Se ha denunciado esta presentación.
Utilizamos tu perfil de LinkedIn y tus datos de actividad para personalizar los anuncios y mostrarte publicidad más relevante. Puedes cambiar tus preferencias de publicidad en cualquier momento.
informs      ®Vol. 25, No. 6, November–December 2006, pp. 740–759issn 0732-2399 eissn 1526-548X 06 2506 0740              ...
Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 20...
Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities742                                       ...
Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 20...
Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities744                                       ...
Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 20...
Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities746                                       ...
Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 20...
Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities748                                       ...
Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 20...
Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities750                                       ...
Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 20...
Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities752                                       ...
Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 20...
Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities754                                       ...
740.full
740.full
740.full
740.full
740.full
Próxima SlideShare
Cargando en…5
×

740.full

1.531 visualizaciones

Publicado el

  • Sé el primero en comentar

740.full

  1. 1. informs ®Vol. 25, No. 6, November–December 2006, pp. 740–759issn 0732-2399 eissn 1526-548X 06 2506 0740 doi 10.1287/mksc.1050.0153 © 2006 INFORMS Brands and Branding: Research Findings and Future Priorities Kevin Lane Keller Tuck School of Business, Dartmouth College, Hanover, New Hampshire 03755, kevin.keller@dartmouth.edu Donald R. Lehmann Graduate School of Business, Columbia University, 507 Uris Hall, 3022 Broadway, New York, New York 10027, drl2@columbia.edu B randing has emerged as a top management priority in the last decade due to the growing realization that brands are one of the most valuable intangible assets that firms have. Driven in part by this intense industry interest, academic researchers have explored a number of different brand-related topics in recent years, generat- ing scores of papers, articles, research reports, and books. This paper identifies some of the influential work in the branding area, highlighting what has been learned from an academic perspective on important topics such as brand positioning, brand integration, brand-equity measurement, brand growth, and brand management. The paper also outlines some gaps that exist in the research of branding and brand equity and formulates a series of related research questions. Choice modeling implications of the branding concept and the challenges of incorporating main and interaction effects of branding as well as the impact of competition are discussed. Key words: brands; brand equity; brand extensions History: This paper was received August 19, 2004, and was with the authors 4 months for 2 revisions; processed by Leigh McAlister.Introduction tasks frequently performed by marketing executivesBrands serve several valuable functions. At their most are discussed in detail: (1) developing brand posi-basic level, brands serve as markers for the offerings tioning, (2) integrating brand marketing, (3) assess-of a firm. For customers, brands can simplify choice, ing brand performance, (4) growing brands, andpromise a particular quality level, reduce risk, and/or (5) strategically managing the brand. We then con-engender trust. Brands are built on the product itself, sider the implications of this work for choice models.the accompanying marketing activity, and the use Finally, we present a simple framework for inte-(or nonuse) by customers as well as others. Brands grating the customer-market, product-market, andthus reflect the complete experience that customers financial-market level impact of brands and how thehave with products. Brands also play an important brand is created and developed by company actions.role in determining the effectiveness of marketingefforts such as advertising and channel placement. Branding Decisions and TasksFinally, brands are an asset in the financial sense. Developing Brand PositioningThus, brands manifest their impact at three primary Brand positioning sets the direction of marketinglevels—customer market, product market, and finan- activities and programs—what the brand should andcial market. The value accrued by these various ben- should not do with its marketing. Brand positioningefits is often called brand equity. involves establishing key brand associations in the Our primary goal in this paper is to both selec- minds of customers and other important constituentstively highlight relevant research on building, mea- to differentiate the brand and establish (to the extentsuring, and managing brand equity and to identify possible) competitive superiority (Keller et al. 2002).gaps in our understanding of these topics. We put Besides the obvious issue of selecting tangible prod-considerable emphasis on the latter and suggest uct attribute levels (e.g., horsepower in a car), twonumerous areas of future research.1 Five basic topics areas particularly relevant to positioning are the rolethat align with the brand-management decisions and of brand intangibles and the role of corporate images and reputation.1 For commentary on the state of branding, see special issues ofInternational Journal of Research in Marketing (Barwise 1993) and exhaustive review of the academic literature on brands and brandJournal of Marketing Research (Shocker et al. 1994). For a more management, see Keller (2002). 740
  2. 2. Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 2006 INFORMS 741 Brand Intangibles. An important and relatively Research Questionsunique aspect of branding research is the focus on 1. How does brand personality affect consumerbrand intangibles—aspects of the brand image that do decision making? Under what circumstances?not involve physical, tangible, or concrete attributes 2. Is brand personality of more strategic or tacticalor benefits (see Levy 1999). Brand intangibles are (e.g., in terms of the “look and feel” of ad executions)a common means by which marketers differentiate importance?their brands with consumers (Park et al. 1986) and 3. What is the value of the different personalitytranscend physical products (Kotler and Keller 2006). dimensions? Are certain personality dimensions moreIntangibles cover a wide range of different types of valuable at driving preference or loyalty than others?brand associations such as actual or aspirational user Does the value vary by product category or by otherimagery; purchase and consumption imagery; and his- factors?tory, heritage, and experiences (Keller 2001). A num- 4. How stable are these various personality dimen-ber of basic research questions exist concerning how sions and what causes them to evolve or change?brand tangibles and intangibles have their effects. How does this stability compare to the stability of Research Questions other types of brand associations? 1. In developing brand equity, what is the role Brand Relationships. Research has also exploredof product performance and objective or tangible the personal component of the relationship betweenattributes versus intangible image attributes? a brand and its customers. Fournier (1998) exam- 2. Are intangible attributes formative (causes) or ined the nature of relationships that customers have—reflective (constructed) reasons for equity or choice? as well as want to have—with companies (seeThat is, are they considered a priori or “constructed” also Fournier and Yao 1997, Fournier et al. 1998).after experience with the brand? Fournier views brand-relationship quality as multi- 3. When and to what extent does recall of pleasant faceted and consisting of six dimensions beyondimages (or “hot” emotions) shield a brand from less loyalty or commit ment along which consumer-positive or even negative cognitive information? brand relationships vary: (1) self-concept connection, 4. How much of brand equity is tied to unique (2) commitment or nostalgic attachment, (3) behav-attributes of a product? What happens when competi- ioral interdependence, (4) love/passion, (5) intimacy,tors copy these attributes? and (6) brand-partner quality. She suggests the fol- 5. Which attribute associations are most stable and lowing typology of metaphors to represent commonbeneficial to a brand over the long run (e.g., “high customer-brand relationships: (1) arranged marriages,quality” and “upscale”) and which have limited use- (2) casual friends/buddies, (3) marriages of conve-ful life (e.g., being “hip”)? nience, (4) committed partnerships, (5) best friend- 6. Can brands be thought of as simply a judgment ships, (6) compartmentalized friendships, (7) kin-bias or in terms of context effects in consumer deci- ships, (8) rebounds/avoidance-driven relationships,sion making? What implications do these perspectives (9) childhood friendships, (10) courtships, (11) depen-have for brand-equity measurement and valuation? dencies, (12) flings, (13) enmities, (14) secret affairs, Brand Personality. Aaker (1997) examined the per- and (15) enslavements.sonalities attributed to U.S. brands and found they While this typology contains most positive relation-fall into five main clusters: (1) sincerity, (2) excitement, ships, it may overlook a range of possible negative(3) competence, (4) sophistication, and (5) ruggedness. (e.g., adversary) and neutral (e.g., trading partner)Aaker et al. (2001) found that three of the five factors ones. Aaker et al. (2004) conducted a two-monthalso applied to brands in both Japan and Spain, but longitudinal investigation of the development andthat a “peacefulness” dimension replaced “rugged- evolution of relationships between consumers andness” both in Japan and Spain, and a “passion” dimen- brands. They found that two factors—experiencing asion emerged in Spain instead of “competency.” Aaker transgression and the personality of the brand—had(1999) also found that different brand personality a significant influence on developmental form anddimensions affected different types of people in differ- dynamics. Aggarwal (2004) explored how relationshipent consumption settings. She interpreted these exper- norms varied for two types of relationships: exchangeimental results in terms of a “malleable self,” which is relationships, in which benefits are given to others tocomposed of self-conceptions that can be made salient get something back, and communal relationships, inby a social situation (see also Graeff 1996, 1997). While which benefits are given to show concern for others’Azoulay and Kapferer (2003) have challenged the con- needs.ceptual validity of this particular brand personality Research Questionsscale, the anthromorphism of a brand is common in 1. How can a customer’s desired relationship beboth casual consumer conversation (e.g., “that brand determined? Have concerns over privacy and theis ‘hip’ ”) and advertising messages. increased use of customer data by firms resulted
  3. 3. Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities742 Marketing Science 25(6), pp. 740–759, © 2006 INFORMSin customers wanting more anonymous, transac- brand identification facilitate experiences? How muchtional relationships, or do customers still desire close does product placement (e.g., in movies) impactrelationships with companies? Does personalization brand equity and how enduring is such equity?of communication make customers feel empowered 5. How can a firm take advantage of unusual cir-and/or valued, or do they feel more exploited? cumstances such as when the brand is associated 2. How can a desired customer relationship be cul- with a positive event? How can a firm minimize thetivated by the company through marketing activities? impact of being associated with a negative event (e.g.,How do different types of marketing activities such a spokesperson behaving badly)?as advertising, customer service, and online resourcescombine to affect customer relationships? Corporate Image and Reputation. Corporate im- 3. In a world where information is widely shared age has been extensively studied in terms of its con-and discrimination is seen as bad, should a firm deal ceptualization, antecedents, and consequences (seedifferently with customers who desire different rela- reviews by Biehal and Sheinin 1998 and Dowlingtionships? Can customer relationships be segmented 1994). Corporate brands—versus product brands—areand can customers who desire different types of rela- more likely to evoke associations of common prod-tionships be identified? Does this vary by product cat- ucts and their shared attributes or benefits, peopleegory or by competing product benefits? and relationships, and programs and values (Barich 4. What is the relative profitability of different and Kotler 1991).types of customer relationships? Should some cus- Several empirical studies show the power of atomers be encouraged and others discouraged or corporate brand (Argenti and Druckenmiller 2004).“fired?” Alternatively, are there systematic ways to Brown and Dacin (1997) distinguish between cor-migrate unprofitable customers into profitable rela- porate associations related to corporate ability (i.e.,tionships? expertise in producing and delivering product and/or Brand Experience. Experiential marketing is an service offerings) and those related to corporateimportant trend in marketing thinking. Through sev- social responsibility (i.e., character of the companyeral books and articles, Schmitt (1999, 2003) has devel- with regard to societal issues), such as treatment ofoped the concept of customer experience management employees and impact on the environment.(CEM), which he defines as the process of strategi- Keller and Aaker (1992, 1998) define corporate cred-cally managing a customer’s entire experience with ibility as the extent to which consumers believe thata product or company. According to Schmitt, brands a company is willing and able to deliver productscan help to create five different types of experiences: and services that satisfy customer needs and wants • Sense experiences involving sensory perception; (see also Erdem and Swait 2004). They showed that • Feel experiences involving affect and emotions; successfully introduced brand extensions can lead • Think experiences which are creative and cogni- to enhanced perceptions of corporate credibility andtive; improved evaluations of even quite dissimilar brand • Act experiences involving physical behavior and extensions. They also showed that corporate market-incorporating individual actions and lifestyles; and ing activity related to product innovation produced • Relate experiences that result from connecting more favorable evaluations for a corporate brandwith a reference group or culture. extension than corporate marketing activity related to Research Questions either the environment or, especially, the community 1. What are the different means by which experi- (see also Gürhan-Canli and Batra 2004). In addition,ences affect brand equity? How can firms ensure that Bhattacharya and Sen (2003) extended the thinking onexperiences positively impact brand equity? Morespecifically, how can advertising trigger positive expe- consumer-brand relationships to consider consumer-riences with a brand or make negative ones less company relationships, adopting a social identity the-salient or influential? ory perspective to argue that perceived similarity 2. How much of brand-related experiences are between consumer and company identities play anunder the control of the company? How can they be important role in relationship formation.effectively controlled? Research Questions 3. When and to what extent do customers re- 1. How much are corporate images created byspond—positively or negatively—to attempts to con- words versus actions? What is the role of public rela-trol their experiences? How do customers make attri- tions and publicity in shaping corporate reputationbutions about company actions and attitudes toward and corporate brand equity?control of experiences? 2. What are important determinants of corporate 4. How does the recognition or realization of com- credibility? How do “corporate social responsibility”pany involvement impact brand experiences? Can or cause marketing programs work?
  4. 4. Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 2006 INFORMS 743 3. How do corporate images affect the equity of and cross-cultural implications of brand names (e.g.,individual products? Alternatively, how do individ- Zhang and Schmitt 2001, Tavassoli and Han 2002).ual product equities build up to corporate equity? Although companies frequently spend considerable 4. What is the impact of corporate image on cus- sums on the design of logos, little academic researchtomer purchases and firm profitability and value? has explored the impact on consumer behavior ofDoes it operate directly or indirectly through its effect logo design or other visual aspects of branding (seeon specific brand equity? Schmitt and Simonson 1997 for background discus- sion). As one exception, Henderson and Cote (1998)Integrating Brand Marketing conducted a comprehensive empirical analysis ofA variety of branding and marketing activities can 195 logos to determine the ability of different designbe conducted to help achieve the desired brand posi- characteristics to achieve different communicationtioning and build brand equity. Their ultimate suc- objectives (see also Henderson et al. 2004, Janiszewskicess depends to a significant extent not only on how and Meyvis 2001).well they work singularly, but also on how they A related area, packaging, has begun to receivework in combination, such that synergistic results greater attention in recent years (e.g., Garber et al.occur. In other words, marketing activities have inter- 2000, Folkes and Matta 2004). For example, Wansinkaction effects among themselves as well as main has conducted several studies related to packagingeffects and interaction effects with brand equity. Three size and shape and consumption (e.g., Wansink andnoteworthy subareas of this topic are the brand- van Ittersum 2003; see also Raghubir and Krishnabuilding contribution of brand elements, the impact 1999).of coordinated communication and channel strategies Research Questionson brand equity, and the interaction of company- 1. What are the brand-building contributions ofcontrolled and external events. brand logos and other nonverbal brand elements? Are names and logos differentially effective or important Integrating Brand Elements. Brands identify and in different circumstances, e.g., for high versus lowdifferentiate a company’s offerings to customers and involvement purchases or early versus late in the lifeother parties. A brand is more than a name (or cycle?“mark”). Other brand elements such as logos and sym- 2. How are visual and verbal effects manifestedbols (Nike’s swoosh and McDonalds’ golden arches), in consumer memory for brand elements? Which arepackaging (Coke’s contour bottle and Kodak’s yellow more accessible? Do more easily accessible elementsand black film box), and slogans (BMW’s “Ultimate influence or bias what is recalled subsequently?Driving Machine” and VISA’s “It’s Everywhere You 3. From both a physiological and psychologicalWant to Be”) play an important branding role as well. perspective, how do brand and design elements gain A number of broad criteria are useful for choos- attention and instill favorable attitudes? How long ofing and designing brand elements to build brand a productive life do they have, i.e., when do theyequity (Keller 2003): (1) memorability, (2) meaning- cease being effective?fulness, (3) aesthetic appeal, (4) transferability (both 4. How do consumers integrate packaging andwithin and across product categories and across geo- other brand element information with informationgraphical and cultural boundaries and market seg- about product performance, marketing communica-ments), (5) adaptability and flexibility over time, and tions, or personal experience?(6) legal and competitive protectability and defensibil- 5. Are there criteria for combining a diverse setity. Brand elements vary in their verbal versus visual of brand elements? How do marketers know if theircontent and product specificity. Although a robust brand elements are “well integrated?” What are theindustry exists to help firms design and implement financial consequences of integration?these various brand elements (Kohli and LaBahn Integrating Marketing Channels and Communi-1997), comparatively little academic research atten- cations. Marketers employ an increasingly variedtion, even in recent years, has been devoted to the means of communication (e.g., various forms oftopic of designing and selecting brand elements other broadcast, print, and interactive advertising; tradethan brand names. and consumer promotions; direct response; sponsor- Brand name properties have been studied exten- ship; public relations; etc.) and multiple means ofsively through the years. For example, researchers going to market (via retailers, company-owned storesstudying phonetic symbolism have demonstrated or outlets, telephone, Internet, mail, etc.). Some mar-how the sounds of individual letters can contain keters have attempted to orchestrate these activitiesmeaning that may be useful in developing a new to create synergistic effects (Duncan 2002).brand name (see Klink 2000, Yorkston and Menon Research has shown that coordinating market-2004 for reviews). Other research has examined global ing activities can lead to beneficial results (Naik
  5. 5. Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities744 Marketing Science 25(6), pp. 740–759, © 2006 INFORMSand Raman 2003). For example, print and radio society that self-selects on a basis of a shared com-reinforcement of TV ads—where the video and audio mitment to a particular product class, brand, or con-components of a TV ad serve as the basis for print and sumption activity. Studying the Harley-Davidson andradio ads—has been shown to leverage existing com- Jeep brands, they explore various relationships thatmunication effects from TV ad exposure and more consumers could hold with the product/possession,strongly link them to the brand (Edell and Keller 1989, brand, firm, and/or other customers and use these1999). Cueing a TV ad with an explicitly linked radio to develop a measure of loyalty (McAlexander et al.or print ad can create similar or even enhanced pro- 2002).cessing outcomes of the radio or print ad that can A number of other researchers have explored word-substitute for additional TV ad exposures. of-mouth effects and their effect on brand evalu- Research Questions ations (e.g., Laczniak et al. 2001, Smith and Vogt 1. Under what circumstances is marketing integra- 1995). Moore et al. (2002) delineated how intergenera-tion more appropriately based on consistency (shar- tional influences affected intrafamily transfer of branding common brand meaning) versus complementarity equity in some product categories. Despite this atten-(presenting different brand meanings)? tion to interpersonal sources of influence and com- 2. How should brand-building activities change as munication, however, research has not systematicallydifferent audiences are targeted (e.g., consumers, dis- contrasted company-controlled and externally-driventributors, press, analysts, etc.)? To what extent can marketing activities.and should a firm tailor different messages to differ- Research Questionsent segments? When does confusion overwhelm the 1. How can brand communities and social net-benefits of more precise targeting? works best be modeled, cultivated, and influenced by 3. What are cost-effective vehicles for building marketers? What is the relative impact on consumersbrands? How do public relations, product placement, of verbal versus other types of communication (e.g.,and experiential marketing approaches compare to mere observation)?traditional advertising and promotion programs? 2. What is the relative impact of company actions, 4. What is the relative impact of third-party agents and evaluators, and customer conversationscommunications (e.g., competitors, rating services, (e.g., web sites) on brand equity? How are sequencesweb communications, or the government) on brand of interactions combined in the customer’s mind?equity? How can a firm utilize positive communica- Does being first or last have any real advantages?tions and counter negative ones? 3. How much do opinion leaders influence other 5. How do customer contact points (personal and consumers? To what extent is communication “ver-automated) influence brand equity? tical” (from expert to novice) versus “horizontal” (experts talking to each other)? Are there “anti-opinion 6. When changing the information communicated leaders,” i.e., people from whom others consciouslyabout a brand over time, how important is it for the try to behave differently? What is their impact?messages to follow a logical progression? 4. For socially conspicuous products, a major Combining Company-Controlled and External association influencing brand equity is other cus-Events. Marketers are increasingly embracing alter- tomers of the product. What is the relative impor-native forms of brand-building activities. In partic- tance of these associations versus company-controlledular, greater emphasis is being placed on “guerilla communications?marketing,” creating emotion-laden experiences, gen- 5. Does the Internet reduce the effects of branderating “buzz” among consumers, and creating online equity and its impact on consumer decision making?and real-world communities. To understand theunderpinnings of these activities, researchers study- Assessing Brand Performanceing interpersonal communication and influence have To manage brands properly, marketers should haveuncovered some important insights. a clear understanding of the equity in their brands— Muniz and O’Guinn (2000) defined “brand com- what makes them tick and what they are worth. Twomunities” as a specialized, nongeographically bound interesting subareas of this topic are the measurementcommunity based on a structured set of social rela- and valuation of brand equity at different levels—tionships among users of a brand. After studying the customer, product market, and financial market—andApple Macintosh, Ford Bronco, and Saab brands, they the relationship of customer equity to brand equity.note that, like other communities, a brand community Measuring Brand Equity. In recognition of theis marked by (1) a shared consciousness, (2) rituals value of brands as intangible assets, increased empha-and traditions, and (3) a sense of moral responsibility. sis has been placed on understanding how to build, Schouten and McAlexander (1995) defined a “sub- measure, and manage brand equity (Kapferer 2005;culture of consumption” as a distinctive subgroup of Keller 1993, 2003). There are three principal and
  6. 6. Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 2006 INFORMS 745distinct perspectives that have been taken by aca- Customer Level. The value of a brand—and thusdemics to study brand equity. its equity—is ultimately derived from the words and 1. Customer based. From the customer’s point of actions of consumers. Consumers decide with theirview, brand equity is part of the attraction to—or purchases, based on whatever factors they deemrepulsion from—a particular product from a par- important, which brands have more equity than oth-ticular company generated by the “nonobjective” ers (Villas-Boas 2004). Although the details of dif-part of the product offering, i.e., not by the prod- ferent approaches to measuring brand equity differ,uct attributes per se. While initially a brand may they tend to share a common core: All typically eitherbe synonymous with the product it makes, over implicitly or explicitly focus on brand-knowledgetime through advertising, usage experience, and other structures in the minds of consumers—individuals oractivities and influences it can develop a series of organizations—as the source or foundation of brandattachments and associations that exist over and equity.beyond the objective product. Importantly, brand To capture differences in brand-knowledge struc-equity can be built on attributes that have no inher- tures, a number of hierarchy of effects models haveent value (Broniarczyk and Gershoff 2003, Brown been put forth by consumer researchers throughand Carpenter 2000, Carpenter et al. 1994), although the years (e.g., AIDA, for Awareness-Interest-Desire-Meyvis and Janiszewski (2002) show irrelevant infor- Action). Customer-level brand equity can largely bemation can be counterproductive in consumer deci- captured by five aspects that form a hierarchy orsion making. chain, which are bottom (lowest level) to top (highest 2. Company based. From the company’s point of level) as follows:view, a strong brand serves many purposes, includ- (a) awareness (ranging from recognition to recall);ing making advertising and promotion more effec- (b) associations (encompassing tangible and intan-tive, helping secure distribution, insulating a product gible product or service considerations);from competition, and facilitating growth and expan- (c) attitude (ranging from acceptability to attrac- tion);sion into other product categories (Hoeffler and Keller (d) attachment (ranging from loyalty to addiction);2003). Brand equity from the company perspective is (e) activity (including purchase and consumptiontherefore the additional value (i.e., discounted cash frequency and involvement with the marketing pro-flow) that accrues to a firm because of the presence gram, other customers through word of mouth, etc.,of the brand name that would not accrue to an equiv- or the company).alent unbranded product. In economic terms, brand Many similar models exist (e.g., Aaker 1996, Kellerequity can be seen as the degree of “market inef- 2003). Several commercial versions are also avail-ficiency” that the firm is able to capture with its able (e.g., Young and Rubicam’s BrandAsset Valuatorbrands.2 (BAV), WPP’s Brand Z, and Research International’s 3. Financial based. From a financial market’s point of Equity Engine), although many focus largely on theview, brands are assets that, like plant and equipment, first three aspects above.can and frequently are bought and sold. The finan- There are several available research techniques tocial worth of a brand is therefore the price it brings measure brands at each of these five levels (Agrawalor could bring in the financial market. Presumably and Rao 1996). In addition, research has providedthis price reflects expectations about the discounted insight into how the value of a brand’s customer basevalue of future cash flows. In the absence of a market relates to stock-market value (Gupta et al. 2004). Intransaction, it can be estimated, albeit with great diffi- the more qualitative realm, a variety of alternativesculty (Ambler and Barwise 1998, Feldwick 1996), from exist for understanding the structure of associationsthe cost needed to establish a brand with equivalent that a customer has for a product. These “mentalstrength or as a residual in the model of the value of maps” rely on concepts such as metaphors (i.e., “It isa firm’s assets (Simon and Sullivan 1993).3 like a ____”) to develop deeper texture in representingComprehensive models of brand equity have been customer reactions to a brand (e.g., Zaltman 2003).developed in recent years to incorporate multiple per- Research Questionsspectives (Ambler 2004, Epstein and Westbrook 2001, 1. How much brand equity can be captured byKeller and Lehmann 2003, Srivastava et al. 1998). Each structured procedures (e.g., conjoint analysis or scan-of the three brand-equity measurement perspectives ner data modeling) and how much requires quali-has produced relevant work. tative understanding (e.g., via metaphors or mental maps)? Are there certain aspects of brand equity that2 See Erdem (1998a, b) for some economic perspectives on branding. can only be uncovered with qualitative research?3 See the special issue on brand valuation in the Journal of 2. Can the value of different qualitative aspects ofBrand Management (1998, 5(4)) for additional discussion and points brand equity be quantified? What is the relationshipof view. between qualitative and quantitative aspects?
  7. 7. Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities746 Marketing Science 25(6), pp. 740–759, © 2006 INFORMS 3. How “independent” versus redundant are the 3. How can brand equity be disentangled from itsnumerous customer-related brand-equity measures causes or source (e.g., product quality)? How canwhich have been studied? Is there a reduced set which brand and category equity be separated?is applicable to all products and/or all countries? 4. How can the impact of the brand be separatedWhat unique measures are relevant in different cat- from that of company market power, entry order, andegories, cultures, locations, or to different customer other possible determinants?groups? 5. What are the best approaches to tracking brand 4. Well-known brands provide a role in reducing performance? How frequently should it be measured?risk: Not only are brands signals of quality (both in 6. How much explanatory power does brandterms of mean and variance), but they also provide equity have after accounting for market-share effectsthe “deep pockets” needed to rectify a product fail- (Uncles et al. 1995)?ure. To what extent is increased confidence in deci- Financial-Market Level. A different approach to mea-sion making a key or even critical factor of brands suring brand equity is based on financial market per-and brand equity; i.e., are standard deviations more formance (Amir and Lev 1996, Barth et al. 1998). Oneimportant than means? measure that has been proposed uses the compo- Product-Market Level. A number of approaches have nent of market value unexplained by financial assetsbeen developed to assess the impact of brand equity and results (i.e., profits). Using Tobin’s Q (the mar-in the product market. These include measures ket value of assets divided by their replacement valueof price premiums, increased advertising elasticity, as estimated by book value) as a proxy of branddecreased sensitivity to competitors’ prices, and the equity, Lindenberg and Ross (1981) found that con-ability to secure and maintain distribution through sumer goods companies such as Coca-Cola, Pepsico,channels (Hoeffler and Keller 2003). Kellogg’s, and General Foods had Tobin Q’s greater Several studies have demonstrated that leading than 2, suggesting that these companies had con- siderable intangible value. On the other hand, morebrands can command large price differences (Simon commodity-like manufacturers such as metal produc-1979, Agrawal 1996, Park and Srinivasan 1994, ers and paper products companies had Tobin Q’s ofSethuraman 1996) and are more immune to price about 1.increases (Sivakumar and Raj 1997). Lower levels Simon and Sullivan (1993) decomposed firm valueof price sensitivity have been found for households into tangible and intangible components: Tangiblethat are more loyal (Krishnamurthi and Raj 1991). components reflected replacement costs and includedAilawadi et al. (2003) proposed that the revenue pre- assets such as plant and equipment and net receiv-mium a brand commands vis-à-vis an unbranded ables; intangible components were broken down intoproduct is a simple useful measure of brand equity industry-wide, cost, and brand factors. The brandand showed how it responds to brand actions. They factors were derived from a market share equationcontend that neither the sales premium nor the using an instrumental variables approach (i.e., brandprice premium alone captures the increased demand value was determined by order of entry and adver-attributable to a brand. tising). As a percent of replacement values, brand Advertising may play a role in decreasing price equity ranged from a low of essentially zero for cat-sensitivity (Kanetkar et al. 1992). Consumers who egories such as paper and allied products; petroleumare highly loyal to a brand have been shown to and coal; stone, glass, and coal; and primary and fab-increase purchases when advertising for the brand ricated metals to as much as 61% for apparel, 58%increased (Raj 1982, Hsu and Liu 2000). Research sug- for printing and publishing, and 46% for tobacco.gests that stores are more likely to feature well-known Firms for which brand value exceeded replacementbrands if they convey a high quality image (Lal and cost included Dreyer’s Ice Cream, Tootsie Roll, andNarasimhan 1996). Fader and Schmittlein (1993) pro- Smucker.posed that differences in retail availability may be a Another approach to assessing the financial valuekey component of the higher repeat-purchase rates for of a brand involves taking customer mindset mea-higher-share brands. sures and relating them to stock-market values. This Research Questions approach is used by Stern Stewart’s Brand Economics 1. What are the advantages of residual versus which link Young & Rubicam’s BrandAsset Valuator,direct measures of the effect of the brand on a survey-based measure of brand strength, to eco-marketing-program effectiveness? nomic value added (EVA), a financial performance 2. How do you assess and identify the “option measure. Along those lines, Aaker and Jacobsonvalue” of the extension potential of a brand? What (1994) relate yearly stock returns for 34 companiesare the “cost savings” that result from higher brand during 1989 to 1992 to unanticipated changes in ROI,equity in terms of advertising effectiveness, etc.? brand equity, and brand salience. Using EquiTrend’s
  8. 8. Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 2006 INFORMS 747perceived quality rating as a proxy for brand equity, Research Questionsthey find that changes in quality and thus equity had 1. What are the links between customer-market,a significant effect over and above that of changes product-market, and financial-market level measuresin ROI. Firms who experienced the largest gains in of brand equity? For example, does customer-levelbrand equity saw their stock return average 30%; con- equity lead to financial-market equity by generat-versely, those firms with the largest losses in brand ing additional cash flow or by directly influencingequity saw stock return average a negative 10%. Inter- investor decisions?estingly, other results suggest that there is a bigger 2. How can the causal impact of brand equity onimprovement when the changes in quality percep- financial market performance be established given thetions occur among heavy users, a result consistent large number of other factors that drive stock price?with suggestions that retention (impacting current 3. Are large values of brand equity limited to con-customers) may often be the best way to increase cus- sumer and hedonic goods or can they also existtomer and, hence, firm value (Thomas et al. 2004). for business-to-business and other high-involvement, Using data for firms in the computer industry utilitarian products?in the 1990s, Aaker and Jacobson (2001) found that 4. Should brand equity be reported on the balancechanges in brand attitude were associated contem- sheet? If so, how?poraneously with stock return and led accounting 5. Which are forward- versus backward-lookingfinancial performance. Awareness that did not trans- brand-equity measures?late into more positive attitudes, however, did littleto the stock price. Adopting an event study method- The Marketing-Mix and Brand Equity. Marketing-ology, Lane and Jacobson (1995) showed that the mix modeling has increased in popularity with indus-stock market response to brand extension announce- try and academics (Gatignon 1993, Hanssens et al.ments depended interactively and nonmonotonically 1998). Considerable research has examined the effec-on brand attitude and familiarity: The stock mar- tiveness of different elements of the marketing mix.ket responded most favorably to extensions of either For example, numerous studies have examined thehigh esteem, high-familiarity brands or those of low short-term and long-term effects of advertising andesteem, low-familiarity brands. Mizik and Jacobson promotion (e.g., Ailawadi et al. 2001, Anderson and(2003) examined the relative importance of value- Simester 2004, Dekimpe and Hanssens 1999, Melaappropriation activities (i.e., extracting profits in the et al. 1997). This research often looks at different out-marketplace via advertising and promotion) versus comes and indicators of marketing effectiveness. Forvalue-creating activities (i.e., through R&D) on the example, Pauwels et al. (2002) found that price pro-stock market. motion has a strong effect on category purchase inci- In an event study of 58 firms that changed their dence for a storable product but a correspondinglynames in the 1980s, Horsky and Swyngedouw (1987) larger impact on brand choice for perishable products.found that, for most of the firms, name changes were Although these research streams have pro-associated with improved performance. The greatest vided considerable insight, they have not typicallyimprovement tended to occur in firms that produced addressed the full breadth of brand equity dimen-industrial goods and whose performance prior to the sions. In particular, it is rare that measures ofchange was relatively poor. Not all changes, however, customer mindset are introduced as possible mediat-were successful. They interpreted the act of a name ing or moderating variables in analyzing marketingchange as a signal that other measures to improve effectiveness.performance—e.g., changes in product offerings and Research Questionsorganizational changes—will be seriously and suc- 1. How stable is brand equity? Does the stabilitycessfully undertaken. depend on the marketing driver involved, e.g., an Mahajan et al. (1994) suggest how to assess the ad versus a personal experience?level of brand equity in the context of firm acquisi- 2. How does the effectiveness of marketing driverstions. Kerin and Sethuraman (1998) also have exam- of brand equity change over time? When are emo-ined the link between brand value and stock value. In tional drivers more important: early on or as a marketthe brand strategy arena, Rao et al. (2004) examined matures? Are emotional drivers more relevant to cor-the question of whether a “branded house” strategy porate brands and rational drivers more relevant towith a corporate brand as an umbrella was associated product brands?with higher stock-market returns than a multiple- 3. To what extent can and should a companybrand “house of brands” strategy. In their data, a try to influence (versus respond to) what the keycorporate branding strategy produced higher average drivers are?return than a multibrand strategy, perhaps to com-pensate for the greater risk due to the nondiversifica- Relationship of Brand Equity to Customer Equity.tion involved. An important emerging line of research concerns
  9. 9. Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities748 Marketing Science 25(6), pp. 740–759, © 2006 INFORMScustomer equity and the antecedents and conse- based on the quality of the original brand, the fitquences of developing strong ties to customers (Rust between the parent and extension categories, and theet al. 2000). A number of researchers have noticed interaction of the two, although cultural differencesthe relationship between the brand-management influenced the relative importance attached to theseand customer-management perspectives (e.g., Ambler model components. Studies have shown how well-et al. 2002). Indeed, under one set of assumptions known and well-regarded brands can extend morethe value of a customer to the firm (i.e., customer successfully (Aaker and Keller 1990, Bottomley andequity) can be shown algebraically to be the sum of Doyle 1996) and into more diverse categories (Kellerthe profit from selling equivalent generic products and Aaker 1992, Rangaswamy et al. 1993). In addition,and the additional value from selling branded goods the amount of brand equity has been shown to be(i.e., brand equity). correlated with the highest- or lowest-quality mem- Research Questions ber in the product line for vertical product extensions 1. Does brand equity management simply reflect (Randall et al. 1998). Brands with varied product cat-an aggregate view of customer equity management? egory associations developed through past extensionsHow do concepts such as customer lifetime value have been shown to be especially extendible (Dacinand CRM relate to brand equity? How can they be and Smith 1994, Keller and Aaker 1992, Sheinin andintegrated? Schmitt 1994). As a result, introductory marketing 2. How closely related are measures of brand programs for extensions from an established brandequity and customer equity (e.g., loyalty and share of can be more efficient (Erdem and Sun 2002, Smithwallet or requirements, brand relationship and cus- 1992, Smith and Park 1992).tomer retention)? A number of other factors also come into play to 3. How can a firm balance a product-driven brand influence extension success, such as consumer knowl-focus with a customer-driven CRM one? Which strate- edge of the parent and extension categories (e.g.,gies are most effective? Under what circumstances? Moreau et al. 2001) and characteristics of the con- sumer and extension marketing program (e.g., BaroneBrands as Growth Platforms and Miniard 2002, Maoz and Tybout 2002, ZhangNo problem is more critical to CEOs than generat- and Sood 2002). Kirmani et al. (1999) found evidenceing profitable growth. Brands grow primarily through of an ownership effect whereby current owners gen-product development (line and category extensions) erally had more favorable responses to brand lineand market development (new channels and geo- extensions.graphic markets). Important subtopics here include One oft-cited concern with brand extensions is thatnew-product and brand-extension strategies and their a failed brand extension could hurt (dilute) the par-effects on brand equity. ent brand in various ways. Interestingly, academic New Products and Brand Extensions. Brand exten- research has found that parent brands generally aresions are one of the most heavily-researched and not particularly vulnerable to failed brand extensions.influential areas in marketing (Czellar 2003). Mar- An unsuccessful brand extension potentially damagesketing academics have played an important role in a parent brand only when there is a high degreeidentifying key theoretical and managerial issues and of similarity or “fit” involved—e.g., in the case of aproviding insights and guidance. failed line extension in the same category—and when Research has shown that extension success depends consumers experience inferior product performancelargely on consumers’ perceptions of fit between a directly (Ahluwalia and Gürhan-Cali 2000, Gürhan-new extension and parent brand (Aaker and Keller Canli and Maheswaran 1998, Keller and Aaker 1992,1990; but see Klink and Smith 2001, van Osselaer Loken and Roedder John 1993, Milberg et al. 1997,and Alba 2003). There are a number of bases of fit— Roedder John et al. 1998, Romeo 1991).virtually any brand association is a potential basis— Several other factors also influence the extent ofbut two key bases are competence (attribute) and damage to a parent brand from an unsuccessfulimage (Batra et al. 1993). Research has also shown brand extension. The more involved the consumer isthat positively evaluated symbolic associations may with the extension decision (e.g., if they own or usebe the basis of extension evaluations (Reddy et al. the parent brand), the more likely it is that harm-1994, Park et al. 1991), even if overall brand attitude ful dilution effects will occur (Kirmani et al. 1999).itself is not necessarily high (Broniarczyk and Alba Importantly, research has shown that a subbranding1994). One key conclusion is that consumers need to strategy, where an extension is given another namesee the proposed extension as making sense. in addition to the parent brand (e.g., Courtyard by Based on a meta-analysis of seven studies using Marriott), can effectively shield a parent brand from131 different brand extensions, Bottomley and Holden dilution from a failed similar extension (Keller and(2001) concluded that brand extension evaluations are Sood 2004, Milberg et al. 1997).
  10. 10. Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 2006 INFORMS 749 Research has also shown that extensions can cre- to maintain brand exclusivity. A subbranding strat-ate positive feedback effects to the parent brand egy, however, protected owners’ parent-brand atti-(Balachander and Ghose 2003). For instance, brand tudes from dilution.extensions strengthened parent brand associations In terms of multiple brand extensions, Keller and(Morrin 1999) and “flagship brands” were highly Aaker (1992) showed that by taking “little steps,” i.e.,resistant to dilution or other potential negative effects by introducing a series of closely related but increas-due to unfavorable experiences with an extension ingly distant extensions, it was possible for a brand to(Roedder John et al. 1998, Sheinin 2000). ultimately enter product categories that would have Research Questions been much more difficult, or perhaps even impossi- 1. How can the long-term new product potential ble, to have entered directly (Dawar and Andersonof a brand be assessed? What is the optimal product 1994, Jap 1993, Meyvis and Janiszewski 2004).breadth for a brand franchise? Joiner and Loken (1998), in a demonstration of the 2. How should a brand be built and managed as inclusion effect in a brand extension setting, showeda growth platform? Which kinds of brand associa- that consumers often generalized possession of antions are most beneficial or detrimental for future attribute from a specific category (e.g., Sony televi-brand growth? What kind of brand associations facil- sions) to a more general category (e.g., all Sony prod-itate versus inhibit the introduction of line and brand ucts) more readily than they generalized to anotherextensions? specific category (e.g., Sony VCRs). Research has 3. What should be built into a pioneer brand to shown that family-brand evaluations depend on theretard future competition? expected variability of individual product quality and 4. For new-to-the-world products, what should be attribute uniqueness (Gürhan-Canli 2003; see alsothe relative emphasis on building the brand versus Swaminathan et al. 2001).establishing and growing the category? More gener- Research has also shown that a subbranding strat-ally, what should be the brand versus product focus egy can enhance extension evaluations, especiallyover the product life cycle? when the extension is farther removed from the prod- uct category and less similar in fit (Keller and SoodStrategically Managing the Brand 2004, Milberg et al. 1997, Sheinin 1998). A subbrandIn many firms, the CEO is effectively the chief brand can also protect the parent brand from unwantedofficer (CBO) as well. Regardless of who (if any- negative feedback (Milberg et al. 1997, Janiszewskione) is in charge of managing the brand, several and van Osselaer 2000, Kirmani et al. 1999), butgeneral strategic issues arise: the optimal design of only in certain circumstances, e.g., if the subbrandbrand architecture, the effects of co-branding and consists of a meaningful individual brand that pre-brand alliances, and cross cultural and global brand- cedes the family brand, e.g., Courtyard by Marriotting strategies. (Keller and Sood 2004). Wänke et al. (1998) showed Brand Architecture. Brand architecture has been how subbranding strategy could help set consumerstudied in the context of line extensions, vertical expectations.extensions, multiple brand extensions, subbrands, Bergen et al. (1996) studied branded variants—and brand portfolios (Aaker 2004). Several researchers the various models that manufacturers offer differ-have examined characteristics of successful line exten- ent retailers (see also Shugan 1989). They showedsions (Andrews and Low 1998, Putsis and Bayus 2001, that as branded variants increased, retailers wereReddy et al. 1994). In the context of fast-moving pack- more inclined to carry the branded product and pro-aged goods, Cohen et al. (1997) developed a decision vide greater retail service support. Other research hassupport system to evaluate the financial prospects of shown how brand portfolios can increase loyalty topotential new line extensions. multiproduct firms (Anand and Shachar 2004). Kumar Although many strategic recommendations have (2003) argues that companies can rationalize theirbeen offered concerning “vertical extensions”—exten- brand portfolios to both serve customers better andsions into lower or higher price points (e.g., Aaker maximize profits (see also Broniarczyk et al. 1998).1997)—relatively little academic research has been Research Questionsconducted to provide support for them (see Randall 1. How do product brands impact the equity of cor-et al. 1998 for an exception). Kirmani et al. (1999) porate brands (and vice versa)?found that owners had more favorable responses 2. How can the interplay and flow of equitythan nonowners to upward and downward stretches between product and corporate brands be measuredof nonprestige brands (e.g., Acura) and to upward (“ladder up” versus “waterfall down”)?stretches of prestige brands (e.g., Calvin Klein and 3. Can and should line extension proliferation beBMW). Downward stretches of prestige brands, how- controlled? What are the design criteria for the opti-ever, did not work well because of owners’ desires mal brand portfolio?
  11. 11. Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities750 Marketing Science 25(6), pp. 740–759, © 2006 INFORMS 4. Does it matter who and where in the organiza- concentrated on when marketers should standard-tion controls the brand? ize versus customize their global marketing programs 5. How should a company deal with differences (e.g., Gatignon and Vanden Abeele 1995, Samiee and(heterogeneity) in terms of what consumers think Roth 1992, Szymanski et al. 1993).about and want from a brand? Research has also examined cultural and linguis- Co-Branding and Brand Alliances. Brand alli- tic aspects of branding, e.g., showing how Chineseances—where two brands are combined in some way versus English brand names differ in terms of visualas part of a product or some other aspect of the mar- versus verbal representations (Schmitt et al. 1994, Panketing program—come in all forms (Rao 1997, Rao and Schmitt 1996, Zhang and Schmitt 2001). From aet al. 1999, Shocker et al. 1994) and have become brand building standpoint, Steenkamp et al. (2003)increasingly prevalent. Park et al. (1996) compared show how perceived brand globalness creates brandco-brands to the notion of “conceptual combinations” value.in psychology and showed how carefully selected Research Questionsbrands could be combined to overcome potential 1. How do consumer schemas and accepted prac-problems of negatively correlated attributes (e.g., rich tices for branding strategies and activities vary acrosstaste and low calories). countries? Simonin and Ruth (1998) found that consumers’ 2. What is the optimal degree of localization forattitudes toward a brand alliance could influence branding and marketing communications? To whatsubsequent impressions of each partner’s brands extent should both the marketing programs for a(i.e., spillover effects existed), but these effects also brand and the product itself (e.g., level of sweetnessdepended on other factors such as product “fit” for Coca-Cola) be varied across locations?or compatibility and brand “fit” or image con- 3. How does global brand management vary bygruity. Desai and Keller (2002) found that although product life-cycle stage? Should it be mandateda co-branded ingredient facilitated initial expansion or encouraged by sharing best practices across theacceptance, a self-branded ingredient could lead to company?more favorable long-run extension evaluations. In 4. To what extent does country image (or equity)other words, borrowing equity from another brand impact the equity of brands from that country?does not necessarily build equity for the parent brand Branding and Social Welfare. Brands would exist(see also Janiszewski and van Osselaer 2000). even if no money were spent on advertising and Research Questions promotion for products. Customers would find some 1. What is the proper executional approach to com- distinguishing characteristics (name, color, shape) tobining brands? What characterizes effective imple- identify products or services that had served themmentation? well and use them to simplify (make more efficient) 2. What are the relative implications of formal future choices. Moreover, as satisficers, customersalliances, co-branding, and ingredient branding on are slow to update performance improvements (orcustomer reactions and company profits? decreases) in their current or other alternative choices. 3. When one brand buys another or is merged with The result, at least in the short run, is market inef-it, how should it be determined whether or not one ficiency in the physical attribute product space. Inbrand should dominate? essence, market inefficiency (see Hjorth-Anderson 4. Does a brand carry the same value after it is 1984) can be seen as the same as brand equity, raisingacquired? several interesting questions. 5. How much brand equity is derived from sur- Research Questionsroundings (e.g., retail stores, distributors) and how 1. Do brands create value, provide value, or reducemuch does a brand contribute to the equity of these value for customers?surroundings? 2. Are there categories of goods for which large 6. What are the roles of different brands in pro- brand equities are acceptable (e.g., luxury goods mar-viding “complete solutions” to consumers? How are keted to affluent customers), and others where they“lead brands” best determined? are not (e.g., pharmaceuticals)? Cross-Cultural and Global Branding. Branding is 3. Is market inefficiency and the creation of brandincreasingly being conducted on a global landscape. equity desirable or undesirable in terms of its effectA number of issues emerge in attempting to build on the overall economy?a global brand. Levitt (1983) has argued that com- 4. How should marketers respond to criticisms ofpanies needed to learn to operate as if the world brands as being overpriced? As creating needs versuswere one large market—ignoring superficial regional satisfying real needs? How about issues of productand national differences. Much research, however, has failure or safety?
  12. 12. Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 2006 INFORMS 751 5. What is the role of brands in acquiring and the last brand purchased (otherwise, Vj and Sj are notretaining employees? Do brands positively impact identifiable).employee effort and hence customer satisfaction or Much of the previous research that incorporateswelfare? brands has focused on assessing the impact by mod- eling consumer choice with a specific brand term (Srinivasan 1979). The rationale is a view that brandImplications for Choice Modeling equity is what remains of consumer preferences andThe previous discussion captures some of the research choices after accounting for physical product effects.progress and gaps in the study of branding. The dis- Several approaches have been suggested:cussion also has some specific implications for incor- • Kamakura and Russell (1993) proposed aporating branding concepts into choice models. scanner-based measure of brand equity that attempts To demonstrate how brands influence consumer to explain the choices observed by a panel of con-choice through their value (utility), we contrast the sumers as a function of the store environment (actualstylized “classic” microeconomic view of utility and shelf prices, sales promotions, displays, etc.), thechoice (Lancaster 1966) with a view which explicitly physical characteristics of available brands, and aand/or implicitly encompasses the impact of brands. residual term dubbed brand equity (here, Vj .In the classic view, the value of brand j is the sum of • Swait et al. (1993) proposed a related approachits I (objective) attributes, net of price, as follows: for measuring brand equity which utilizes choice experiments that account for brand name, product VBj = Bi Xji − Pj (1) i=1 I attributes, brand image, and differences in consumer sociodemographic characteristics and brand usage. Essentially, at the customer level, a brand is the They define the equalization price as the price thatlens through which the words and actions of a com- equates the utility of a brand to the utilities that couldpany, its competitors, and the environment in general be attributed to a brand in the category where noare converted to thoughts, feelings, images, beliefs, brand differentiation occurred.perceptions, and attitudes, etc., about a product (or • Park and Srinivasan (1994) proposed a method-family of products). Much of the value of a branded ology for measuring brand equity based on theproduct is in these subjectively determined compo- multiattribute attitude model. The attribute-basednents. The manner by which consumers transform component of brand equity is the difference betweenobjective product value to create additional (intangi- subjectively perceived attribute values and objectively ∗ble) value leads to four components of brand value: measured attribute values, essentially the Xji − Xji ∗ • Biased Perceptions Xji − Xji , i.e., the extent to terms in (2) (i.e., the “halo effect,” Beckwith andwhich specific product attribute perceptions are influ- Lehmann 1975, 1976). The nonattribute-based com-enced by the halo effect (Beckwith and Lehmann ponent of brand equity is the difference between1975). subjectively perceived attribute values and overall • Image Associations Zjk , i.e., nonproduct-related preference and reflects the consumer’s configural rep-attribute beliefs such as “friendly” or “stylish.” resentation of a brand that goes beyond the assess- • Incremental Value Vj , an additive constant asso- ment of the utility of individual product attributes.ciated with the brand name that is not related to any • Dillon et al. (2001) presented a model for decom-particular attribute or benefit. posing ratings of a brand on an attribute into two • Inertia Value Sj , the value to consumers of sim- components: (1) brand-specific associations (i.e., fea-ply choosing the same option rather than spend- tures, attributes, or benefits that consumers link to aing effort to consider others, e.g., due to switching brand), and (2) general brand impressions (i.e., overallcosts, or the confidence (less uncertainty) of a known impressions based on a more holistic view of a brand,alternative. here the Zjk s).The value of a branded product (VBP) can be seen One clear and important implication of the aboveas the sum of the objective value of a product as well discussion is that the value of a brand is greateras the four components of brand value listed above: than either its additive (main effect) incremental value (e.g., in a conjoint study or logit model) or its impact ∗ VBj = i Xji − Pj + i Xji − Xji on perceptions, and it needs to be separated from i=1 I i=1 I state dependence. + Ck Zjk + Vj + Sj (2) k=1 K Influences on Brands Brands are made, not born. The process of their con-Note that Sj is not strictly a brand term but rather struction is complex. From a manufacturer’s point ofreflects state dependence and can be modeled using view there is a reduced form, “stimulus-response”
  13. 13. Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities752 Marketing Science 25(6), pp. 740–759, © 2006 INFORMSstyle simplicity to it: (1) the manufacturer takes at least include a brand main effect, brand interactionactions (e.g., the marketing mix) and that leads to effects, and the impact of competition. This is obvi-(2) customer mental responses towards the brand ously a very complex model so that simplifications(perceptions, beliefs, attitudes, and so on). These per- are needed. For example, we can assume brand equityceptions (and the resulting willingness to pay) in turn modifies the impact of marketing activities throughlead to (3) customer behavior in the product market a varying parameter formulation such as Djn = Dn +(e.g., sales), which in turn generates (4) financial value wVj . It is also difficult, of course, to separate thein general and stock market and market capitalization impact of a brand from its unique attributes orin particular. attributes not included in the analysis. This separabil- This framework or value chain is a useful basic con- ity problem makes it hard to identify whether appar-ceptualization. Still, it obscures some important com- ent brand equity is due to brand image or attributeplexities. The first is that a brand’s position is heavily differences; attributing it all to the attributes mayinfluenced by others, e.g., competitors, governmental induce omitted variable bias whereas attributing it allbodies, and interest groups, as well as by actions of to the brand may overstate brand impact.employees and the identity and behavior of customers Further levels of complication are also possible,of the brand. Analogous to the customer level, high although rarely considered. For example, the decisionlevels of brand equity reduce price sensitivity and of channels to stock and support a brand depends onmake advertising more effective. Perhaps most impor- how much revenue it will generate which, in turn,tant, it ensures distribution in channels with limited depends in part on brand equity (e.g., see Besankoselection (e.g., convenience stores or small distribu- et al. 2005). Similarly, brand equity can have indi-tors), making it available in more locations. Greater rect cost effects through its impact on volume (i.e.,availability may in turn impact (signal) perceptions: economies of scale) or by providing the confidence“If a brand is widely carried and displayed, then it to suppliers for them to commit resources to “part-must be good.” nering” with a firm and supporting its product. It is Thus, another complexity is that the impact of also possible that brand equity influences competitivewhat the brand does depends on the brand itself actions and reactions. For example, will a competitor(i.e., is endogenous), particularly in terms of its over- be more or less likely to cut price when faced with aall strength. Considerable evidence exists that strong high equity competitor who is to some degree insu-brands have lower price elasticity with respect to their lated from the impact of their price cuts? While weown price increases or price decreases of their com- have no specific answers to these issues, these areaspetitors. Similarly, the advertising elasticity of strong are promising and underdeveloped avenues for futurebrands may be larger. This leads to different decisions. modeling research. Consider the impact of advertising on one compo- Allowing mix elements to have different, compet-nent of brand equity—image associations. Specifically, itor-specific effects, greatly complicates modeling byconsider a simple model of how a specific image asso- introducing more parameters than can be effectively estimated. One issue, therefore, is whether it is worth-ciation k is related to a specific marketing program while trying to capture such complexity, i.e., byactivity Mn for brand j: adding the large number of possible interaction (mod- Zjkt = Zjkt−1 + Djn ∗ Mjn (3) erating) effects. Said differently, for some purposes, is a “wrong” but simple model likely to outperformFor a strong brand, the marginal impact of its adver- an extensive but likely misspecified more complete/tising Djn may be greater than for a weak brand. complex model? Another related issue, particularlyThus, strong brand j can spend less than a weak relevant for modelers, is how to capture such com-brand and still improve its image. More generally, the plexity in structural models of brand evaluation andimage of a brand depends on the N marketing activi- competition. For purposes of this review, we leaveties of the various R competitors as well as main and these as an area for future analysis.interaction effects of its own activities: More generally, there may be a “virtuous circle.” As brands develop positive brand equity, it becomes Zjkt = Zjkt−1 + Djn Mjn easier for them to develop further (and harder for n=1 N competitors to compete with them). The obvious + Djnp Mjn Mjp implication is that there are increasing returns to n=1 N p=n+1 N scale to building a brand, at least up to a point. The research question then becomes when, if ever, + Djrn Mrn (4) and under what conditions additional brand build- r=1 R n=1 N ing becomes less efficient (e.g., see Naik et al. 2005). The multiple consequences of brand equity mean Combined with the earlier discussion on the multiplethat an aggregate product-market level model should ways a brand manifests its extra value, this suggests
  14. 14. Keller and Lehmann: Brands and Branding: Research Findings and Future PrioritiesMarketing Science 25(6), pp. 740–759, © 2006 INFORMS 753an important measurement issue: how to capture its their own current market (as in Equation (4)) but sototal value and how to determine the relative contri- they provide a basis for both expanding the existingbution of its multiple sources. market and the option of entering others. Finally, it should be noted that the topics concern-ing brand-management decisions discussed above A Systems Model of Brandhave direct implications for these modeling formu- Antecedents and Consequenceslations. Developing brand positioning relates to how A number of brand dashboards have been devel-marketing activities (Ms) lead to the formation of oped by firms which capture, but rarely link, manyattribute perceptions (Xs) and image associations aspects of brand equity and performance. For brand-(Zs) in Equation (2). Integrating brand marketing ing research to be scientifically rigorous, it is impor-addresses, in part, consistency issues and is implic- tant to develop a comprehensive model of how branditly related to the interaction terms among market- equity operates and to develop estimates of the var-ing activities in general and making their impact ious cause-and-effect links within it. To that end,positive in particular. Assessing brand performance we expand on the notion of a “brand value chain”relates to metrics which both measure the elements (Keller and Lehmann 2003) discussed earlier. Theof Equation (4) and their consequences in the prod- chain focuses on the following four major stages (seeuct and financial markets. Brands as growth platforms Figure 1):addresses the key strategic issues of how to orches- 1. What companies do. Marketing programs, as welltrate efforts over time to develop brands not just for as other company actions, form the controllableFigure 1 A Systems Model of Brand Antecedents and Consequences Company actions Strategy Programs Direction Quality Specifics: Quality type, budget Partners’ actions Competitor’s (channels, actions What customers think employees) and feel about a brand Industry/environmental conditions Awareness Associations Attitude Attachment Activity (Satisfaction) What customers do about a brand Financial market impact
  15. 15. Keller and Lehmann: Brands and Branding: Research Findings and Future Priorities754 Marketing Science 25(6), pp. 740–759, © 2006 INFORMSantecedents to the brand value chain. Importantly, These expectations should increase the number ofthese activities can be characterized along two sepa- people willing to buy the brand extension initiallyrate dimensions: quantitative factors such as the type (p, the coefficient of innovation) and the speed of dif-and amount of marketing expenditures (e.g., dollars fusion of the extension through word of mouth (q, thespent on media advertising), and qualitative factors coefficient of imitation) since it will seem less risky tosuch as the clarity, relevance, distinctiveness, and con- those consumers who wait for others to buy it first.sistency of the marketing program, both over time A stronger brand can more easily gain wider distribu-and across marketing activities. tion which will also lead to faster trial among innova- 2. What customers think and feel. Customer mindset tors (in effect, makes the market potential m larger).consists of the “Five As” discussed above. Impor- Thus, a reasonable prediction is that stronger brandstantly, there are feedback effects here, as demon- will, ceteris paribus, have both faster diffusion andstrated by the “halo effect” where brand attitudes greater market potential.affect perceptions of brand associations (Beckwith To move branding toward becoming a rigorousand Lehmann 1975, 1976). Moreover, what customers science, a general model similar to Figure 1 needsthink and feel about brands is obviously not under the to be tested and calibrated. Currently, little progresssole, or often even primary, control of the company. has been made toward estimating such a compre-Individual customer characteristics as well as compe- hensive model, or even a reduced form version oftition and the rest of the environment help shape what the model, such as marketing activities → product-is thought of the brand, e.g., by influencing expec- market results → financial impact. As noted above,tations (Boulding et al. 1993). Both personal expe- there are certainly scattered empirical generalizations.rience (feedback from use and product satisfaction) For example, we know increasing ad budgets has lit-and the experience of others (through word of mouth tle impact on current sales unless either the prod-and “expert” ratings) also determine what a customer uct or the use that is promoted is new (Lodish et al.thinks of a brand. 1995, Assmus et al. 1984). What is badly needed are: 3. What customers do. The primary payoff from cus- (1) metaanalyses that combine partial tests of modeltomer thoughts and feelings is the purchases that they components (i.e., only relating a subset of variables)make. This product-market result is what generates into an overall estimate of the average links andrevenue, share, and other metrics commonly used key contingencies in the model, and (2) comprehen-to evaluate the effectiveness of marketing programs. sive studies that systematically examine the model, orOf course, other things customers do, especially word at least a large part of it, in its entirety.of mouth, impact future product-market results andneed to be considered in any comprehensive model. 4. How financial markets react. For a publicly held Conclusioncompany, stock price and market capitalization, as Branding and brand management has clearly becomewell as related measures such as Tobin’s Q, are critical an important management priority for all types ofmetrics. In essence, these measures are the ultimate organizations. Academic research has covered a num-bottom line. As such, they are relevant at the CFO and ber of different topics and conducted a number ofCEO level, unlike most marketing metrics which are different studies that have collectively advanced ourat the customer level or product-market level. Impor- understanding of brands. Table 1 summarizes sometantly, stock price is impacted by a number of other of the generalizations that have emerged from thesevariables such as the growth potential of the indus- research studies that were reviewed in this paper.try as a whole, general economic trends, and stock- To put the academic literature in marketing in somemarket dynamics, which need to be controlled for in perspective, it could be argued that there has beenassessing the financial value of brands. somewhat of a preoccupation with brand extensions The overall model is thus conceptually fairly sim- and some of the processes that lead to the develop-ple (i.e., it has only four main components), but in ment of brand equity. By contrast, there has been rel-practice is both complicated (to account for all the atively limited effort directed toward exploring theinfluences and feedback effects) and stochastic. financial, legal, and social impacts of brands. In terms The model reflects and accounts for a number of of methodology, considerable effort has been devotedmarketing principles. Consider the impact of a brand to controlled experimentation (often with student sub-extension in the context of the Bass model of new jects), although some work has focused on choiceproduct diffusion. Assuming there is some level of modeling of scanner data. Little integration of thesefit with a parent brand which has positive equity, a two streams with each other or the qualitative workbrand extension has advantages in terms of assumed on branding has appeared.product quality and the willingness of the firm to Although much progress has been made, especiallystand behind the product in the event of problems. in the last decade or so, a number of important

×