Brand and Line Extensions: Role of Brand Associations in Choosing the Better Alternative

ABSTRACT - Brand and line extensions have become important means of growth for companies that have strong brand names. There may be instances where brand managers have to choose between brand and line extension as the primary form of growth for the brand. In order to be able to choose the better alternative, brand managers need guidelines which relate consumers perceptions of the brands, in the form of brand associations, to decisions on line or brand extension. This paper makes an attempt to develop propositions and formulate guidelines which may be used to make such decisions.

Citation:

Sridhar Samu (1994) ,"Brand and Line Extensions: Role of Brand Associations in Choosing the Better Alternative", in AP - Asia Pacific Advances in Consumer Research Volume 1, eds. Joseph A. Cote and Siew Meng Leong, Provo, UT : Association for Consumer Research, Pages: 311-315.

Asia Pacific Advances in Consumer Research Volume 1, 1994      Pages 311-315

BRAND AND LINE EXTENSIONS: ROLE OF BRAND ASSOCIATIONS IN CHOOSING THE BETTER ALTERNATIVE

Sridhar Samu, Indiana University

ABSTRACT -

Brand and line extensions have become important means of growth for companies that have strong brand names. There may be instances where brand managers have to choose between brand and line extension as the primary form of growth for the brand. In order to be able to choose the better alternative, brand managers need guidelines which relate consumers perceptions of the brands, in the form of brand associations, to decisions on line or brand extension. This paper makes an attempt to develop propositions and formulate guidelines which may be used to make such decisions.

INTRODUCTION

New product introductions have inherent risks in terms of cost and other lost opportunities. Data from previous research show that new brands have to deliver sales of more than $50 million to be successful and that the true cost of establishing a new brand could be as much as three times this (Tauber 1988). Consumer product manufacturers have used various strategies to reduce such risks and one important strategy that has been found to work is the use of brand names that have proven successful (Tauber 1988; Aaker and Keller 1990). Using successful brand names may ensure that a competitive advantage is gained for the new product even before it is introduced. Two different approaches utilizing existing brand names commonly used are line and brand extensions. Line extension is the process where a current brand name is used to enter a new market segment or new usage situation for an existing market segment in the same product class (Aaker and Keller, 1990). An example would be the case where Diet Pepsi was launched to target a different market or a different usage than the Pepsi drinkers market. This may be perceived as a natural extension by consumers and would match well with existing distribution outlets and trade policies.

Brand extension is the process through which manufacturers use a current brand name to enter a different product class. Examples of this would be Zenith computers, and AT & T credit cards. The major similarities between the parent brand and the extension would be the brand and manufacturer's name. Depending on the product class of the extension and the fit between the current brand and the proposed extension, consumers may or may not perceive it as a natural extension.

There have been a number of brands that have used either a line (Diet Pepsi, Caffeine Free Diet Pepsi, etc) or brand extension (Ford and GM credit cards, Oral-B toothpaste, etc) to improve company wide sales and/or to enter new markets. However, there has been limited research that has studied the type of brands that can be either line or brand extended successfully and it will be useful to generate a framework that can be used by management to decide on the most suitable form of growth for their brand. Currently, such general guidelines are not easily available.

Brand or line extension?

Existing brands may grow through the use of line or brand extension strategies, or both. It is important for organizations to identify which strategy is more suited at a given point of time and to utilize their current strengths in their growth strategy. It is possible that brand managers may chose to change the characteristics of the brand to be able to use either of the two strategies. However, this may involve use of advertising/promotion to create or change consumer perceptions of the brand before selecting a particular strategy, and may result in diluting the characteristics of the parent brand and it is not proposed to tackle this issue in this paper. Rather, the focus is limited to the strategy that may be pursued in terms of leveraging existing brand strength to other product categories or usage situations. For some companies, a successful brand name may be the single most important asset they possess and they may want to use this brand name to the maximum extent possible and expand to new product classes or new consumer segments. Line and brand extensions may be targeted at different target markets, and the opportunities and risks of using either strategy may be different. In situations where the brand manager has to chose one of the strategies as the primary form of growth for the brand, it is important to correctly choose the best alternative so at improve the chances of success and that is the focus of this paper.

One possible starting point for managerial guidelines could focus on the basis used by consumers to decide on similarities between the parent product and the extension product. Line extensions may share the brand name and the basic structure and strength of the product (e.g., both Pepsi and Diet Pepsi may share associations related to the product - thirst, sold in cans, etc) and evaluations of extensions may be influenced more by the associations related to the product. Brand extensions may not share all the strengths of the parent brand as it would be in a different product class (e.g., AT & T credit card, while not sharing any product associations with AT & T telephone service, may share other associations like good service, prestige, etc). Evaluations of brand extensions may be influenced more by the associations that reflect other benefits offered by the brand. If consumers are using different associations to evaluate line and brand extensions, this may indicate the need for guidelines to be based on the inherent strength and characteristics of the brand itself.

Previous research in brand extension has tended to look at consumers perception of fit between two product classes as a function of the product feature similarity judgements where consumers compare some aspects of the existing set of products with those of the extension product (Aaker and Keller 1990). Other studies in marketing have looked at brand extension from the aspect of the similarity required in terms of both product feature and brand concept, defined as brand-unique abstract meanings that typically originate from a particular configuration of product features (Park, Milberg, and Lawson 1991; Boush and Loken 1991). Park, Milberg, and Lawson (1991) have shown that even if the amount of product feature similarity is low, it is possible to overcome this with a high level of concept consistency if the brand concept is in terms of prestige as against a brand concept in terms of functional aspects. This result is important as it provides an insight into the basis of comparison utilized by consumers (similarity of non-product attributes) when evaluating brand extensions.

Chakravarti, MacInnis, and Nakamoto (1990) view the fit between a parent brand and an extension at the level of shared and unique associations. This is an interesting perspective and considers the role that different associations could play in the success of brand extensions. However, the type of associations that are shared needs to be analyzed further, and there is a need to include line extensions also.

Limited research in marketing has looked at specific types of brands that can be extended to other product classes. Park, Milberg, and Lawson (1991) limit their focus to brands that have either a prestige or functional brand concept and suggest that it would be easier to extend brands with a prestige brand concept to a prestige product category and brands with a functional brand concept to a functional product category. Herr, Farquhar, and Fazio (1990) consider brands that are prototypical of their product category and using the relational model conclude that prototypical brands can be extended to related categories more easily than to non-related product categories. However, the investigation has only focussed on a limited range of brands and has also not considered possibilities for line extension. Desai and Hoyer (1993) address the theoretical principles involved in the processing and evaluation of line extension by consumers and the focus is on what needs to be done to ensure the success of line extensions. There is a need to combine research in the area of line and brand extensions and develop a framework that can address the question of what form of extension would be appropriate for different brands and this paper attempts to make a beginning in this direction.

Brand characteristics

When a brand is being considered for extension into another product class (brand extension) or in the same product class (line extension), it is important to understand how consumers perceive that brand. Brand associations have been suggested as one method for finding out the characteristics that consumers associate with the brand. Brand associations have been defined as the thoughts that are linked to the brand in the consumers memory (Aaker 1991) and a set of associations may represent the underlying value of the brand. It may be possible to use the underlying set of brand associations as a basis to evaluate and select brands for different types of extensions (both brand and line).

A simple and first step to understanding the usefulness of brand associations in extension decisions may be to classify the associations into groups based on shared similarities. A number of content based classifications have been proposed (Aaker 1991; Biel 1992). However, as we are interested in extensions within and outside the product category, it may be better to classify the associations as either product class or non-product class related. Product class related associations are those associated with the performance or use of the product category and would be somewhat analogous to the functional characteristics as defined by Park, Milberg, and Lawson (1991). Non-product class related brand associations are those which are not directly related to the use of the product class and may be associated with the super-ordinate category (soft drinks for a cola brand), user experiences (thirsty), usage situation (party), etc.

Product and non-product class related associations provide us with an initial criteria to understand how a brand may be extended in the same or different product category. The type and characteristics of the associations may be used to decide on the type of line or brand extension suitable for a brand. By analyzing the individual and groups of associations and measuring their strength, brand managers may be in a position to decide the product category (in the case of brand extension) or product usage (in the case of line extension) that may be most appropriate for the brand, from the consumers perspective. In this paper, we will develop this concept further by identifying and classifying brand associations for a number of different brands. We will also develop guidelines and propositions that may be empirically tested.

THEORETICAL BACKGROUND

Brand associations and sets of associations represent not only the underlying value of the brand, but also the meanings that consumers attribute to brands and brand names. Brand associations may very well be the basis on which consumer decisions like brand loyalty and purchase decisions are made (Aaker 1991). In order to better comprehend how brand associations play a role in the consumers attitude toward brand extensions and decision making, it is necessary to understand how associations are developed in memory and how they are activated when consumers are exposed to a brand.

Associative network model

Memory models based on the associative network theory view semantic memory or knowledge as a set of nodes and links (Anderson 1983). Nodes are points in memory that store information or knowledge and there are a number of links that connect these nodes. These links are, in turn, connected to other nodes. Any activation of a node will also activate the links that are connected to it and the other nodes that are connected to those links.

The associative memory model provides us with an excellent framework to represent brand knowledge and associations. It is possible to view brand names or brands being represented as nodes in memory, and the various associations as links connected to that brand name. This conceptualization provides us with a theoretical foundation to understand how associations are formed and developed.

Associations are created by anything linked to a brand (Aaker 1991). Let us consider a new brand that does not have any associations. When this brand name and product class is introduced to a consumer, either through advertising (print or television) or promotion, the consumer may represent knowledge of the brand name as a node in memory. It is also possible that the consumer may have certain thoughts associated with that brand name and product class, creating links attached to that brand node. These associations may be based on the brand name itself (a brand like Federal Express may create an association like fast or quick, based on the 'Express' part of the name) or it may be based on the product class to which the brand belongs (Bic pens may create associations related to the product class of pens and may be based on the consumers previous knowledge of that product class). These initial associations need not be based on any knowledge of the brand or its qualities or performance. Once a consumer actually sees a brand, a different set of associations may be created based on the shape or size of the product, its packaging, or its distribution channel. And once a consumer starts using a brand, yet another set of associations may be formed based on actual usage experiences, usage situations, etc. Thus, it is possible for brand associations to be formed at each stage of interaction between the brand and the consumer.

Strength of associations

Once the initial set of associations are formed, it is likely that further exposures to the same brand name would strengthen these associations. These exposures may be through advertisements or actual product experiences. Each exposure to the brand name would activate both the brand node and the linkages that have already been formed and would strengthen these associations. Not every association may be activated with equal strength on each activation of the brand node. Only those associations that are relevant would be activated and become stronger. For example, if a consumer has the associations of Ronald McDonald, having fun and good service for the McDonald's brand name, and the consumers brand node for McDonald's is activated in a service context, i.e., when buying something at McDonald's, the association of Ronald McDonald or having fun may not be activated, and only the association of good service may be activated. Such selective activation may lead to a strengthening of the good service association only and not of the Ronald McDonald or having fun associations. Thus, we can see that certain associations or sets of associations may be activated with specific experiences. We can also infer that it may be possible to deliberately activate specific sets of associations and not all of the associations at all times. Thus, over a period of time, consumers could develop associations that vary in strength and type.

Content based classifications

Aaker (1991) contends that brand associations may be classified based on content into 11 different groups. The proposed groups are celebrity/person, life style/personality, product class, competitors, country/geographic area, product attributes, intangibles, customer benefits, relative price, use/application, and user/customer. However, so far this proposed classification has not been tested empirically. Of these groups, some like product attributes, customer use/application and customer benefits could be thought to consist of product class related associations and the rest of the groups to consist of non-product class related associations. Product class/non-product class related classification may also be compared to the context independent and context dependent characteristics (Barsalou 1983) where the context independent characteristics would be similar to the product class related associations and the context dependent characteristics would be similar to the non-product class related associations. Product class related brand associations are defined as those associations directly linked with the attributes or characteristics of the product. Examples of product class related brand associations would be easy to use, durability, etc. Non-product class related associations are defined as all the associations not linked to the attributes or characteristics of the product. These may include association groups like intangibles, life style/personality, etc. The focus is on associations that link the brand to a general rather than specific benefit that can be obtained because the brand possesses a specific attribute. The first step would be to identify all the different associations that consumers actually possess. It should then be possible to classify these associations into product and non-product related groups based on whether the associations are directly related to the performance of the product or not. This leads us to our first proposition:

P1: Brand associations can be classified into product class and non-product class related groups.

When a brand is being considered for line or brand extension, the groups of brand associations and their strengths may play an important role in how the extensions are evaluated. In terms of the associative network model (Anderson 1983), the strength of the brand association in the consumers memory would be the strength of the association between the brand node and the association concept node. A brand having strong product class related associations would indicate that consumers have associations linked to the actual use or performance of the brand and also that the brand is perceived to represent the product class quite well. This strength of association may make it difficult to extend the brand to another product class as consumers may still associate characteristics of the parent product class with the extension product class and this may lead to some confusion. However, this strength may be extremely useful in extending the brand within the same product class, but to another segment of consumers using line extension. If consumers already have strong product class related associations, it may be more efficient and effective for the firm to consider line extensions. On the other hand, a brand with stronger non-product class related associations may indicate that the brand is not very closely linked to the product class and may, therefore, be easier to extend to another product class using brand extension. The classification of brand associations into product class and non-product class related associations may be used as a basis to initially decide on the growth strategy that the brand may be able to pursue. If a particular product class association or group of associations becomes strong, consumers may even base their purchasing decisions on that association (Aaker 1991). If the product class related brand association of "roller coasters" becomes extremely strong for the Disney brand name, consumers may expect all Disney brands to possess that particular brand association. This may limit the Disney brand name to consider pre-dominantly line extensions with "roller coasters" as an important physical attribute. If the company were to consider brand extension, consumers may try to transfer some of the strongly held product class related brand associations to the extension brand, creating a situation of poor fit between the extension brand and its product class. This leads us to our next proposition:

P2: Line extension would be the pre-dominant choice of extension for brands that have stronger product class than non-product class related associations.

Consumers may be influenced by groups of associations in different ways. The user/customer group of associations may indicate the typical user of the brand, and if the consumer identifies with that particular group, then these associations may have a positive influence on brand extension evaluations. On the other hand, if the consumer has no reason to associate with that user group, then there may be no influence or there could even be some negative influence. Brand extensions to other product classes associated with the same user group may be favorably perceived by the customer. Similar results may be possible with any of the different groups of associations. Any or a combination of the association groups may be used as a basis to develop possible brand extension opportunities. For example, if McDonald's found that consumers had high strength on a particular group of associations (the intangible group of associations, say), it may become important to focus on this group of associations when considering extensions. If we compare the product class and non-product class related associations with context independent and context dependent characteristics (Barsalou 1983), it may be possible to think of non-product class related associations as being not too strongly linked to the characteristics of the product class. If the non-product class related brand association of "having fun" is strongly linked to the Disney brand name, consumers may expect all other Disney products to have a similar association. This gives Disney a broad range of product categories to choose from when they consider extending into other product classes. On the other hand, if there is a strong association with the type of people who use the Disney brand name, this provides different types of products into which the Disney brand name can be extended. If the company were to consider line instead of brand extension, the lower strength of the product related brand associations may not provide the necessary additional benefits of staying in the same product class. This leads us to our third proposition:

P3: Brand extension would be the pre-dominant choice of extension for brands that have strong non-product class related associations.

The above three propositions provide us with a general idea of the direction that a brand may use to pursue its growth opportunities. However, these propositions still do not clearly show the type of brand or line extension that may be successfully used by a brand. In order to determine this, the brand manager has to be able to analyze the individual brand associations for strength and other characteristics like favorability, etc. Associations that are strong and favorable are more likely to indicate opportunities for the brand to utilize for growth and, therefore, it becomes important to be able to measure these characteristics.

TABLE 1

BRAND ASSOCIATIONS: PRODUCT/NON-PRODUCT CLASSIFICATION AND STRENGTH

Research in psychology has shown us that it is possible to identify word associations through free recall (Nelson et. al., 1992) and this can be extended to associations that consumers may have for brands. Typical experiments have asked subjects to write down the first word that comes to mind when a particular word is activated, the assumption being that when the node containing the name is activated, the relevant associations would also be activated. It is also possible to calculate the strength (based on frequency) of various brand associations by measuring the number of times an association is named by a group of consumers out of the total number of associations. These findings in psychology (Nelson et. al. 1992) are used to test proposition 1 in a pre-test, the details of which are given below. Propositions 2 and 3 are not tested in this paper.

Pre-test

36 undergraduate marketing students participated in the pre-test to fulfill course requirements. Subjects were told that when evaluating an existing brand name, marketers are often interested in finding the associations that brands possess and that one way of finding these associations is through a typical brand name association task. Subjects were provided with complete names of 25 existing brands (e.g., Miller Lite beer, Lee Jeans, etc) selected from the list of strong and weak brand equity (Miller 1993) and told that their task was to write down two words or phrases that they associated with each brand name. Nelson et al. (1992) have recommended that subjects should be asked to list only one thought in a free association task in order to avoid problems of response chaining and retrieval inhibition. However, these problems were reduced by asking subjects to think of the brand name again before coming up with the second associated word.

The list of associations generated for each brand was tabulated separately. Two judges who were blind as to the purpose of the study were given definitions and examples of product class related and non-product class related associations (Appendix 1) and asked to classify the associations into these categories. The inter-coder reliability was 0.82 and all disagreements were resolved by discussion between the two judges and the author. Table 1 gives the list of brands, the number, and strength of product class and non-product class related associations.

RESULTS

The results obtained in Table 1 show that most brands used in this study (15/25) have stronger product class related than non-product class related brand associations. We were able to identify only 2 brands which clearly had higher strength (0.60 or higher) of non-product class related brand associations. Fifteen of the twenty five brands were perceived to have stronger (0.60 or higher) product class related brand associations while eight of the brands had about equal strength (between 0.4 and 0.6) for the product class and the non-product class related brand associations.

APPENDIX 1

DISCUSSION

The results show that a higher number of brands have stronger product class related than non-product class related brand associations. This suggests that consumers may be relating more to the particular features and benefits offered by the product class to which the brand belongs rather than benefits at a more general level. Brands having stronger product related associations may be able to utilize extensions that retain this strength by using line rather than brand extension as a route for expansion and growth. Brands in the introductory stage of their life cycle may be keen to become a part of the product category and therefore advertising and promotion may focus on achieving this objective by creating product class related brand associations. If the intent is to remain as a prominent member of the product category only, the focus of later advertising may be to strengthen the product class related brand associations rather than on creating or developing new associations. For brands with stronger product class related associations, brand extension may be a risky venture as consumers may not be able to understand how the new extension makes sense given the fact that they associate the parent brand so strongly with its product category. Brand extension may lead to perceptions of lesser fit thus reducing chances of success in the marketplace.

Brands with stronger non-product related brand associations can consider brand extension as the primary means of expansion. The strength of the non-product class related brand associations may indicate that consumers perceive that the brand offers benefits at a more general than at a specific product feature level. Benefits at a more general level may be easier to transfer across product categories using brand extension.

IMPLICATIONS

The preliminary results of this study offer a broad framework for brand managers for identifying growth opportunities in the same and different product classes. Brand managers need to first identify the different associations that consumers have for their brand. Free recall could be one of the techniques used for this purpose. Aaker (1991) offers a comprehensive list of techniques that may be used to generate brand associations. These associations need to be analyzed to determine whether they are more strongly related to the product class or whether they indicate that the benefits and associations are at a more general, non-product class related level. This classification can be used to make preliminary decisions with respect to the growth of the brand - line extension if the associations are more strongly linked to the product class, and brand extension if the associations are more strongly linked to non product features and benefits. These are steps which can be undertaken by all brand managers for all brands. Once a decision is taken for a line or brand extension, the nature of the associations can be further analyzed to determine the particular segment to which the brand can be targeted to (line extension) or the particular product class to which it can be extended to (brand extension). This analysis needs to be based on the characteristics of the individual brand associations.

REFERENCES

Aaker, David A. (1991), Managing Brand Equity: Capitalizing on the Value of a Brand Name. New York: The Free Press.

Aaker, David A. and Kevin Lane Keller (1990), "Consumer Evaluations of Brand Extensions," Journal of Marketing, Vol 54 (January), 27-41.

Anderson, John R. (1983), The Architecture of Cognition, Cambridge, MA: Harvard University Press.

Barsalou, Lawrence W. (1983), "Ad Hoc Categories," Memory and Cognition, 11 (May), 211-227.

Biel, Alexander L. (1992), "How Brand Image Drives Brand Equity," Journal of Advertising Research, Vol 32, 6 (November), pp RC6-RC12.

Boush, David M. and Barbara Loken (1991), "A Process Tracing Study of Brand Extension Evaluation," Journal of Marketing Research, Vol 28 (February), 16-28.

Chakravarti, Dipankar, Deborah J. MacInnis, and Kent Nakamoto (1990), "Product Category Perceptions, Elaborative Processing and Brand Name Extension Strategies," in Advances in Consumer Research, Vol 17, Association for Consumer Research, pp 910-916.

Desai, Kalpesh Kaushik, and Wayne D. Hoyer (1993), "Line Extensions: A Categorization and an Information Processing Perspective," Advances in Consumer Research, Vol 20, Leigh McAlister and Michael L. Rothschild (eds.,), Provo, UT: Association for Consumer Research, pp 599-606.

Herr, Paul M., Peter H. Farquhar, and Russel H. Fazio (1990), "Extending Brand Equity to New Categories," Working paper, Graduate School of Business, Indiana University.

Miller, Cyndee (1993), "Upbeat Outlook for U.S. Brands," Marketing News, Vol 27, 11 (May).

Nelson, Douglas L., Thomas A. Schreiber, and Cathy L. McEvoy (1992), "Processing Implicit and Explicit Representations," Psychological Review, Vol 99, 2, 332-348.

Park, C. Whan, Sandra Milberg, and Robert Lawson (1991),"Evaluation of Brand Extensions: The Role of Product Feature Similarity and Brand Concept Consistency," Journal of Consumer Research, Vol 18 (September), 185-193.

Tauber, Edward M. (1988), "Brand Leverage: Strategy for Growth in a Cost-Controlled World," Journal of Advertising Research, 28, (August/September), 26-30.

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