the scope of the present chapter. Chapter 10 of this volume focuses more specifically on communication in a personal selling context.

Direct response and interactive marketing involve the use of non-personal media to conduct a dialog with consumers. The role of such marketing communications is to facilitate reciprocal response between customers and marketers. It includes electronic shopping, voice mail, e-mail, facsimile, telemarketing, catalogs and TV shopping. The Direct Marketing Association defines an interactive marketing system as one that uses one or more advertising media to affect a measurable response and/or transaction at any location. Day and Montgomery (1999) observe that the adaptive company of the future will utilize interactive strategies to become more market driven. That is, the company will have ‘the ability to address an individual and then remember the response of that individual. With each successive interaction, the firm learns more and uses this information from customers to personalize further communications in ways that take into account that unique response and the value proposition.’

Word-of-mouth communication (WOM) refers to communication among and between individual customers. Such communication is very common in many markets, and the role of customer-to-customer communication cannot be underestimated. Customer-initiated communication tends to have very high credibility and may have wide-ranging effects: it may influence where customers shop, what information customers seek, what customers buy, and how customers evaluate a purchase after the sale. Although marketers often attempt to facilitate word-of-mouth communication, it is not directly under the marketers’ control. There is evidence that word-of-mouth communication may be growing in importance as interactive media, such as the Internet, make it easier for customers to communicate with one another (Pavlou & Stewart, 2000).

Integrated Marketing Communications. In the past decade there has been an increasing recognition by both marketing scholars and practitioners of marketing communications that the various forms of marketing communication can and should work together to reinforce the goals and objectives of the marketer. This recognition has produced a focus on what has come to be called Integrated Marketing Communications (IMC) (Thorson & Moore, 1996). IMC explicitly recognizes the fact that consumers are exposed to information from many different sources. This exposure, in turn, results in the need for marketers to coordinate communications across media and consider how various forms of communication will reinforce and complement one another. In addition, various new media, such as the Internet, tend to blur the elements of the marketing mix (Stewart et al., 1996). Thus, the same medium may be a communications channel, a distribution channel, and, in some cases, even the product itself. This blurring of the boundaries of communication channels and of the larger marketing mix suggest a need to understand the context in which an individual medium or message may be embedded. It is certainly the case that marketers need to understand how all the elements of the communication and marketing mix will work together to produce a particular outcome.

Effects of Marketing Communication

The questions as to whether marketing communication is effective and what its effects are, are, on their face, simple questions. Yet, they are not so simple. For example, the question of effectiveness leads to issues regarding how success or failure is measured, and what is the time frame to be considered? A particular communication may be readily recalled but only because it is so irritating and obnoxious that it persuades consumers not to buy the product. By one measure, awareness, the communication ‘worked’; by another, the number of persons persuaded to buy the product, it is a dismal failure. One cannot make a determination of whether communication has had an effect or has been successful without first specifying the purpose of the communication. Specification of purpose, in turn, requires an understanding of how communication influences the customer. Recognition of this fact led marketers to develop various approaches for identifying communications objectives and for conceptualizing the relationships among these objectives and their in-market measurement. The ubiquitous hierarchy-of-effects model, in its many forms (Moriarity, 1983; Vaughn, 1980), is one example of such an approach. Much of the research and theory related to the effects of marketing communications, at least communications involving impersonal media, has focused on advertising. For this reason, much of the discussion of the effects of marketing communication in the remainder of this chapter will focus on advertising effects.

Advertising and Primary Demand

A reading of marketing textbooks suggests that it is almost axiomatic that advertising increases product category sales. A frequent criticism of marketing communication—that it creates demand for products that have little utility—is also predicated on the assumption that one effect of communication is the creation of generic demand (Galbraith, 1967). This assumption rests on what appear to be reasonable premises. There is little doubt that various types of marketing communication are very efficient means for creating awareness of new products. Communication may also be a very useful and efficient means for stimulating the trial of a new product when the product itself is a new-to-the-world product. Thus, it is not difficult to identify a link between advertising, or marketing communication more generally, and product purchase early in the life cycle of a product. This link does not establish that communication is the primary determinant of demand, however, even for new-to-the-world categories. More importantly, there is little evidence that advertising, or other types of marketing communication, can increase the consumption of products within a mature product category (Ehrenberg, 1983; Lambin, 1976). The evidence that does exist suggests that such effects are small and short-lived. One of the most comprehensive studies of the effects of advertising on sales was carried out by Lambin (1976). He investigated the effects of advertising for 16 different product categories in as many as 9 different countries over a 10-year period. With respect to the question of advertising’s effects on aggregate category demand, Lambin concluded (1976: 136): ‘Limited empirical support is given to the view that advertising increases primary demand.’ In the four product categories where advertising did have a significant effect on primary demand, all four product classes were early in the product life cycle, and

The economic power of advertising ‘per se’ has been overstated by advertising critics and apologists. Socioeconomic forces are more important, and advertising has a limited capacity to stimulate total market growth. (Lambin, 1976: 136)

It is useful to note that some advertising campaigns that appear directed at increasing sales for a product category as a whole are, in fact, competitive advertisements in a broader context of competition. Examples of such campaigns include the California Raisins campaign (the dancing raisins) and the V-8 Vegetable Juice campaign (‘Wow! I could have had a V-8.’). This type of marketing communication tends to be rare, however, since many products do not have close indirect substitutes. Other exceptions to these general findings are cases where marketing communication suggests new uses for an established product (the well-known Arm and Hammer Baking Soda campaign is an example), and where a product with a low level of top-of-mind awareness is made more salient to consumers (advertising campaigns for http://dot.com companies such as EBAY and IVILLAGE are recent examples). In all of these instances the product on which communications focuses is more similar to a new product than an established product. The suggestion of new uses for a product is essentially the same as developing a new market. In the case of a product or brand with low awareness and usage, the influence of marketing communication is closely akin to the development of a new market.

Several of the empirical studies that suggest communication has an effect on a primary demand have, in fact, been carried out in markets for various agricultural products, where low levels of awareness or indirect substitutes exist (Henderson & Brown, 1961; Henderson et al., 1961; Hoofnagle, 1963). Thus, these studies are consistent with the proposition that marketing communication has an effect on aggregate demand only under a narrow set of circumstances.

Empirical support for the proposition that marketing communication creates generic demand largely rests on studies that have shown correlations between growth of aggregate sales in a market and increases in total advertising expenditures (see, for example, the study by Taylor and Weiserbs, 1972). These studies seldom consider the influence of market conditions in interpreting results, however. For example, in markets that are growing, it is likely that advertising expenditures will increase in response to the growing market. The growing market, in turn, will result in increases in sales. When advertising budgets are based on a percent-of-sales rule, which remains the most common budgeting practice, it will always be the case that rising demand will be associated with rising expenditures on advertising. It would be incorrect to conclude that the increase in expenditures was the cause of the increase in sales in this instance. Rather, market growth drives the increase in both sales and advertising.1

Market growth may be attributable to factors other than a product being early in its life cycle. Demographic changes in a population often increase demand for whole product classes. As this demand is recognized by marketing organizations, expenditures on marketing communications are also likely to grow. This will produce a correlation between expenditures on advertising and growth in market demand. Such expenditures did not create the demand, however. The increase in expenditures is a response to the growth in demand.

Among the most frequently cited studies of the purported effect of marketing communications on aggregate demand is that of Taylor and Weiserbs (1972). This study, which concluded that advertising does influence the aggregate consumption function, has been criticized for its methodological and conceptual flaws (see Lambin, 1976: 137–8). Among these flaws is a failure to consider marketplace factors other than advertising that may influence demand: the Taylor and Weiserbs study (1972) did not account for new product introductions during the period of the study.

Another frequently cited study (Roberts & Samuelson, 1988) examined data on the United States’ cigarette market during 1971–1982. This study concluded that cigarette advertising expenditures had a significant positive impact on total demand in the category, while individual market shares of competing firms depended more on the number of brands offered. This study failed to consider an important determinant of demand during the period of study, however. The period 1971–1982 was a period in which large numbers of consumers attained an age when smoking often begins. The baby-boom generation entered adolescence and young adulthood during this period. In fact, the final wave of the baby-boom generation, those born in 1964, would have been 18 years of age in 1982. Indeed, a subsequent study of the cigarette market (Pollay et al., 1996) concluded that the primary effect of advertising in this market was to influence brand choice.

The influence of demographic trends can be observed in a variety of markets and the effects of marketing communications on primary demand can only be estimated correctly when such trends are also considered. For example, during the same period of the Roberts and Samuelson study, 1971–1982, there was an explosion of demand in the soft drink category. This huge increase in demand was also accompanied by large increases in expenditures on marketing communications by soft drink marketers. Such increases in marketing communications expenditures was a response to the increased demand that was created as a large segment of the population moved into the years of high soft drink consumption.

There are, of course, determinants of demand other than demographics. There are cultural, social, and familial influences. People often want things that society suggests are important; they often want what others have. Most products have utility of one type or another; they make life easier, they move us from place to place, they cover our bodies, they provide excitement or relaxation, they make us feel better about ourselves, they provide nourishment, or they just taste or feel good. Some of this utility is inherent in the product itself; some of it may be a kind of learned utility. Such learned utility is a function of the culture, social relationships, and personal and observed experiences. Insofar as marketing communication reflects the values of a culture, it may serve to reinforce these values.

Increases and decreases in the demand for products are generally the result of broad changes in a society. These changes may be demographic; they may be economic; they may be technological; or they may be related to the fabric of values that define a society. Sometimes, these changes may be the result of new information. For example, more consumers are concerned about the amount of cholesterol in their diets today than 20 years ago. This concern has changed the eating habits of many consumers. It has also resulted in changes both in food products and in what is communicated about food products. That is, attributes of the product linked to cholesterol and, more broadly, to health concerns, become more salient to the consumer (e.g., fat and vitamin content) and hence are emphasized more frequently and prominently in marketing communications.

Both the empirical evidence and logical deduction offer compelling evidence that marketing communication does not create demand; it is a response to demand. People buy things because they want them, not because advertising somehow compels them to purchase. When the influence of primary drivers of demand, like demographic changes, broad societal changes, and the effects of other marketing actions, such as lower price, are controlled, there are no studies that demonstrate that marketing communication creates demand for established products.

In light of the evidence to the contrary, why does the belief that marketing communication creates demand persist? History and the human tendency to confound correlation with causation are the likely culprits. Much of the history of advertising was written during a period of very rapid growth in markets. As Blair et al. (1987) note, the period of history that began with the end of World War II and ended in the late 1970s was a time of enormous growth in the domestic market of the United States. This growth was driven by both a growth in population and a growth in the affluence of the population. This growth offered opportunities for business firms who responded by introducing a plethora of new products. At the same time, there was rapid growth in the media available for communicating with consumers about goods and services. One result of these broad societal changes was increased demand for a wide array of products and a concomitant increase in expenditures on marketing communication. Marketing communication during this period did speed the rapidity of the product adoption process by facilitating consumer awareness. It is not too difficult a leap to conclude that marketing communication was at least partially responsible for the growth in aggregate demand. Galbraith (1967) popularized this view of advertising in his book, The Affluent Society. Yet, this view was fundamentally incorrect.

Advertising and Selective Demand

If advertising or other types of marketing communications do not create aggregate product demand, at least in relatively mature product categories, it may, nonetheless, increase demand for the advertised brand. A common assumption about advertising is that its effect should always be manifest in an increase in demand. This increase in demand may be the result of one or both of two events: 1) communication may induce users of competing brands to switch brands; or 2) communication may induce users of a particular brand to buy more of the brand. In order to examine the influence of advertising on these two events, it is necessary to distinguish between two classes of consumers.

It has long been recognized that product categories differ appreciably in terms of the amount of switching among brands that takes place over time (Adtel, 1974; Carpenter & Lehmann, 1985; Colombo & Morrison, 1987; Ehrenberg, 1972; Frank & Massy, 1967; Jeuland, 1979; Raj, 1982; Schwerin & Newell, 1981; Stewart & Furse, 1986; Tellis, 1988). Some product categories are characterized by very high rates of brand switching among consumers, while others are characterized by relatively high brand loyalty and very low rates of brand switching among product users. These differences in switching rates are the result of a multitude of factors that will not be reviewed here. It is sufficient to note that these differences tend to be very stable over time (Ehrenberg, 1972). Further, there is evidence that within certain product categories, individual brands may also have unique switching rates.

The differences in brand switching rates across product categories are related to the mix of consumers who are loyal to a given brand or brands and those consumers who are not brand loyal. Certain product categories, for example cigarettes and laundry detergent, tend to exhibit very low rates of brand switching and high loyalty (as measured by repeat purchase rates) to individual brands. These types of product categories may have many or few product alternatives, but there is little switching from one product to another, i.e., the rate of switching does not appear to be related to the number of alternatives available. On the other hand, some product categories, e.g., paper towels and ready-to-eat cereals, tend to exhibit very high levels of brand switching. Thus, product categories differ with respect to the relative mix of brand loyal consumers and brand switchers.

Empirical research has consistently shown the existence of these two classes of consumers and has clearly demonstrated that marketing communication has differential effects on these two classes (Belch, 1981, 1982; Cacioppo & Petty, 1985; Carpenter & Lehmann, 1985; Eskin & Baron, 1977; Information Resources, Inc., 1989; Jeuland, 1979; Naples, 1979; Pechmann & Stewart, 1990; Raj, 1982; Schwerin & Newell, 1981; Simon & Arndt, 1980; Stephens & Warren, 1984; Tellis, 1988). Unfortunately, much of the empirical research on marketing communication has examined the effects of communication at the aggregate market level, rather than examine the effects within these two classes of consumers (Columbo & Morrison, 1987). This has frequently resulted in contradictory findings and misleading conclusions about the effects of marketing communication.

Advertising Effects on Loyals

The effect of advertising on loyals is obviously not to persuade them to buy the advertised brand, since they already do so. Rather, if there is an effect at all it must be to reinforce loyalty to the brand or persuade the consumer to buy more of the brand. Thus, in markets with very large numbers of loyal consumers relative to switchers, it would be unreasonable to expect that marketing communication should produce incremental sales unless the loyal users could be induced to buy more product.

There is some empirical evidence that suggests that advertising may induce loyal consumers to purchase larger quantities of their preferred brand, at least in the short run (Raj, 1982; Tellis, 1988). Such volume effects may be explained, at least in part, by a lessened tendency on the part of the consumer to purchase competitive brands (Carpenter & Lehmann, 1985; Raj, 1982), however. It is not at all clear that these effects are the result of an increase in the long-term demand for either the specific brand or the generic product. In other words, the effect of marketing communication on loyals appears to be to discourage switching (reinforce loyalty). Such reinforcement effects may or may not manifest themselves in increases in purchase volume, depending on the propensity of consumers to switch among brands and the frequency of purchase. Among the most loyal of consumers, those who purchase a single preferred brand, there would be no manifest effect of advertising even when the communication serves to reinforce loyalty.

An important question, with both theoretical and empirical implications, is the process by which marketing communication has its effect on loyal consumers. A substantial body of research speaks directly to this issue. Empirical studies that have examined the influence of advertising on loyals have consistently found that these consumers are more likely to pay attention to and remember information about products (brands) that they use, and are less likely to pay attention or remember information about products (brands) they do not use (Belch, 1981; Craig et al., 1976; Pechmann & Stewart, 1990; Politz Media Studies, 1960; Tolley, 1994). Such findings are quite consistent with theories of attention (Broadbent, 1977; Greenwald & Leavitt, 1984; Krugman, 1988) and selective exposure (Atkin, 1985; Axsom & Cooper, 1981; Baugh, 1982; Condry, 1989; Fenigstein & Heyduk, 1985; Gunter, 1985; Ratneshwar et al., 1989; Zillmann & Bryant, 1985), and with theories of cognitive dissonance (Akerlof & Dickens, 1982; Calder, 1981; Cotton, 1985). The former theories simply assert that people are more likely to attend to things that are more relevant to their needs and consistent with their own beliefs. The latter theory asserts that consumers often seek justification for making a decision or behaving in a particular way. Thus, Ehrlich et al. (1957) found new car buyers more attentive to automobile advertising after the purchase than before.

Studies in contexts other than marketing also support the view that people selectively attend to information (Axsom & Cooper, 1981; Brehm, 1956; Davis & Jones, 1960; Glass, 1964; Knox & Inkster, 1968). Thus, it is not surprising to find studies that reveal a high correlation between advertising recognition and product use (Chapman & Fitzgerald, 1982; Goldstein et al., 1987; McCarthy, 1986). Such findings suggest that attention to advertising is a natural consequence of using a particular product or brand.

The phenomena of selective exposure and cognitive dissonance explain why it is so hard to change brand loyal behavior once it is established. Consumers are simply less likely to attend to marketing communications for a brand they do not currently purchase. These phenomena also explain why repetition and creativity are such important elements in advertising. Attracting the attention of the non-user of a brand or product with a short message that is surrounded by enormous clutter is very difficult. Creative content or frequent repetitions of a product message represent the only avenues for attracting attention from an otherwise disinterested consumer. Attention and memory are not the only factors that are influenced by loyalty, however.

Modern attitude theory holds that individuals are not mere passive recipients of messages (Petty & Cacioppo, 1986). Rather, they actively screen and operate on information as it becomes available (assuming they choose to attend to it at all). They tend to seek information that reinforces existing beliefs, a phenomenon referred to as perceptual vigilance (Runyon & Stewart, 1987), and selectively ignore information that is counter to existing beliefs, a phenomenon known as perceptual defense (Runyon & Stewart, 1987). Individuals also elaborate on messages. They add arguments of their own, they counter-argue, and they have thoughts about the veracity of the source (Belch, 1981; Wright, 1973). These thoughts, not the message that gave rise to them, are what form or change attitudes. The numbers of these thoughts and the strength of resultant attitudes are a function of how involved the consumer is in the processing of a message (Petty & Cacioppo, 1986). When there is the potential for serious consequences associated with a decision (the decision is one of high involvement), individuals tend to do more thinking and form attitudes that are more resistant to change. When the potential for consequences is small (the decision is one of low involvement), the number of thoughts tends to be small and the attitudes that are formed tend to be weak (Kamins & Assael, 1989; Wright, 1974). There is little incentive in these types of low involvement situations for consumers to revisit decisions and attitudes once they are formed, however.

Most decisions about brands of frequently purchased consumer goods are low involvement decisions. Among loyal consumers of these products there are few incentives for switching brands. As a consequence, marketing communication related to competitive brands is unlikely to attract the attention of the loyal consumer, and this consumer is likely to have few thoughts in response to any given communication that does happen to obtain his or her attention. This is not a scenario in which marketing communication is likely to have a strong effect.

For loyal consumers, marketing communication simply serves to reinforce decisions of little consequence. Recognition of this role of advertising led Andrew Ehrenberg (1983, 1988) to propose the Awareness-Trial-Reinforcement (ATR) model of advertising. The ATR model suggests that advertising is a weak force rather than a strong, persuasive influence. It asserts that advertising may influence the consumer in one of several ways; advertising may create awareness, induce trial, or reinforce product purchase. It is the last stage, where the role of advertising is to reinforce feelings of satisfaction after purchase or to re-awaken awareness, that Ehrenberg emphasizes. He argues that the role of repetitive communication for mature products is predominantly defensive—to reinforce already developed repeat buying habits—since most buying behavior involves the repeat purchase of familiar products. Since the consumer is more likely to attend to communications for the brands he or she is already buying, repetition enables the purchase habit for a particular brand to continue to operate in the face of competitive advertising and other marketing programs.

Ehrenberg mounts a strong argument in defense of his position. He notes that there is little direct evidence that advertising for established brands operates by any other means, and such evidence as does exist is negative (Achenbaum, 1972) or only partially supportive (Barnes, 1971; McDonald, 1971; Palda, 1966). He argues that other models of advertising response do not explain how small- to medium-brands retain share in the face of vast amounts of advertising by brand leaders or why cuts in advertising seldom result in large losses of sales over the short run. Ehrenberg also notes that advertisements frequently provide little product information, a fact that would be more consistent with advertising being suggestive or reinforcing rather than strongly persuasive. Because the buyers of frequently purchased goods are not ignorant of them, Ehrenberg reasons that the aim of marketing communications is to inform the experienced customer that brand ‘X’ is as good as other brands.

Thus, ATR suggests that advertising is a price that firms pay to stay in the market by defending or maintaining their share of a market that is well established. Advertising might also create new sales for a given brand (but not for the category as a whole) by reawakening consumers’ awareness and interest in it and by stimulating trial, but ultimately these consumers will be lost if they do not like the product and their new purchase behavior is not reinforced. Indeed, this is a particularly likely occurrence among brand switchers, whose response to marketing communication will be discussed below. The ATR model suggests that reinforcement reduces cognitive dissonance, increases brand satisfaction, and rekindles brand awareness. Communication has such a reinforcing effect because it predisposes the consumer to notice certain aspects of the brand, while ignoring others.

The marketing literature has for some time acknowledged that one effect of advertising is to provide a rationale for purchasing a specific brand within a category by calling attention to significant product differences or minor differences among products that are otherwise commodity products. This view of how marketing communication works is consistent with an impressive array of empirical studies and theory development in marketing, psychology, and economics, which suggests that providing a way to think about a decision, ‘framing the problem,’ has a significant effect on behavior (Hauser, 1986; Kahneman & Tversky, 1979, 1982; Plott & Levine, 1978; Stewart & Furse, 1986; Thaler, 1985; Tversky & Kahneman, 1974). One mechanism by which communication may reinforce habitual purchase behavior is through reminding consumers of a specific decision rule. The conceptualization of marketing communication as a rein-forcer of the purchase behavior of loyal consumers is also consistent with the empirical observation that the more time that elapses between purchase events, the less likely that consumer loyalty will persist over time. This is one reason for findings that advertising has differential effects on heavy users and light users of a product (Ackoff & Emshoff, 1975a, 1975b, 1975c).

The preceding discussion suggests that most advertising is defensive; it is designed to hold current share of marketing. Marketing communications is part of the price of doing business in a mature market. This view is also consistent with empirical findings that market share is frequently closely related to share of voice (the percentage of a brand’s advertising relative to all advertising for the product category (Horsky, 1977; Little, 1975; Stewart, 1989)).

This is not to suggest that advertising cannot ever be used for building market share in a product category. A particular competitor within a given category may choose to spend well above the equilibrium level of expenditures (that is, this competitor might raise its relative spending or share of voice above its market share). Research suggests that such competitive tactics are most likely to succeed when accompanied by a significant product improvement, a particularly compelling message, or when at least a subset of competitors is unable or unwilling to respond to the competitive threat. Short of the presence of these situational factors in the market, market response to increases in advertising in mature markets appears to be very short-lived (Little, 1979; also see below).

Several empirical studies offer evidence that is consistent with this view of marketing communication as a reinforcer of brand loyalty. In a carefully controlled study of print advertising carried out by Time, Inc. and Seagram (Time, Inc./Seagram, 1981), the largest changes in awareness, brand attitude, and purchase behavior were associated with brands that were low in awareness among consumers. While there were gains associated with advertising for high awareness brands as well, it is important to note that none of the products used in the study had been recently advertised in the test markets in which the study was carried out. Among high awareness brands, the greatest increase in awareness occurred following the first insertions of the print advertising. Changes in brand ratings (attitudes toward the brand) were very modest for high awareness brands. Although the advertising did increase consumers’ willingness to buy the high awareness brands, willingness to buy was unrelated to the number of opportunities consumers had to see the advertising. These findings are consistent with the view that for established brands, the role of advertising is largely that of reinforcing existing attitudes and behavioral patterns. When such brands have not been advertised for some time, advertising may serve to raise the general awareness of the brand, and thereby remind consumers of the availability of the brand and their previously formed attitudes toward the brand. For new brands, or brands with low levels of awareness, regardless of the length of time on the market, advertising may not only increase awareness, it may also serve as one means, along with product trial, for creating attitudes toward the brand among consumers who have not previously formed such attitudes.

The role of marketing communication as a rein-forcer of behavior may also be the reason that one of the more comprehensive studies of advertising (Information Resources, Inc., 1989; see also Lodish et al., 1995) found that about half of all of the advertising for more than 360 frequently purchased consumer goods produced no increase in sales. Among new products, almost 60% of the commercials examined in the study produced immediate increases in sales, but only 46% of the advertising executions for established products manifested any immediate increase in sales.

The results reported by Information Resources, Inc. are consistent with a study conducted by Raj (1982), who found that advertising had its only effect on brand loyal consumers in the product category he studied. Likewise, Tellis (1988) found that consumers who are loyal to a brand responded more strongly to advertising for that brand than those who were not loyal. He concludes, ‘advertising has a small effect in winning new buyers (from competitors’ offerings) but a relatively strong effect in reinforcing intensity of preference.’ Several other studies also suggest that taking share away from competitive products is difficult, expensive, and often of only short-term benefit (Adams & Blair, 1989; Appel, 1984; Axelrod, 1980; Haley, 1978; Simon, 1982; Tellis, 1988).

Advertising Effects on Switchers

Advertising appears to have different effects on switchers than loyals. Much of the research in both industry and academe has focused on this capricious group. Switchers have little or no brand loyalty. They most certainly make purchase decisions on the basis of something other than brand. Switching may be associated with price, availability, top-of-mind awareness, or any number of other factors that may enter into decision making (Tellis, 1988). In the case where price is the determinant characteristic, price promotions are likely to be the most effective vehicle for inducing brand switching. Where the determinant factor in decision making is familiarity, top-of-mind awareness, image, or some other communications-driven element, marketing communications may play an important role. It is useful to note, however, that the effect of advertising on switchers is likely to be short-lived.

There is a significant body of literature that suggests brand familiarity, or top-of-mind awareness, plays an important role in consumer choice (Aaker & Day, 1974; Axelrod, 1968; Burke & Schoeffler, 1980; Grubar, 1969; Holman & Hecker, 1983; Sutherland & Galloway, 1983; Taylor et al., 1979; Wilson, 1981). The role of awareness also appears to be far greater for low involvement types of decisions, where the consumer is looking for a simple rule for decision making (Batra & Ray, 1985). Further, there is evidence that the repetition of brand advertising and the intensity of competitive advertising play key roles in producing these effects. Several studies have demonstrated that repetition has a far greater effect on awareness than on attitude (Ray, 1982; Ray & Sawyer, 1971; Ray et al., 1971; Sawyer, 1974; Strang et al., 1975). There is also evidence that competitive marketing communications can reduce the salience or top-of-mind awareness of a brand (Alba & Chattopadhyay, 1985; Burke & Srull, 1988; Geiger, 1971; Stewart, 1989). Thus, in product categories where consumers use a simple familiarity rule, switching may simply be a manifestation of changes in the most salient brand. In the most extreme case, this may be the brand to which the consumer was most recently exposed.

Familiarity is but one type of decision rule that consumers faced with low involvement choices may use for deciding among purchase alternatives. Virtually any differentiating characteristic of a product or its advertising will do. The use of a likable spokesperson, a credible sales person, a strong image with which consumers identify, a catchy tune, or any of dozens of other creative devices might serve to predispose the consumer to purchase one brand over another. Such predispositions in low involvement situations have been shown to be highly transient and easily changed (Chaiken, 1980; Petty & Cacioppo, 1986; Petty et al., 1983). These predispositions must be constantly reinforced or they will fade rapidly, often within a week (Petty & Cacioppo, 1986). This is one reason that research on advertising execution has found that more effective commercials, both in terms of memorability and in terms of eliciting brand switching behavior, contain brand differentiating messages, that is, statements, demonstrations, images, or other devices that distinguish the advertised brand from competitors (Stewart, 1986; Stewart & Furse, 1986; Stewart & Koslow, 1989). This research has also shown that differentiating messages are even more important for established products than for new products, where being new may, in itself, be a basis for differentiation.

Unlike firmly established attitudes, these transient predispositions are likely to change in the face of marketing communications by competitors that offer different cues that the consumer might use to make a brand selection. As in the case where familiarity is the critical factor in determining switching behavior, predispositions based on other factors must be constantly reinforced. This is a different kind of reinforcement than is required for loyal consumers, and it is more expensive to execute. This is one reason why product categories characterized by a high number of switchers typically require substantially greater expenditures on advertising.

In any given product category there exists some mixture of both loyals and switchers. Thus, marketing communication must reinforce loyal consumers and reinforce the decision rules of consumers with a predisposition to switch brands. If the objective of marketing communication is to attract loyal users of competing products, the communication must also cut through the selective attention of these consumers and offer a compelling reason for switching. These are not easy tasks. This is one reason good creative directors are well paid. Indeed, research suggests that the quality of advertising, i.e., the creative execution, the strength of the selling message, plays a very substantial role in determining the effectiveness of communication in the marketplace (Adams & Blair, 1992; Arnold et al., 1987; Blair, 1987/88; Drane, 1988; Stewart & Furse, 1986; Stewart & Koslow, 1989). Even if such effective advertising can be created, there is still the question of how long it will have an effect. The evidence suggests that the effects tend to be short-lived (Appel, 1984; Axelrod, 1980; Blair, 1987/88; Krishnamurthi et al., 1986; Tele-Research, Inc., 1968).

Temporal Characteristics of Advertising

There is a pervasive view that advertising has very long-term effects. This view received some early support from studies by Palda (1964) and Ackoff and Emshoff (1975a, 1975b). These studies suggested that the effects of advertising might persist for a year or more. There are numerous other studies that indicate that communication effects may persist beyond the initial or last exposure (Bass & Clarke, 1972; Clarke, 1976, 1977c, 1977d; Clarke & McCann, 1973; Dean, 1966; Houston & Weiss, 1974, 1975; Kuehn et al., 1964; Lambin, 1972; Lavidge & Steiner, 1961; Sawyer & Ward, 1977, 1979). Thus, there is no question that advertising, and other marketing communication effects, can be long lasting: a consumer who learns about a product via advertising does not immediately forget the product. Having learned about a product via advertising, tried the product, and had a satisfactory experience, a consumer may continue to buy the product into the future. These are certainly examples of the long-term effects of marketing communication. The mechanism by which these effects occur is a different matter, however. There are two competing views as to how these long-term effects arise. The differences in these two views are subtle, but they have important implications.

Cumuluative Effects versus Current Effects

One perspective, the cumulative effects view, suggests that marketing communications do not have their full impact on sales in the period in which the communications occur. Rather, this view holds that the effect of marketing communication builds over time; the full effect of communication in a given period can only be identified by looking across multiple periods after exposure to the communication. A competing perspective, the current effects, hypothesis, suggests that sales are a function of current expenditures on marketing communication and a carryover effect that cannot be completely attributable to past expenditures. These carryover effects are posited to be the result of an interaction of marketing communication and product use. For example, marketing communication in the form of advertising may create awareness of the product and induce trial (Marks & Kamins, 1988). Product trial may, in turn, result in a positive product experience. If the product experience is positive, a loyal user may be created. This loyalty may persist over time. It would be incorrect to attribute this effect exclusively or even directly to advertising, however. Product satisfaction is the primary determinant of loyalty (Churchill & Surprenant, 1982).

In this view, advertising contributes to loyalty only insofar as it generates product awareness and trial. The generation of awareness and trial are not unique functions of advertising; there are other means by which consumers become aware of products and are encouraged to try a product. Thus, the current effects model suggests that advertising has its primary effect in the period in which it occurs. It manifests longer-term effects only because there is a long-term consequence associated with a positive product experience.

The cumulative effects model suggests that the effects of marketing communications build over time. Thus, a single exposure to a communication may be insufficient to have a detectable effect on consumers, but it does have an effect that is added to subsequent exposures until a detectable effect occurs. This view is at least implicit in any research that looks at the relationship of aggregate spending on marketing communications and market demand over multiple time periods. Indeed, the notion that multiple exposures may be necessary for communication to manifest a measurable and/or optimal effect has a long history, and is frequently referred to as ‘wearin.’ For example, Zielske (Zielske, 1959; Zielske & Henry, 1980) observed that recall increased for a message with repeated exposures. However, the pattern was different as a function of whether the messages were viewed monthly or weekly. A review of advertising repetition by Pechmann and Stewart (1991) led them to the general conclusion that both cognitive and affective responses to an advertisement (e.g., attention, recall, attitude) initially increase with increasing repetition over the first few months of an ad campaign.

The current effects model, on the other hand, suggests that the only appropriate analysis of the relationship between demand and marketing communications is one that examines the immediate effect of communication, that is, an analysis of the influence of the communications in a given period on sales in the same period. The most appropriate definition for a given period will vary by product category, but it is probable that the interval would correspond closely to the inter-purchase time, that is, the length of time that elapses between purchases of a brand (Krishnamurthi et al., 1986). For frequently purchased consumer goods this inter-purchase time interval is likely to vary from a few days to a few weeks.

A study of 22 frequently purchased brands (Harvey et al., 1989) concluded that advertising had a very immediate effect, but that gains began to fall back after eight to twelve weeks. Similarly, Krishnamurthi et al. (1986) found that the effect of advertising was immediate and had a duration of roughly the same interval as the purchase cycle for the brand (one-and-a-half weeks in this particular case). A collaborative study carried out by Time, Inc. and Seagram (1981) found that response to advertising was immediate for eight low usage products. Finally, the results of a large-scale study of 360 brands of consumer-packaged goods over a 10-year span (Information Resources, Inc., 1989), while finding that advertising effects may be long lasting, concluded:

For new products, advertising can provide significant help when it is fulfilling its role of communicating product news. The data show that increasing weight behind new product advertising is a very productive strategy. Deferring spending may not work. Because new product advertising is primarily affecting trial, its effectiveness is likely to be long term because of the repeat purchases of these new triers generated by advertising …

Once the new product advertising has performed its role of generating trial and positioning the new product in the market, the same large advertising budgets may not always be needed. Without new, compelling copy, the evidence shows that some established brand advertising might not be causing incremental sales. New, fresh copy for established products can be very effective even if run for a relatively short period. The data show that the incremental sales effects of new copy are found within six months.

Various researchers (e.g., Gibson, 1996; Jones, 1995; McDonald, 1996; Schroeder et al., 1997) have argued and offered empirical demonstrations that only one exposure is enough to trigger a maximum response to advertising. All of these findings are consistent with a current effects model of advertising response. If communication is effective, it works quickly upon exposure to the individual consumer. To the extent that different types of communication, media schedules and expenditures result in more or less rapid exposure of all consumers in the market, there may appear to be some lagged effects of advertising. These effects are artifacts of aggregating data, however; they are not real effects.

There is an impressive array of studies that can be brought to bear on the question of the length of communication effects, though almost all of this work has focused only on advertising. Econometric studies carried out using historical data, behavioral studies carried out in the laboratory, and field experiments involving controlled market testing and split cable testing, agree that the impact of advertising occurs quickly (Assmus et al., 1984; Bagozzi & Silk, 1983; Batra & Ray, 1984; Clarke, 1976, 1977b; Frank & Massy, 1967; Haley, 1978; Massy et al., 1970; Nakanishi, 1973; Sawyer & Ward, 1979; Schwerin & Newell, 1981; Sethi, 1971; Sexton, 1970; Zielske, 1959; Zielske & Henry, 1980). These studies also suggest that the effects of advertising erode over time and that this erosion is most rapid in the face of competitive advertising and other marketing programs. Generally, the more intense the competitive pressure, the more rapid the loss of advertising effects. This dissipation of effect holds for top-of-mind awareness, attitude, and purchase behavior.

These studies do not resolve the question of whether the cumulative effects model or the current effects model is the more appropriate, however. They do suggest that the direct effects of marketing communication tend to be immediate and short-lived. Indeed, even though Lavidge and Steiner (1961) took the position that ‘the effects of much advertising are long term,’ they also acknowledged, ‘if something is to happen in the long run, something must be happening in the short run, something that will ultimately lead to eventual sales results.’ Similarly, Tellis and Weiss (1995) raise the question of how communication can have a long-term impact if it has little impact in the period in which exposure occurs.

The Role of Message Frequency

There are a variety of potential explanations for the appearance of cumulative effects of communication. As noted above, the effects of communication may persist through an influence on product awareness, satisfaction and loyalty. In addition, many studies that appear to provide support for the cumulative effects model have examined aggregate market data. Such data provide insight into how communication effects influence the market as a whole, but do not indicate anything about what is happening at the level of the individual consumer. Any apparent cumulative effect in such studies may simply reflect the differential rates of exposure within the target audience. Ephron (1997) argues that advertising works most directly with the few consumers who are in the market at the time of exposure to the communication. Since consumers are continually entering the market, it may appear that advertising has a long-term, albeit weak impact on sales.

Differential exposure to a message among consumers is a function of both the consumers in the market at any one point in time and the marketer’s selection of schedule and budget for advertising. Effective communication will have an impact in the marketplace only to the extent that it reaches consumers. The level and timing of media expenditures influence the rapidity with which consumers are exposed to an advertising message. Heavy expenditures result in reaching consumers most quickly. Lesser expenditures result in a more gradual reach of consumers. Blair (1987/88) demonstrated that advertising loses its effectiveness once all consumers have been exposed to the message. Adams and Blair (1992) also found that market share gains were immediate in response to a new commercial, but the gains peaked within a few weeks and brand share started to slip back, partially as a result of a new commercial launched by a competitor. Use of a new commercial resulted in sales again increasing. Similar results have been reported by Drane (1988).

Tellis (1997) also argues that frequency in and of itself is not the key driver of communication effectiveness. He maintains that the effect of the frequency of exposure to a communication is more complex than the question of the number of exposures alone. Rather, he suggests that three factors influence the effectiveness of any given exposure. The first factor is brand familiarity. A consumer’s familiarity with a brand (measured by knowledge or purchase behavior) tends to produce communication effects with fewer exposures as a result of increased attention to communications for a familiar brand, desire for consistency between beliefs and behavior and the habituation-tedium theory of ad response (Sawyer, 1981). Second, message complexity requires more exposures for the ad to be effective since more complex concepts are being conveyed (Pechmann & Stewart, 1988; Sawyer, 1981). Finally, the need for message novelty results in either the need for new executions to refresh the message or the need for more spacing between executions to delay wearout of the message (Zielske, 1959; Zielske & Henry, 1980). The issue of wearout is considered later in this chapter.

A key factor in reconciling research on the relationship between frequency of exposure to a message and the effect of the message is recognition of the difference between opportunities for exposure to a message and actual exposure. Research that focuses on such measures of advertising intensity such as expenditures, rating points, appearances or mentions, or similar measures of the frequency of message occurrence are, in fact, measuring only opportunities for exposure to a message by individual consumers. An opportunity for exposure is not the same as an actual exposure. In addition, not every exposure that occurs will necessarily represent an undistracted, high quality exposure to the message. Thus, it may be necessary for a marketer to repeat a message several times before an individual consumer actually attends to and processes the message. This need for repetition, and the differential impact of such repetition across the target market, may produce what appear to be cumulative effects of communication. In reality, it is the cumulative effect of opportunities for message exposure that is at work rather than the cumulative effect of actual exposure to the message.

Also confounding research on the cumulative effect of advertising is that research that focuses on expenditures, or opportunities for exposure, tends to ignore the important role played by message quality and changes in messages over time. As discussed below, message quality is an especially critical factor in determining whether communication has any effect at all. In addition, studies that focus on expenditures or opportunities for exposure frequently ignore the fact that messages change over time (Stewart, 1999). There may well be a cumulative effect of multiple messages over time as consumers learn more about a product or service.

Communication Wearout

Wearout has been defined by Corkindale and Newall (1978) as: ‘That level of advertising which corresponds to the point at which an individual, or group of individuals, fails to respond to the advertising stimulus. Beyond this point, the likelihood that the individual or group of individuals will fail to respond increases despite continued repetition of the stimulus.’ The presence of advertising wearout has important implications for the debate regarding the cumulative effects model of communication and the current effects model. If communication does ‘wear out’ it is unlikely that it would have a cumulative effect. However, as Stewart (1999) observes, wearout is an ill-defined term; it means many different things, and whether it even exists is a function of how it is defined and how it is measured. Because communication has many potential goals, various studies have examined wearout with respect to very different goals: attention (Grass & Wallace, 1969), attitude (Greenberg & Suttoni, 1973), recall (Appel, 1984; Craig et al., 1976), sales (Alexrod, 1980), and brand choice (Adams & Blair, 1992; Blair, 1987/88; Drane, 1988). In addition, wearout has been defined as the absence of further incremental effects (Blair & Rabuck, 1998) and as a diminution in or negative effect in response to additional exposures to the message (Pechmann & Stewart, 1988; Scott & Solomon, 1998). Finally, there has been confusion in the literature between wearout of an individual message and wearout of entire advertising campaigns consisting of many different messages. Thus, it should not be a surprise that the literature on advertising wearout is filled with seeming contradictions.

Quite apart from issues of definition and measurement, empirical determination that wearout has occurred is difficult for two other reasons. First, marketing communication does not occur in a vacuum, there are exogenous factors and components of the communication strategy itself that may influence various measures of interest. Competitors’ actions (including competitors’ own responses to a successful communications program), changes in communication strategy, and general economic conditions, may produce changes in various in-market measures of communication effectiveness. Second, it may not be easy to verify the presence of wearout, particularly if results for one important dependent measure (e.g., attention) are stable or increasing, whereas those for another (e.g., recall) are declining. Indeed, it is probably not appropriate to discuss wearout in a general sense. Rather, it may be more useful to define wearout with respect to a specific measure of communications effect.

Given the differences in definitions and measures of wearout, it is not surprising that different reasons have been offered for its occurrence. Some researchers (e.g., Grass & Wallace, 1969; Greenberg & Suttoni, 1973; Jacobovitz, 1965) suggest that ceiling effects are in part responsible for wearout. It is certainly the case that no incremental effect of communication can be observed with respect to such measures as recognition and recall once 100% of the message recipients have processed the message. In addition, repeated exposure to a given message may result in a lack of attention to the message with additional repetition. This lack of attention may reduce recall of the message over time through a process of forgetting or through the interference of competing messages (Stewart, 1989).

On the other hand, Calder and Sternthal (1980) hypothesize that wearout is a function of changes in information processing in response to a given message over time. These researchers focus more on changes in attitude rather than memory. They posit that thoughts in response to initial exposures to a message are likely to be closely linked to the message. As message repetition continues, however, thoughts of message recipients turn from ‘message-related thoughts’ to ‘own thoughts.’ Since message-related thoughts are likely to be more positive toward the product than one’s own thoughts, wearout in this begins to occur. Wearout this context is defined as a more negative attitude toward the product after multiple exposures to a given message. Interestingly, this view predicts that wearout can occur despite the implementation of strategies designed to increase attention.

The most common definition of wearout has focused on the absence of incremental effects associated with the repetition of a message (Stewart, 1999). Pechmann and Stewart (1988) observe that negative effects associated with repetition of a message are most likely to be observed in laboratory settings where the message is repeated frequently within a very short period of time. They suggest that such negative effects are far less likely to occur in the marketplace since marketing communications tend to be spread over time and are frequently varied. On the other hand, an absence of incremental effects is far more likely. A communication that is fully learned across all consumers cannot exhibit higher awareness, recall or comprehension. Similarly, a communication that has already produced a strong change in attitude or behavioral intention is unlikely to produce further change. It is important to recognize, however, that failure to observe further incremental change is not the same as the communication having no effect. A communication may very well reinforce memory or an existing attitude, and failure to repeat the communication (in the absence of other communications) may well result in decay in memory or positive attitude over time. Thus, wearout need not mean that a communication is not producing an effect; rather it means only that the communication is not producing any observable incremental effect.

Various strategies have been suggested for dealing with communication wearout. For example, Rossiter and Percy (1997), as well as Grass and Wallace (1969), suggest that strategies designed to increase attention to the message may be effective in countering wearout. This can be accomplished by using multiple executional variations on a similar theme so that the rate of wearout at any given exposure frequency could be slowed (cf Gorn et al., 1997). Similarly, new messages that contain additional information may provide a means for obtaining incremental effects in attitude, intention, and purchase behavior (Blair & Rabuck, 1998; Lodish et al., 1995).

Evidence regarding communications wearout further weakens the case for the cumulative effects model of advertising. If repetition of a message produces no incremental effect, and, on occasion, can produce negative effects, there is little basis for concluding that communication effects build over time. Such building of effects over time and with repetition appears likely to occur only when the communications involved are themselves incremental in terms of the information provided to the consumer. Repetition alone does not appear to increase the effect of a message.

Long-term Effects of Advertising

Although the optimal frequency of communication is arguably a function of other identifiable factors, the question remains as to how long exposure to a communication remains effective over time. There are numerous studies focusing on sales that indicate that advertising effects may persist beyond the initial or last exposure (Bass & Clarke, 1972; Clarke, 1976, 1977a, 1977b; Clarke & McCann, 1973; Dean, 1966; Houston & Weiss, 1974, 1975; Kuehn et al., 1964; Lambin, 1972; Lavidge & Steiner, 1961; Sawyer & Ward, 1977, 1979). Thus, there is no question that communication effects appear to be long lasting. However, as noted above, these effects appear to be the result of an indirect effect that communication has on sales. This indirect effect is mediated through product awareness and trial and product satisfaction. Marketing communication precipitates and reinforces a process that produces a tendency to respond in a particular way toward a product. Communication is an important part of this process, but it is not the entire process.

Factors Moderating the Effects of Advertising

In the course of the preceding discussion of communication effects, several important characteristics of the marketplace in which communication takes place, and of communications strategy, have been identified. These characteristics have been shown to influence the effect of marketing communication. They include:

  • The stage in the life cycle of the product category, that is, whether the category is new or established;
  • The competitive position of the advertised product and the intensity of competitive activity within the market;
  • The familiarity of the product within the target audience;
  • Whether the decision to purchase a specific brand is highly involving or one that is ‘low involvement’;
  • The nature and execution of the communication message.

All of these factors interact with marketing communication in ways that change the nature of consumer response to the communication. While there are, no doubt, numerous other mediating factors, these five are among the most frequently studied, and are among the most general of factors that may influence the effects of communication in the marketplace.

The Role of the Product Life Cycle

As noted earlier, the objectives of marketing communication are frequently different for a new product and an established product. There is sound empirical evidence that different types of messages, executional factors, and exposure schedules have differential effects for new versus established products (Belch, 1981; Calder & Sternthal, 1980; Politz Media Studies, 1960; Ray & Sawyer, 1971; Stewart & Furse, 1986; Stewart & Koslow, 1989; Stephens & Warren, 1984; Time, Inc./Seagram, 1981). Thus, any discussion of how marketing communication works must consider the life cycle of the product category, as well as the life cycle of the individual brand.

In product categories that are new (in contrast to new brands in established categories), the market itself is growing. Thus, expenditures on marketing communications and sales will tend to grow together as the market expands, but both are driven by market growth. Communication may stimulate sales in the sense that it increases awareness of the product among individuals who are not yet aware of it, but this effect will not influence sales unless the product itself provides a benefit or relative advantage.

Marketing communication for new brands within an established product category clearly must have as its objective the creation of awareness and the stimulation of trial. In established categories, however, awareness and trial are more likely to be generated among current users of competitive products. For established products, awareness and trial are less important objectives since high levels of awareness and trial are likely to be present among consumers. In an established product category, emphasis shifts from building awareness and trial to the reinforcement of brand loyalty. In this latter case, the effect of marketing communication will not be revealed in incremental sales. Rather, the effect of marketing communication will only become apparent when it is eliminated.

The Role of Competition

In most markets the purchase of a specific brand is but one of a number of possible responses available to consumers. There are frequently directly competitive products, that is, different brands of essentially similar products, and indirect competitors, that is, products that are perceived and used by the consumer as substitutes for the same purpose even though the products themselves are not necessarily comparable. Examples of these latter types of substitutes would be candy vs. raisins, vegetable juice vs. soft drinks, and an in-theater movie vs. a rented videotape.

Within any given market the availability of competitive products, of whatever type, is an important determinant of the extent to which advertising has longer or shorter-term effects. Further, the intensity of the marketing efforts for these competitive products mediates the effect of advertising. Thus, a firm that advertises a brand that holds a virtual monopoly, or that operates in a market composed of relatively weak competitors, may find that its advertising has effects that last over some time. Some of the early empirical studies of the duration of advertising effects, for example, Palda’s (1964) classic work on the advertising of Linda Pinkham’s Vegetable Compound, and the work of Ackoff and Emshoff (1975a, 1975b, 1975c) with Anheuser-Busch, suggested that the effects of advertising might persist for a year or more after its elimination. In both of these cases the brands were the dominant product in the category and faced relatively weak competition. However, in competitive markets, the communications and other marketing activities of competitors tend to mitigate the effects of marketing communications. Thus, the intensity of competition is one important factor that must be examined when examining the effects of marketing communication.

Stewart (1989) has offered a theoretical model to explain the effects of communication within a competitive market. This model is based on a well-known model of learning within the field of psychology, the accumulation model (Herrnstein, 1970, 1974; Mazur & Hastie, 1978). The model simply states that, at any one point in time, there exist numerous competing response tendencies within an individual consumer, some of which are stronger than others. Marketing communications for a particular brand, even if effective, does not displace dispositions to purchase competitive products: it merely strengthens the tendency to buy the brand for which the new communication has been received. Likewise, communications about competitive products, if effective, simply increases the tendency to purchase those products. Ultimately, the brand purchased will be the one with the strongest response tendency at the time of purchase.

Numerous studies have offered empirical evidence of the important role that competitive actions have in mediating the influence of marketing communication (Alba & Chattopadhyay, 1985; Burke & Srull, 1988; D’Souza & Rao, 1995; Geiger, 1971; Keller, 1991; Lambin, 1976; Moran, 1988; Pechmann & Stewart, 1988, 1990; Roediger & Schmidt, 1980; Schwerin & Newell, 1981; Unnava & Sirdeshmukh, 1994). This evidence suggests that competitive communications have an effect on memory as well as purchase behavior (Burke & Srull, 1988; Geiger, 1971). In addition, Lambin (1976: 109) concluded, ‘the order of magnitude and the opposite sign of own and competitive advertising coefficients indicate a tendency toward reciprocal cancellation in the market as a whole.’ This conclusion is consistent with the view offered earlier that much of the marketing communication for mature products serves to defend the firm’s share of market from loss to competitors, rather than taking share from competitors.

The Role of Product Familiarity

Product familiarity has been shown to have a pronounced influence on the effects of advertising in the marketplace (Adtel, 1974; Alba & Chattopadhyay, 1985; Belch, 1981, 1982; Cacioppo & Petty, 1985; Calder, 1981; Calder & Sternthal, 1980; Craig et al., 1976; Dedeo, n.d.; Krugman, 1972; Naples, 1979; Petty & Cacioppo, 1986; Petty et al., 1983; Politz Media Studies, 1960; Raj, 1982; Ray & Sawyer, 1971; Sawyer, 1973, 1981; Simon & Arndt, 1980; Stephens & Warren, 1984; Tellis, 1997). As noted above, there is a significant body of literature that suggests brand familiarity, or top-of-mind awareness, plays an important role in consumer choice (Aaker & Day, 1974; Axelrod, 1968; Burke & Schoeffler, 1980; Grubar, 1969; Holman & Hecker, 1983; Sutherland & Galloway, 1983; Taylor et al., 1979; Wilson, 1981).

The role of familiarity also appears to be far greater for low involvement types of decisions, where the consumer is looking for a simple rule for decision making (Batra & Ray, 1985). Further there is evidence that the repetition of brand advertising and the intensity of competitive advertising play key roles in producing these effects. Several studies have demonstrated that advertising repetition has a far greater effect on awareness than on attitude (Ray, 1982; Ray & Sawyer, 1971; Ray et al., 1971; Sawyer, 1974; Strang et al., 1975). There is also evidence that competitive advertising can reduce the salience or top-of-mind awareness of a brand (Alba & Chattopadhyay, 1985; Burke & Srull, 1988; Geiger, 1971; Stewart, 1989). Thus, in product categories where consumers use a simple familiarity rule, it is important for advertisers to be constantly in the marketplace with communications, in order to maintain consumer awareness and combat competitive efforts to build even higher levels of awareness.

The Role of Consumer Involvement

Another factor that appears to mediate the influence of advertising is the level of involvement of the consumer in the purchase decision. Certain decisions, those that involve high risk for the consumer, appear to predispose consumers to greater information search and more effortful decision-making strategies.

It is useful to recognize that in most purchase situations, two different types of decisions are involved: 1) the decision to use (purchase) the generic product (product category), and 2) the decision about which specific brand within a product category to select. Most brand decisions involve rather little risk, and for most consumers are relatively trivial. Choices among brands of a frequently purchased consumer good are relatively trivial, inasmuch as there are usually relatively small differences among competing products. When making such decisions consumers frequently look for very simple rules of choice, or heuristics, that serve to reduce the effort involved in the decision. Marketing communication may be a particularly useful means for suggesting such heuristics.

The Role of the Message and Execution Factors

A very significant literature on message factors that influence the effectiveness of marketing communications now exists. This literature suggests that a focus of the message on the product or service is especially important, as is the offering of a brand differentiating message: a statement of how the product differs from competitive products, or a reason to buy the product instead of a competitive product (Stewart & Furse, 1986; Stewart & Koslow, 1989). In a carefully controlled study of 20 different commercials in several product categories, Blair (1987/88) found that the effects of advertising were highly dependent on the ‘quality’ of the advertising message. When the message was not persuasive, as measured by a laboratory measure of pre-post exposure brand switching, no amount of spending on advertising media produced any effects on sales. When the message was persuasive, increased spending on advertising resulted in an increase in market share. This sales increase was immediate and the duration of the effect was directly related to media expenditures. The findings of this study are consistent with the findings of several other empirical studies that have found that the content and execution of advertising is an important determinant of consumer response to such advertising (Adams & Blair, 1992; Arnold et al., 1987; Drane, 1988; Jones, 1995; Stewart & Furse, 1986; Stewart & Koslow, 1989).

Research on advertising effects has demonstrated positive effects for changes in either the advertising execution or the advertising campaign itself on sales (Batra et al., 1995; Eastlack & Rao, 1989; Lodish et al., 1995). But an important question still remains: what message factors should be considered in the creation of a ‘effective/quality’ communication?

One such message factor involves message content in the form of the order of presentation of persuasive arguments. That is, should the strongest arguments be placed at the beginning, middle or end of the communication? Research generally shows that people best remember the first item in a series of similar items they are exposed to, remember second best the last item, and remember the items in between less well (Ebbinghaus, 1885). Therefore, it is common knowledge that strong arguments should be either placed first or last in a promotional communication. However, if the goal is attitude change, and the target market is opposed to the content of the communication, the communicator may be better off by presenting strong arguments first in an attempt to diffuse ensuing counter-arguments.

Such strong arguments may also serve to attract attention to the communication and to cut through advertising clutter. In fact, the use of a two-sided refutational argument early on in the communication may be extremely effective in diffusing counter-arguments specifically, if the targeted consumers entertain negative thoughts toward the product or service advertised. Using a biological analogy that McGuire (McGuire & Papageorgis, 1961) termed ‘inoculation theory,’ Kamins and Assael (1987) found that the use of a communication that presented both product benefits and refuted weaknesses led to less counter argumentation on the part of the customer, than did a communication that only presented the product in a positive light. Thus, by presenting diluted product negatives in advance, the consumer is inoculated against the effectiveness of a competitor’s use of these negatives in an attempt to derail the purchase of the firm’s brand. Consistent with Inoculation Theory, overall counter argumentation may be reduced as refuted arguments generalize (McGuire & Papageorgis, 1961). However, this counter argumentation reduction comes at a cost of reducing the credibility of the advertiser (Kamins & Assael, 1987). That is, two-sided appeals that do not refute a negative claim are viewed by the recipient as significantly more credible than those that do refute the claim. This finding has a basis in attribution theory in that it is expected that the communicator would refute a negative claim.

Another message-related consideration deals with whether the communicator should draw a conclusion for the audience (closed-ended ads) or let the audience freely draw their own conclusion (open-ended ads). Although one could argue that consumers may resent being told what to infer from a given advertisement, there is a danger in not summarizing the main advertising claims for the reader. Jacoby and his colleagues (Jacoby & Hoyer, 1987; Jacoby et al., 1980) demonstrated that for advertising communications appearing in mass media magazines, the median miscomprehension rate for individual meanings is 11.8%. That is, the probability is greater than one in ten that a given consumer will miscomprehend a specific meaning conveyed in an advertisement (e.g., such as who is sponsoring the advertisement). Typically, since a given advertisement contains multiple meanings, Jacoby’s finding suggests that the probability is probably close to certainty that a given consumer will miscomprehend some component of an advertisement that has just been read. This finding suggests there is an obvious need to summarize advertising messages for consumers.

Interestingly, research has shown that the open-ended ad is particularly effective in enhancing consumer attitudes and purchase intentions for those who are highly involved in the product class (Sawyer & Howard, 1991). This is because those consumers with high involvement tend to be motivated to draw their own conclusion from an advertisement. According to Sawyer and Howard (1991),

Attitudes resulting from effortful self-generated conclusions should be more positive than attitudes resulting from less effortful processing of conclusions explicitly provided in a message (Linder & Worchel, 1970).

The goal(s) and objective(s) of advertising play, or should play, a significant role in influencing the need for and impact of various message and execution factors. For example, if an advertisement is designed to create an affectively based desire for the product, then a fear appeal as an executional approach may be appropriate (Ray & Wilkie, 1970; Sternthal & Craig, 1974). These appeals are typically operationalized by showing the consumer the unfavorable consequence of not using a particular brand of product (e.g., the consequences of losing your traveler checks if they are not American Express; the risk of a poor connection if you are not using AT&T). The effectiveness of fear appeals in advertising has been linked theoretically to the learning process (McGuire, 1969) through both drive and cue functions. That is, as a drive, fear has the impact of increasing message acceptance (facilitating effect). However, as a cue, fear has the potential to arouse message avoidance, which decreases message processing and ultimately attitude change (inhibiting effect). Therefore the impact of a fear appeal is hypothesized to be a function of two opposite forces and the empirical question that remains, related to the ‘optimal’ level of fear that should be induced by the advertiser. The research of Keller and Block (1996) answered this latter question by showing that moderate levels of fear led to the greatest degree of message persuasion consistent with the perspective that the decrease in message reception accelerates faster than an increase.

If an advertising goal is cognitively based, then the use of humor may be appropriate (Sternthal & Craig, 1973). Humor works at three levels. First, it helps to cut through the clutter by attracting the receiver’s attention to the advertisement. Secondly, it serves to weaken counter-arguments to the message through distraction, by directing the receiver to focus on the message form as opposed to the message content. Finally, it works on an affective level by enhancing the receiver’s mood. There are some concerns regarding the use of humorous ads, however. First, it can be argued that if the receiver’s attention is focused on the joke then the processing of the advertisement content is compromised. Therefore humor may be more appropriate for those with low as opposed to high involvement with the product, since such an ad may cut through the clutter with its entertaining characteristics (Weinberger & Campbell, 1990/91). However, the longer term impact of the humorous commercial may be compromised for all consumers because the humor in the advertisement may lose its impact with repetition. One could easily construct an argument that humorous ads may wear out at a faster rate than if the humor were absent.

Some 30 years ago the FTC decided to allow the use of comparative advertisements as an executional device, so that consumers would have more complete information as they moved closer to the purchase decision. However, research has shown that direct comparative claims are differentially effective as a function of the relative market share of the sponsoring brand. That is, for low share brands direct comparative ads enhance attention, resulting in increased purchase intentions. However, for high share brands the use of comparative advertising may be detrimental, due to exposure of the competition and confusion about the sponsor (Pechmann & Stewart, 1990, 1991). This finding suggests a continued need for specific research on how marketers can increase top-of-mind awareness of their brands relative to competing brands in any context, as well as for a more compete understanding of those mediating factors which impact advertising structure.

Finally, there is a stream of research that examines the combined impact of visual and verbal components of advertising through what is known as the Dual Component Model (Mitchell & Olson, 1981). This research shows that both verbal and visual cues affect brand attitudes in two ways. First, brand attitude may be impacted through the effect of the visual/written stimuli on product attribute beliefs. Secondly, attitude might be changed as a function of the impact of the visual/written stimuli upon one’s attitude toward the advertisement.

In general, research has shown that pictures that elicit a favorable mood (i.e., a tropical sunset, a canoe gently drifting on a lake at dusk) have an impact on attitude toward the advertisement and brand that extends beyond the effect they have on product beliefs (Miniard et al., 1991, 1992). This can be explained by the fact that a visual image has many different interpretations, and therefore while a favorable image may be reflected in terms of a generally more favorable attitude toward the advertiser or product, it would be difficult to expect such an image to impact specific product characteristics that make up brand belief.

For advertisements in which the visual image conveyed in the advertisement is designed to be congruent with the advertising copy, findings vary as a function of the degree to which the verbal copy is laden with imagery value. For copy with low imagery value, inclusion of visuals in the ad increased both immediate and delayed recall of product attributes. When the copy was high in imagery value, inclusion of visuals in the advertisement had little impact (Unnava & Burnkrant, 1991). Some research has been undertaken to examine the impact of advertising content that is inconsistent with respect to copy and visual imagery. Here it has been found that inconsistency results in generally greater recall and processing of the information presented (Houston et al., 1987). Such an approach offers a possible solution for those consumers who normally would not attend to the advertisement. However, although recall may be enhanced, the link to more favorable attitude and purchase intention may still be lacking.

Clearly the advertiser has many different message and execution factors at his/her fingertips. The use of a specific factor should be considered in light of the intention or goal of the advertisement, with the knowledge that sometimes increases in cognition, for example, may come at a cost of a decrease in affect or purchase intention. Changes in ad content or message structure should come incrementally so that the effect of each change can be studied and examined independently of the interaction of other factors that may be introduced.

Whereas marketers have control over the content of advertising and other forms of marketing communications, they do not have control over an especially powerful form of communications in the marketplace. This form of communication is communication between and among consumers, generally called word-of-mouth (WOM) communication. While marketers do not have the type of control over WOM communication that they have over advertising, considerable time and expenditures go into efforts to influence and induce WOM among consumers. Thus, marketers often attempt to create advertising that will be talked about at the water cooler. Event marketing, in which products or marketing communications play a prominent role, may also facilitate WOM among consumers. Thus, it is appropriate to consider the role and influence of WOM communication within a marketing communications context.

Word-of-Mouth Communication

WOM is a form of media and is probably the oldest, the most extensive and one of the most effective. In a marketing context it is a form of communication that conveys information about products and services in a verbal format mainly through conversations and group discussions (Brown & Reingen, 1987; Herr et al., 1991; Katz & Lazarsfeld, 1955; Reingen & Kernan, 1986). WOM communication about products has been shown to influence product evaluation to an even greater extent than information from a well-known objective source such as Consumer Reports (Herr et al., 1991). In fact, estimates have maintained that as much as 80% of all buying decisions are influenced by an individual’s direct recommendation (Voss, 1984).

The question remains as to why WOM exerts such a strong influence on consumer choice and judgment? To answer this question Herr et al. (1991) manipulated information vividness as a function of the manner in which information was presented to respondents. That is, half of their subjects were presented with anecdotal information presented in a face-to-face manner (the vivid WOM condition), and the other half were exposed to the same information in a printed mode (the pallid condition). The authors concluded that the effectiveness of WOM information could be explained by the fact that information that is received in a face-to-face manner is more accessible from memory than information presented in a less vivid format. Others seem to suggest that the effectiveness of WOM can be directly attributed to the confidence and perceived credibility the receiver has in the information. That is, oftentimes it is sought out from people in whose opinions the receiver has extreme confidence (Kapferer, 1990). Although some advertisements may achieve vividness through the use of imagery (Kisielius & Sternthal, 1986), advertising is not as credible or influential a source of information as that gleaned from family and friends.

WOM has been found to be more pervasive under certain market conditions inclusive of the evaluation of high involvement products and services (Feldman & Spencer, 1965). Kapferer (1990) explains this phenomenon in that high involvement products are important to the consumer and services can be characterized as credence goods (i.e., goods that the buyer normally finds hard to evaluate even after consumption (Ostrom & Iacobucci, 1995)). Moreover service quality is not necessarily standardized (as is production quality) and hence can vary considerably. Therefore, under these conditions, the consumer must rely on information that represents both the experiences of others and is perceived as credible. Similar to what has been observed in a personal selling context, under high involvement purchase conditions, the information supplied verbally by the salesperson was found to be important, not necessarily, however, for its credibility, but rather for its flexibility in responding to purchase concerns.

The importance of WOM as an informational source also seems to become prominent during the final stages of the consumer’s decision-making process when a choice must be made. Here consumers seek out credible sources of information, and speaking with opinion leaders, market mavens and experts (Feick & Price, 1987) and/or friends and family is often a strategy of choice. It is clear that advertising and WOM information can effectively interact with each other. As Kapferer (1990) notes, advertising can be used to inform the consumer that a product or service has entered the market (e.g., advertising for a new movie). WOM informs the viewer whether it is worth going to see it.

One of the limitations of WOM is that the company has little control over its valence or degree to which it is spread. Therefore it is important to consider strategies designed to minimize the potential for negativity. One such strategy involves the obvious—producing quality products and services under strict, quality control standards. Responding to consumer complaints in a timely manner can also go far to diffusing potential negative WOM (File et al., 1992; Folkes, 1984).

As noted earlier, unlike advertising and other forms of marketing communication, neither the timing nor the content of WOM is under the control of the manufacturer. Compounding this is the fact that WOM communication often includes negative accounts of products or services because consumers often use WOM to express dissatisfaction (Folkes et al., 1987; Richins, 1983; Swan & Oliver, 1989). Hence, firms are often particularly concerned about negative WOM, especially when there is no evidence of it being true—that is, when it is a rumor.

Kamins, Folkes and Perner (1997) reported that, on average, consumers are exposed to approximately one rumor about companies or brands in a given year. Although this may not appear to be extensive, it should be noted that 92.6% of these rumors were classified as negative. Consistent with prior research, the authors observed that consumers typically attach lower credibility and importance to rumor relative to other sources of marketing communication (e.g., advertising, published sources and product trial). Although, rumor’s low credibility and importance might be suggestive of a minimal impact on purchase decisions, Tybout et al. (1981) suggest that this is not the case. According to these authors, ‘even when an association is disbelieved, it may be stored in memory and may influence evaluation.’ Hence rumor is impactful simply because it is processed. This led these authors to propose three potential strategies based on information processing theory, in order to forestall the impact of rumor.

The first strategy, called ‘refutation,’ involved an attempt to discredit the rumor by presenting facts that supposedly proved it false. This approach was found to be ineffective since by refuting the rumor, it served to increase the rehearsal of it in the consumer’s mind, thereby strengthening the association of the product or brand with the rumor itself. A second strategy involved what was termed a ‘storage’ strategy. Here at the point at which the rumor is spread, an attempt is made to associate the focus of the rumor with a second, more favorable object rather than with the object originally specified in the rumor. So, for example in the context of the rumor that a McDonald’s hamburger had worms in it, one could mention that worms are a delicacy and that they are often used in fancy sauces at expensive French restaurants (Tybout et al., 1981). Finally, a third strategy used to counteract rumor is known as ‘retrieval,’ since it impacts information retrieval. Here subjects’ attention is directed to a new stimulus that is designed to redirect thoughts about the rumor object away from the rumor. Again, in terms of McDonald’s, consumers could be asked about the cleanliness of the nearest location, whether it has a children’s playground, and the price of the Big Mac. Even if the new stimulus does not completely inhibit the retrieval of associations between the object and the rumor, it does have the potential to dilute these associations with other thoughts in active memory. It seems that the most effective strategy to forestall rumor is that of ‘retrieval,’ since ‘storage’ only works at the point at which the rumor is being spread. Clearly, it is impossible to be everywhere a rumor is disseminated. These strategies, designed to forestall the impact of rumor, are extremely important in light of the findings of Kamins, Folkes and Perner (1997), who reported that the majority of consumers made no attempt to verify the truthfulness of rumors they confronted and that consumers are more likely to spread rumors with negative as opposed to positive outcomes.

According to Shibutani (1966), rumors develop as a function of the importance and ambiguity of the issue in question. This of course is suggestive of future research in the area designed to forestall rumor. For example, would an approach designed to clarify as opposed to refute be more effective in forestalling rumor? Can the spread of rumor be slowed by de-emphasizing the importance of the rumor situation or the rumor attribute? Given the increasing prevalence of rumor in relation to global companies, answers to these research questions are imperative.

The Future of Research on Marketing Communications

Marketing communications is one of the most highly investigated areas within marketing. As this review has demonstrated, there is an impressive body of research and theory related to how marketing communications works, what factors influence its effectiveness, and how its effects should be measured. Few areas within marketing have received as much attention from academic and industry researchers. Given the impressive body of research on marketing communications it might seem that the area is largely understood. In fact, important questions remain to be examined.

Within the past decade, a new form of communication has emerged. This new form of communication is electronic, but it is similar to other forms of communication: 1) it shares many of the characteristics of print and/or broadcast advertising, at least with respect to the more traditional advertising that appears on it (banner ads, e-announcements); 2) it can be interactive, but without the interpersonal dimensions of personal selling; and 3) it provides the opportunity for direct response from and to the customer. The fact that electronic interactive marketing communication shares many of the characteristics of other forms of communication implies that much of what is known about communication applies in this new medium.

On the other hand, there are unique dimensions of electronic interactive media. There is interaction but this interaction is without the personal, face-to-face dimensions of such interpersonal communication modes as personal selling and WOM. Electronic interactive communication also provides a wider and more immediate scope of communication among customers (Pavlou & Stewart, 2000). Such communication remains a source of new ideas for empirical research and development of theory. Although the Internet is widely heralded as a new medium for interactive communications (Alba et al., 1997; Hoffman & Novak, 1996; Stewart & Zhao, 2000), consumers have already begun to provide evidence that they have integrated the Internet experience into their broader media use. Almost half of all personal computers are in the same room as the television set, and simultaneous viewing of television and access to the Internet are common (Cox, 1998). Such consumer-directed integration of television and the Internet is but one example of interactivity involving the integration of media by consumers. Combinations of older media, such as traditional print and broadcast advertising with the telephone (especially, but not exclusively, 800 telephone numbers), have long provided a degree of interactivity. At the most general level, feedback via sales reflects interactivity. Interactivity is, therefore, a characteristic of the consumer, not a characteristic of the medium; consumers can choose to respond or not. Thus, in this sense interactivity is not really new. What is new are the speed, scope, and scale of interactivity that is provided by new information and communication technologies.

Investigation of the role of interactivity is still in its infancy. Much remains to be learned about how consumers will use the new interactive media. In addition, much remains to be learned about how the various elements of the marketing mix interact to produce outcomes in the marketplace. A great deal of the extant research on marketing communication has tended to focus on one or a few isolated elements of the marketing mix. Rather little systematic work has been done on how various forms of marketing communication and of the larger marketing mix interact. Much of this integration is self-directed by consumers, at least in part. Research that integrates the behavior of consumers and their goals into an understanding of the effects of marketing communications would be especially useful for developing theory and guiding practice.

The rise of new interactive media and the growing ability of marketers (and others) to capture and manage data at the level of the individual consumer using information technology also raises new questions about the economics of marketing activity. Interactivity and data capture are not costless. In addition, individual consumers have preferences related to how they wish to interact (or whether they wish to interact at all), and concerns about the use of personal data. Thus, for the marketer there are questions of how to determine the optimal level of interaction with an individual consumer. This question has given rise to questions about the lifetime value of consumers under different scenarios of interaction (Niraj et al., 2001; Reinartz & Kumar, 2000). From the perspective of the consumer, a variety of issues related to privacy and intrusiveness beg investigation (Milne, 2000). Finally, the ease with which consumers can communicate with one another about common interests, about experiences with products, services and companies, and otherwise share information raises interesting questions regarding the relative balance of power between buyers and sellers. Thus, despite the substantial body of literature that already exists on the effects of marketing communications there remain important and largely unexplored areas of research for the future.