The success of durable brand
extensions
- and the role of consumer innovativeness -
Master thesis, MscBA, specialization Marketing
University of Groningen, Faculty of Management and Organization
August, 2009 MARIJKE HOOGHIEMSTRA Studentnumber: 1386212 De Gilden 9 9247 CX Ureterp e-mail : s.hooghiemstra@student.rug.nl Supervisors
Management Summary
One of the most frequently used branding strategies is to launch brand extensions; companies use an established brand name to launch new products. 82% of all new product introductions are brand extensions (Simms, 2005). This high percentage shows the importance of identifying factors that influence the success of brand extensions. Brands extensions have generated a lot of research in consumer behavior (Ahluwalia, 2008). Most of the time, these researches are focused on fast mover consumer goods (FMCG’s).
In this thesis the different effects that influences the success of brand extensions of durable, or luxury goods are researched. In earlier research different direct, mediating and moderating effects are found that influence the success of brand extensions of FMCG’s. In this research is tested if the effects found in earlier research are applicable as determinants of success of durable brand extensions. The focus of this research is on the consumer innovativeness, in combination with the quality of the parent brand and the marketing support. The problem statement in this research is: How do the consumers perception of the overall quality of the parent brand and the marketing support effect the success of a brand extensions, and is there a difference between the different intensities of the consumer innovativeness?
The aim of this research is to explore and refine the existing theory. Therefore, it is based on research of the core articles. In line with this, the method that is used is also based on these articles. The success of three imaginary brand extensions is measured with a questionnaire. With a regression analysis, and by comparing the correlations, the different effects that influence the success of a brand extension are measured.
Table of content page Management Summary 2 Table of content 3 1. Introduction 5 1.1 A brand 5 1.2 Brand extensions 5
1.3 Advantages of brand extensions 5
1.4 Disadvantages of brand extensions 6
1.5 The success of brand extensions 7
2. Research proposal 8
2.1 Background 8
2.2 Problem statement 9
2.3 Conceptual model and research questions 9
2.4 Theoretical and practical relevance 10
2.5 Structure of the thesis 11
3. Theoretical framework 12
3.1 Direct effects on the success of a brand extension 12
3.1.1 Parent brand characteristics 12
3.1.2 Extension’s marketing context 14
3.1.3 Relationship of parent brand to extension product 15 3.1.4 Extension’s product category characteristics 16
3.2 Mediating effects 17
3.3 Moderating effects 18
4. Research design 26
4.1 Research method 26
4.2 Data collection 28
4.3 Plan of analysis 29
5. Results and conclusions 30
5.1 Internal consistency 30
5.2 Direct effects 31
5.3 Mediating effects 31
5.4 Moderating or interaction effects 32
5.5 Conclusion 32
6. Discussion and recommendations 34
6.1 Differences with earlier research on FMCG’s 34
6.1.1 Direct effects 34 6.1.2 Mediating effects 36 6.1.3 Moderating effects 37 6.2 Measures of fit 37 6.3 Managerial implications 39 6.4 Recommendations 40 6.5 Reflection 41 7. References 42 8. Appendices 46
Appendix 1 – Introductory research 47
Appendix 2 – Measurements and scales 50
Appendix 3 – A print of the questionnaire, available at the internet 52 Appendix 4 – The regression model of the effects toward the extensions 70
1 Introduction
The subject of this thesis is brand extensions. In the first and second paragraph a definition of the concepts “brand” and “brand extensions” is given. In the third and fourth paragraph the advantages and disadvantages of brand extensions will be discussed. The last paragraph covers the importance of brand extensions.
1.1 A brand
In this paper the definition of a brand according to Keller (2003) is used, because this is the leading author in the literature about brands and brand extensions, and this definition is also used by the American Marketing Association (Keller, 2003). A brand is a “name, term, sign, symbol, or design, or a combination of them, intended to identify the goods and services of one seller or a group of sellers and to differentiate them from those of competition”
1.2 Brand extensions
When a company wants to introduce a new product, it has three main choices as how to brand it (Keller, 2003). The first option is to develop a new brand, individually chosen for the new product. A second option is to apply one of the existing brands. The last option is to use a combination of a new brand with an existing brand.
A brand extension occurs when a company uses an established brand for a new product. Which means the last two options that Keller mentioned are approaches for brand extensions. When a new brand is combined with an existing brand, it is called a sub-brand. The existing brand is the parent brand.
According to Keller (2003) brand extensions can be classified in two general categories. The first category is line extensions. In this case the parent brand is used to brand a new product that targets another market segment within a product category currently served by the parent brand. An example for Schwarzkopf is the introduction of Schwarzkopf Dry Shampoo. This is a shampoo with another application than the other shampoos of Schwarzkopf, but it is in the same product category. The second category is category extensions. In this case the parent brand is used to enter a different product category than the category currently served by the parent brand.
1.3 Advantages of brand extensions
First, the brand extension can rely on the parent brand. Consumers make inferences and expectations to the likely performance and composition of the new product based on what they already know about the brand itself and to the extent to which they feel this information is relevant to the new product. These inferences may improve the favorability, the strength and the uniqueness of the extension’s brand associations.
Second, the most important factor for predicting initial trying of a product is the extent to which a known family brand is involved (Keller, 2003). The established reputation of the parent brand can reduce the perceived risk to consumers.
Third, introducing a new brand as an extension may increase the consumer demand, and as a consequence convince retailers to stock and promote the brand extension easier.
Fourth, a company can increase the efficiency of promotional expenditures, and can reduce the costs of introductory follow-up marketing programs. Collins-Dodd and Louviere (1999) state that brands that are already known and recognized require lower new product introduction expenses. Examples of this are the trade deals, price promotions and advertising.
Other advantages are that companies are able to avoid the costs of developing a new brand, have packaging and labeling efficiencies, and can permit consumer variety-seeking.
1.4 Disadvantages of brand extensions
Despite the advantages, there are also some disadvantages.
First a brand extension can frustrate or confuse consumers, especially the different varieties of a product. Often, retailers can not shelve all the variants, and as a consequence consumers can be disappointed when their product is no longer available. A new brand extension could encounter retailer resistance.
Second, a brand extension might fail and hurt the parent brand image. This is the worst case scenario for a brand extension, because the extension not only fails, but it also harms the parent brand. Another disadvantage that influences the parent brand is that a succeed brand extension can cannibalize the sales of the parent brand.
Third, an extension can succeed, but diminish the identification with one category. A risk of multiple products to a single brand is that the brand may not be strongly identified with one product. According to Keller (2003) the potential drawbacks from this weakened image and the lack of identification may be especially evident with prestige or high-quality brands.
1.5 The success of brand extensions
An important determinant of the success of a brand extension is the fit to the parent brand. That fit has been conceptualized as “the extension’s perceived similarity to the parent brand primarily on dimensions such as product category and attributes” (Ahluwalia, 2008). Usually, the higher the perceived fit of an extension with its parent brand, the more positive the extension evaluation and the greater the gain from introducing the new product as a brand extension as compared to under a new brand name (Keller, 2003). One of the reasons of this is that consumers who perceive a good fit between the extension and the core brand category may consider that the extension is a part of this category, and as consequence base their evaluation of the extension on a previously formed attitude to the parent brand. In opposite, consumers who perceive a lack of fit between the extension and the parent brand are less likely to transfer their attitude of the core brand categoryto the extension and may even generate undesirable beliefs about the brand extension (Aaker and Keller, 1990).
2 Research proposal
In the previous chapter an introduction about brand extensions is given. In this chapter a proposal for further research about this topic will be given. This proposal will provide the structure of this research. First the background of this proposal will be discussed, followed by the problem statement, conceptual model and research questions. Successively the theoretical and practical relevance of this research and the structure of the thesis will be presented.
2.1 Background
There are a lot of different variables that effect the success of a brand extension in different ways. In the next chapter, these variables will be discussed more widely.
First, there are the direct effects. Different studies tried to research all the possible variables, by interviewing managers and testing different hypotheses. The results of these researches are outlined in the next chapter. In the mentioned studies, most of the time the researchers have chosen to analyze fast mover consumer goods (FMCG’s) only, like Aaker and Keller (1990), Reddy, Holak and Bath (1994), and Völckner and Sattler (2006). It would be interesting to investigate the extent to which these findings could be generalized to other fields as well, such as consumer durables or services. Therefore, this research will focus on consumer durables in stead of FMCG’s.
Because of the expectation that the consumer innovativeness is of greater importance for durable goods than FMCG’s, another focus of this research will be on consumer innovativeness. The reason for this expectation is that these durable goods are commonly relatively more expensive than FMCG’s and are used for a longer period. As a consequence, the perceived risk for durable goods is higher, which influences the consumer innovativeness. The expectation is that the difference between early and later adopters, and their buying behaviour, will increase.
Second, there are the mediating effects. In this research will be tested if these effects are also significant for durable brand extensions.
2.2 Problem statement
The aim of this study is to explore and refine the existing theory about the factors of success of brand extensions. As mentioned before, a special attention will be on the consumer innovativeness, the difference between early and later adopters.
The research question is :
“How do the consumer’s perceptions of the overall quality of a parent brand and the marketing support effect the success of a brand extension,
and is there a difference between the different intensities of the consumer innovativeness?”
Although the expectation is that the effect of the consumer innovativeness will be different the most between durable and FMCG’s, all the other effects found in earlier research will be tested as well, and will be researched to find out whether these are applicable for durable brand extensions. Through this the existing theory can be explored and refined. For this purpose, the different research questions are formulated.
2.3 Conceptual model and research questions The following research questions were addressed:
1. How do the different direct, mediating and moderating effects mentioned in earlier researches effect the success of a brand extension for a durable good?
2. What is the role of consumer innovativeness in the relation between the quality of the parent brand and the success of a brand extension?
3. What is the role of consumer innovativeness in the relation between the marketing support and the success of a brand extension?
Figure 1 shows the conceptual model for this research. All these variables are tested in earlier research, except the bold italicized ones, that will be discussed in the next chapter. Völckner and Sattler (2006) did a meta-analysis, which combines the results of several studies about the success of brand extensions. These variables are used in the conceptual model, because the expectation is that these variables are better estimates of the true effect size than those derived in single studies, under a given set of conditions and assumptions.
FIGURE 1
Overview of the Conceptual Framework
Source : See figure 2 (Chapter 3)
2.4 Theoretical and practical relevance
In the available research about the success of brand extensions different factors of success are found. In total, there are ten variables that directly effect the success of a brand extension, and there are different mediating and moderating effects. However, this is tested for FMCG’s only. There is a lack in the existing literature concerning the success of durable brand extensions; the theoretical relevance for this research.
There is also an obvious practical relevance ; to managers who are selling durable goods or services, it is interesting to know the effects that influence the success of a brand extension. Managers can take this in account when selecting an extension product, and managers should give the greatest
Success of a brand extension Extension’s marketing context
- Marketing support - Retailer acceptance
• Marketing support -> Retailer acceptance • History of previous BE -> PB conviction • PB experience -> PB conviction
• Marketing support -> Fit
• PB conviction -> Quality of the PB
• History of previous BE -> Quality of the PB • History of previous BE -> Retailer acceptance • Fit -> Retailer acceptance
• Perceived risk of unknown brands -> consumer innovativeness
Relationship of parent brand to extensions product
- Fit
- Linkage of the utility of the PB to product attributes of the original product category
Parent-brand characteristics - Quality (strenght) of the PB - History of previous BE - PB conviction
- PB experience
Extension’s product category characteristics
- Perceived risk
- Consumer innovativeness
BE = Brand Extension, PB= Parent Brand • Fit x Quality of the PB
• Fit x PB conviction
• Consumer innovativeness x Fit • Fit x Linkage of the utility • Marketing support x Fit
attention to the factors that significantly influence the success of a brand extension. For example, managers can influence the marketing support of a brand extension on a short term, so it is useful to know if this effects the success of the brand extension.
One emphasis in this research is the interaction effects of consumer innovativeness with the quality of the parent brand and the marketing support. If it will be fulfilled that these interaction effects exist, managers can give more attention to the type of consumers they serve, and adjust their marketing support. In that case managers also can give the desirable attention to the importance of the quality of the parent brand. So, if the consumer innovativeness is a moderator in this, the managerial implication is that managers can change their marketing communications to the distinctive adoption tendencies. Later adopters adopt new products on the basis of much more information than earlier adopters, like multiple exposures to marketing communications, word of mouth and observations of the product used by others. Often, this information is not available.
2.5 Structure of the thesis
3. Theoretical framework
One of the most frequently used branding strategies is to launch brand extensions; companies use an established brand name to launch new products. According to Simms (2005) 82% of all new product introductions are brand extensions. This high percentage shows the importance of identifying factors that influence the success of a brand extension, or the potential of a brand into new markets. Brand extensions have generated a lot of research in consumer behavior (Ahluwalia, 2008) because brand extensions generate important growth and new product introductions by new companies. This paper covers a part of that research.
There are a lot of direct, mediating and moderating effects that possibly influence the success of a durable brand extension, too much to mention all of them in this research. In the conceptual model a selection of variables is presented to investigate. This selection is based on the research that is done about the success of brand extensions in FMCG’s. The effects that were significant in earlier research are taken into account, therefore a comparison can be made with earlier research. The expectation is that the sign of the different effects is the same for durable brand extensions, but that the importance of each effect differs. For example, the success of the extension still is positively dependent on the marketing support, but this effect can be greater for durable brand extensions in comparison with the extension of FMCG’s.
Another expectation is that the consumer innovativeness, in combination with the quality of the parent brand and the marketing support, is more important to durable then to FMCG’s brand extensions.
3.1 Direct effects on the success of a brand extension
There are different factors of success for a brand extension. The relevant determinants can be classified in four categories (Völckner and Sattler, 2006):
1) the parent-brand characteristics 2) extension’s marketing context
3) relationship of parent brand to extension product 4) extension’s product category characteristics
Each of these four categories will be discussed in the next paragraphs.
3.1.1 Parent brand characteristics
quality (strength) of the parent brand, the history of previous brand extensions, the parent brand conviction and the parent brand experience.
The quality (strength) of the parent brand
The quality is conceptualized here in terms of the consumer’s perception of the overall quality of the brand. The perceived quality has received considerable attention in the marketing literature. Zeithaml (1988) defines the perceived quality as “a global assessment of a consumer’s judgment about the superiority or excellence of a product”. The difference between the quality of a product and other specific attributes is that the perceived quality is at a higher level of abstraction.
Aaker and Keller (1990) also conceptualized quality in this way. Smith and Park translated this definition into a conceptualization of brand strength. In this case, brand strength has been articulated implicitly in terms of consumer’s predispositions toward the brand (following the Marketing Science Institute, 1988).
Völckner and Sattler (2006) translated this brand strength directly in brand quality, when they did an analysis of the determinants of the success of a brand extension. The hypothesis was based on the quality (strength) of the parent brand. The corresponding expectation was that if the quality of the parent brand is high, the brand extension is more successful (based on Smith and Park, 1992).
In another research brand strength constitutes the brand associations held by consumers (Lassar, Mittal and Sharma, 1995). This includes the quality of the brand, but other brand attributes as well. Because of the focus of this research on the quality of the parent brand, the definition according to Zeithaml (1988) will be used.
There is a positive relation between the quality of the parent brand and the success of a brand extension. If the quality of the parent brand is high, the brand extension is more successful (Smith and Park, 1992).
The history of previous brand extensions
however this determinant has a significant and positive effect on the parent brand conviction, the quality of the parent brand and the retailer acceptance.
The parent brand conviction
If the parent brand conviction is high, the brand extension is more successful. The parent brand conviction can be measured in different ways. The parent brand conviction is high for example, when consumers generate a liking for and trust in the brand name, have parent brand knowledge and strong brand associations (Kirmani, Sood and Bridges, 1999).
The parent brand experience
This variable measures if the respondent has experiences with the parent brand, for example by measuring the frequency of using or purchasing the parent brand. If the parent brand experience is high, the brand extension is more successful (Broniarczyk and Alba, 1994 ; Swaminathan, Fox and Reddy, 2001)
Völckner and Sattler (2006) found that the last two determinants (parent brand conviction and parent brand experience) are real important to managers. The implication of these last two determinants is that there are favourable strategies, like building customer-based brand equity or acquiring strong brands. These strategies represent a prerequisite for the successful leveraging of an existing brand to get some financial benefit. Managers who have access to a portfolio of more parent brands might consider influencing the conviction and the experience of the parent brand by selecting a brand that performs well on these two factors.
3.1.2 Extension’s marketing context
There are two determinants of the success of an extension in this category; the marketing support and the acceptance of the retailer.
The marketing support
If the marketing support is high, a brand extension is more successful (Reddy, Holak and Bhat, 1994). Examples of marketing support are advertising support or marketing competence of a company.
Acceptance of the retailer
Völckner and Sattler (2006) found that both of these factors are real important to managers. For products that are FMCG’s the marketing support plays a critical role in determining the success of a brand extension. Managers can influence this factor on the short term. The retailer acceptance can be influenced by managers as well. Promotional allowances, for example, reduce retailers’ costs of telling consumers that the brand extension product is available. Consumer advertising also increases the demand for the product.
3.1.3 Relationship of parent brand to extension product
There are two determinants of the success of an extension in this category; the fit between the parent brand and the extension product, and the linkage of the utility of the parent brand to product attributes of the original product category.
The fit between the parent brand and the extension product
The fit is the similarity between the product classes of the parent brand and the brand extension. The degree of a fit between the product categories of the parent brand and the brand extension is an important factor of success for a brand extension (Aaker and Keller, 1990). A lot of research is done about this factor (e.g. Bottomley and Doyle, 1996 ; Broniarczyk and Alba, 1994), and the scales of these three researches will be used.
There are three dimensions of fit. The first fit measure is ‘complement’, which indicates the extent to which consumers view two product classes as complements. The second fit measure is ‘substitute’, which is the extent to which consumers view to product classes as substitutes. The last fit measure is ‘transfer’, which is how consumers view relationships in product manufacturing (Aaker and Keller, 1990). If the fit between the parent brand and the extension is high, a brand extension is more successful. Examples of this fit are: if there is a high global similarity, if there is a high ability of the owner of the parent brand to make a product in the extension class (Aaker and Keller, 1990) or when there is a high relevance of the extended associations for the extension product (Broniarczyk and Alba, 1994).
The linkage of the utility of the parent brand to product attributes of the original product category Rangaswamy, Burke, and Oliva (1993) investigated the linkage of the utility of the parent brand to product attributes of the original product category, thus how closely tied these attributes are. The less the utility of the parent brand is linked to product attributes of the original product category, the more successful the brand extension is (Rangaswamy, Burke, and Oliva, 1993).
3.1.4 Extension’s product category characteristics
There are two determinants of the success of an extension in this category; the perceived risk and the consumer innovativeness.
The perceived risk
The perceived risk to a consumer by buying a product can be measured for example through asking about the uncertainty that consumers feel when buying products. If the perceived risk is low, the brand extension is more successful (Nijssen and Bucklin, 1998).
The consumer innovativeness
because this signals greater risk. In contrast, less innovative consumers may be effected by the relatively higher risk associated with the lower marketing support and quality of the parent brand.
In conclusion, Klink and Smith (2001) found the direct effect that if the consumer innovativeness is high, the brand extension is more successful
The conclusion of Völckner and Sattler (2006) is that there are five important determinants of the success of a brand extension for managers. These five can be ordered, from the most important to the less important one, as followed: the fit between the parent brand and an extension product, followed by marketing support, parent brand conviction, retailer acceptance and parent brand experience.
3.2 Mediating effects
In the previous paragraphs the relationship between some determinants and the success of the brand extension are mentioned. There are mediating effects as well; the impact of a certain success factor on another one.
Quality of the parent brand -> retailer acceptance
Retailers do not expect that manufacturers place a strong brand name on a product that is inferior. Therefore, the quality of the parent brand is expected to have a positive influence on retailer acceptance of a brand extension (Collins-Dodd and Louviere, 1999). According to Völckner and Sattler (2006) this effect is not significant, therefore, this effect is not included in the conceptual model.
Marketing support -> retailer acceptance
Retailers like a great marketing support of a brand extension, because that creates consumer awareness. Therefore, greater marketing support – in terms of consumer advertising – generates more retailer acceptance (Collins-Dodd and Louviere, 1999).
History of previous brand extension -> parent brand conviction
Parent brand experience -> parent brand conviction
Direct experiences with the parent brand might increase the personal relevance to the brand, and consequently generate a greater liking for the brand, greater parent-brand knowledge, stronger brand associations, and this might result in a higher level of parent brand conviction (Kirmani, Sood and Bridges, 1999).
Marketing support -> fit
Repeated exposure to a brand extension can elevate the consumers’ perception of fit. The exposure can help consumers to identify more shared attributes between the parent brand and the extension. Thus, higher levels of marketing support – in terms of advertising – lead to higher levels of perceived fit (Klink and Smith, 2001).
During interviews with managers Völckner and Sattler (2006) found five other possible mediating effects; the parent brand conviction also influences the quality of the parent brand, the history of the previous brand extension influences the quality of the parent brand, as well as the retailer acceptance. The retailer acceptance is influenced by the fit as well, and the last expectation is that the perceived risk influences the consumer innovativeness.
According to Völckner and Sattler (2006) the effect of structural relationships among the success factors on the success of an extension is still not clear, because there is no study available that considered the extension success as a component of the estimated model.
3.3 Moderating effects
Besides the mentioned direct and indirect (mediating) effects, there are some moderating or interaction effects as well that may play a role in determining the success of a brand extension.
Quality of the parent brand x fit
There is a positive interaction between the quality of the parent brand and the fit between the brand and the extension. The degree to which the associations of the brand are transferred to the extension depends on the level of perceived fit. The attitude of a consumer toward an extension is higher when there is both a perception of fit between the two product classes, and a perception of high quality of the original brand (Aaker and Keller, 1990).
when there is a basis of fit between the two product classes of the parent brand and the brand extension. However, in line with Smith and Park (1992), Völckner and Sattler (2006) found that the quality of the parent brand has a significant direct effect on the extensions success (with an estimated structural parameter of + .10).
In 2001, Bottomly and Holden investigated the empirical generalization of the model of Aaker and Keller (1990) that hypothesizes that an evaluation of a brand extension is based on the quality of the parent brand, the fit between the parent and the extension category, and the interaction of the two. Despite published results of earlier researchers, these authors found support for the full model of Aaker and Keller (1990). Bottomly and Holden (2001) found evidence that the level of contribution of each of the components of fit in the model varies by brand and culture.
Quality of the parent brand x history of previous brand extensions
Consumers tend to be more confident about judgments on the quality of the parent brand if these are based on large samples; products that bear the parent brand name. If a brand has a lot of extensions the consumer can give a better judgment. Consequently, the positive effect of higher levels of quality of the parent brand on the success of a brand extension increases as the number/success of previous brand extensions increases (Dacin and Smith, 1994). According to Völckner and Sattler (2006) this effect is not significant, therefore, this effect is not included in the conceptual model.
History of previous brand extensions x fit
The negative effect of low levels of fit on the evaluation of extensions diminishes as the number/success of previous brand extensions increases. According to Völckner and Sattler (2006) this effect is not significant, therefore, this effect is not included in the conceptual model.
About the interaction effect between the fit and the parent brand conviction: the research of Yeung and Wyer (2005) provides evidence for the issue that consumers who feel good about a core brand, also evaluate the extensions of this brand positively, even if the extension is dissimilar to the parent or core brand. Hence, there is no interaction effect between the parent brand experience and the fit between the parent brand and the extension product.
Often, consumers may form an initial impression of a product based on its brand name alone, even before they learn about the specific features. After this impression, this could influence the evaluation of the extension independently of any conscious categorization processes that occur afterwards.
Yeung and Wyer (2005) have tested the hypothesis that when consumers evaluate an extension of an affect-eliciting brand, the effect it causes can have a positive impact on their evaluations of the brand extension, regardless of whether the extension is similar or not to the core brand. The authors define an affect-eliciting brand as brands where by consumers have subjective feelings that they experience when they encounter the brand.
When a core brand did not elicit affect, the evaluation of the extension would be dependent on the similarity to the core brand. And vice versa, when the core brand elicited affect, this affect would influence the evaluation of an extension regardless of the core-extension similarity. So by an affect-eliciting brand, similar and dissimilar extensions can both be evaluated positively by a favorable brand.
There is also a significant effect to the evaluation of an extension for an affect-elicited brand when consumers have a positive mood before they where exposed to the core brand. Consumers in general evaluate extensions more positively when they are moderately similar to the core brand category, then when they are different. This effect of similarity is higher when the core-extension similarity was estimated before the evaluation. Thus, when the extension’s similarity to the core product is not mentioned, it has a relative small impact on the evaluation of the extension. In conclusion; the participants of the research learned something about the core brand before they received information about the nature of the extension and the attributes of the extension. Therefore they are particularly likely to form an expectation of what the extension would be, based on this information only. This influences the evaluation of the extension.
the brand and the product category of the extension. The resulting managerial implication is that brands should not be extended to perceptually categories with a distance. However there are many brands that have been extended successfully with no or a little perceived fit between the brand and the product category of the extension as well. There are different possible explanations for this. Klink and Smith (2001) found three possible background traits that may help explain the difference between these prior findings and marketplace observations.
1) Limited extension information
In previous research about brand extensions the researchers restricted the amount of attribute information that is available to subjects. Often only the brand name and the product category are used as the stimulus to be evaluated. But in real life, subjects will rely on different available cues like price, the brand name, experiences, and so on. And when other cues are available, the impact of a single cue is diminishing. So, if there are more cues to evaluate, the importance of the perceived fit becomes smaller.
2) Failure to account for consumer’s new product adoption tendencies
Another issue is, that in previous research a distinction between early and later adopters is not made, while this can make a difference. One of the basic factors underlying the effects of fit on brand extensions is the perceived risk of a consumer. As the level of fit increases, the perceived risk decreases, which can have a positive effect on the evaluation of a brand extension. According to Rogers (1983), later adopters tend to be more risk averse than earlier adopters. Because there are relatively a lot of later adopters this can raise the adverse effect of a low perceived fit. Another variable that can determine the adoption tendency is the consumer innovativeness. The more innovative the consumer, the more the consumer adopts new or innovative products. Völckner and Sattler (2006) found that the perceived risk has a direct effect on the consumer innovativeness.
3) Single exposure to proposed extensions
According to Klink and Smith (2001) previous research also limits subjects to one exposure of proposed extensions. But, in reality, new products are launched with substantial communication efforts. These often result in repeated exposure to many consumers in the target market. This can be important to brand extensions with a low perceived fit.
available, which raises the effect of perceived fit. In addition, although the exposure setting was fine for the innovator, a setting of only one exposure is not natural for the later adopter. The experimental conditions of prior research may have subtly created a relatively high-risk setting in which to evaluate the alternatives.
In conclusion, higher levels of perceived fit between the parent brand and proposed extension will lead to more favorable evaluations of the extension. But the effects of fit on extension evaluations will diminish as the amount of product-related information increases. A higher level of exposure to the brand extension gives a higher level of perceived fit, and the effects of fit on extension evaluations will be greater when consumer innovativeness is low than when it is high. Klink and Smith (2001) found that the effects of fit disappear when attribute information is added to extension stimuli and are applicable for later product adopters only.
Similar to Klink and Smith (2001) Czellar (2003) notices that external information is an important factor in research of brand extensions. Competitor marketplace activity, distributor activity and other external information directly effect the perceived fit between the brand and the brand extension. Czellar (2003) adds another explanation for differences in earlier research as well:
4) Positioning of the brand – functional versus symbolic brands
Brands with an experiential or functional positioning are influenced more by providing product-related information than by providing non-product-product-related information. For brands with a symbolic brand positioning, non-product-related information influences fit perceptions more than providing product-related information.
In conclusion, these conditions contain three moderating or interaction effects on the success of brand extensions.
Fit x Product related information (linkage of the utility of the parent brand to product attributes of the original product category)
The effects of fit on extension evaluations will diminish as the amount of product-related information increases
Consumer innovativeness x fit
The effects of fit on extension evaluations will be greater when consumer innovativeness is low than when it is high.
Marketing support x fit
The effects of fit on extension evaluations will diminish as the amount of marketing support increases.
And last, in this research two new possible moderating effects are included:
Consumer innovativeness x quality of the parent brand
Later adopters adopt new products on basis of much more information than earlier adopters, they are more risk averse, and are commonly not the ones who first test the newest innovations, enjoy buying the latest products or buy new products before others do. This group of consumers tends to be more confident about judgments of the quality of the parent brand if they take less risk, in this case the quality of the parent brand is more important. Consequently, the positive effect of higher levels of quality of the parent brand on the success of a brand extension increases as the consumer innovativeness increases.
The expectation is that the effects of the quality of the parent brand on extension evaluations will be greater when consumer innovativeness is low than when it is high.
Consumer innovativeness x marketing support
The expectation is that the marketing support on extension evaluations will be greater if the consumer innovativeness is low than when it is high.
3.4 The conceptual model
The conceptual model that is presented in paragraph three of chapter two is based on collecting the variables that influence the success of a brand extension. In this research, the acceptance of the retailer is not measured, because of the imaginary character of the brand extension introduced in the questionnaire. The lack of information of this effect on the success of a durable extension can be a direction for further research about this subject.
FIGURE 2
Hypotheses Regarding Success Factors of Brand Extensions
Expected Source Sign
H1 Quality of the parent brand + Smith and Park (1992)
H2 History of previous brand extensions + Dacin and Smith (1994)
H3 Parent brand conviction + Kirmani, Sood and
Bridges (1999)
H4 Parent brand experience + Swaminathan, Fox and
Reddy (2001)
H5 Marketing support + Reddy, Holak and Bhat (1994)
H6 Fit between the parent brand and the + Aaker and Keller (1990) extension product
H7 Linkage of the utility of the parent - Rangaswamy, Burke and
brand to product attributes of the Oliva (1993)
original product category
H8 Perceived risk - Nijssen and Bucklin (1998)
H9 Consumer innovativeness + Klink and Smith (2001)
H10 History of previous brand extensions -> + DelVecchio (2000) Parent brand conviction
H11 Parent brand experience -> Parent brand + Kirmani, Sood and
H12 History of previous brand extensions -> + Völckner and Sattler (2006) Quality of the parent brand
H13 Marketing support -> Fit between the + Völckner and Sattler (2006) parent brand and the extension product
H14 Parent brand conviction -> Quality of the + Völckner and Sattler (2006) parent brand
H15 Perceived risk -> Consumer innovativeness - Völckner and Sattler (2006) H16 Fit between the parent brand and the + Aaker and Keller (1990) extension product x Quality of the parent brand
H17 Fit between the parent brand and the + Völckner and Sattler (2006) extension product x Parent brand conviction
H18 Consumer innovativeness x Fit between + Klink and Smith (2001) the parent brand and the extension product
H19 Fit between the parent brand and the + Klink and Smith (2001) extension product x Linkage of the utility of
the parent brand to product attributes of the original product category
H20 Marketing support x Fit between the parent + Czellar (2003) brand and the extension product
Expected Moderating Effects:
H21 Consumer innovativeness x +
Quality of the parent brand
H22 Consumer innovativeness x +
Marketing support
4. Research design
In this chapter the research method, the data collection and the plan of analysis will be presented.
4.1 Research method
The explanatory research that will be done is based on the research of the core articles mentioned in the earlier chapters. This is closely related to the aim of this research, to refine the conclusions of earlier research. A questionnaire is developed using variables and rating scales that closely coincide with those used in previous studies. In line with the earlier researches, the questions will be scaled with a 7 point – Likert scale. In these researches different scales are used, and in this research a scale from -3 to 3 will be used (e.g. in line with Völckner and Sattler, 2006). For this reason a bipolar scaling method will be used in this research, measuring either positive or negative response to a statement( -3 = Strongly disagree, -2 = Disagree, -1 = Slightly disagree, 0 = Neither agree nor disagree, 1 = Slightly agree, 2 = Agree, 3 = Strongly Agree )
So, a 0 means that a respondent is undenominational, a negative outcome means that the respondent (on a different intensity) disagrees, and a positive outcome means that the respondent (on a different intensity) agrees, with the statement. The questions which are used all have their origin in previous researches, these are all converted in this scale, which means the results can be compared per question with earlier research outcomes.
The basic elements in this research include the parent brand, the brand extension, and the variables that influence the success of the extension. First, all the involved variables will be mentioned.
The parent brand
In this research three different parent brands will be tested; Canon, Sony and Philips. These brands are selected because of different reasons. First these brands have done different brand extensions, so the history of earlier brand extensions can be measured, and people can make an imagination of the brand and the possible extension by evaluating the earlier extensions. Second, these are brands that produce durable, luxury goods, on which is being focused, in contradiction to earlier research that has focused on FMCG’s. Third, these brands are supposed to be well-known. In the pre-test, which will be shown in the next subsection, all respondents could fill in the questions about these brands, and had an opinion about it.
The extension
introduced to the market). Respondents were asked to rate independent (the success factors) and dependent (the success of the extension; for example the perceived quality of the extension) variables on rating scales. Aaker and Keller’s (1990) fundamental study represents a similar example. A few studies used market data as dependent variables (e.g., market share or stock market value; Lane and Jacobson 1995; Reddy, Holak, and Bhat 1994; Smith and Park 1992).
In this study the survey is also based on hypothetical brand extensions. These three extensions are different from each other in the level of fit between the parent brand and the extension because of the relative great importance of the fit as influencing factor on the success of brand extensions according to earlier researches (e.g. Völckner and Sattler, 2006).
The imaginary brand extensions are tested in a small introductory research (Appendix 1), under a group of students. This is a pretest for the final questionnaire, for which the target group are students as well. The first extension is the parent brand “Canon”, with a flat screen television. The result of the introductory research for this extension was that there is no distinct fit or misfit. The average fit of this extension is 0,07 (n=23).
The second extension is the parent brand “Sony”, with a motor cycle. The result of the introductory research for this extension was that there is no fit. The average fit of this extension is -0,92 (n=24). Most of the respondents (n=23) disagreed with the statement that there is a global similarity between “Sony” and a motor cycle. However, this extension is imaginable. The brand “Yamaha” on the other hand produces keyboards, DVD-players and motorcycles as well.
The third extension is the parent brand “Philips”, with a fitness training machine. The result of the introductory research for this extension was that there is a fit. The average fit of this extension is 1,21 (n=24). Most of the respondents (n=20) agreed with an estimated level of fit for this parent brand and extension. Overall, the answers of the questions about fit of the mentioned extensions varied from -3 to 3, so there is a variation in perceived fit in these three extensions overall.
The influencing variables
In paragraph 3.1, the definitions of the variables that can influence the success of the brand extensions are shown. In appendix 2 the measurements and scales of these variables are shown.
Here, only the marketing support will be discussed further; in most researches that were done the only information the respondents received was the name of the brand and the product class (e.g. Aaker and Keller (1990), and Boush and Loken (1991)). The reason for this was to avoid confounding the reactions. So the experiment provides information about the baseline reactions that occur and the associations that respondents have. In real life, during a launch of an extension a firm can influence the perception of an extension by providing information cues through advertising and other marketing mix activities. Klink and Smith (2001) found that the effects of fit on the evaluation of the extension will diminish as the amount of product-related information increases.
So in this case, an overview and examples of different types of marketing support of the brand extension will be given in the interviews. After that, the marketing support can be measured, with scales of Nijssen (1999) and Reddy, Holak and Bhat (1994) as to how the extension receives marketing support.
Success of the brand extension
The success of a brand extension can be measured with a lot of different variables, like consumer evaluations of brand extensions linked, for example, to the perceived quality of the extension. In the questionnaire of this research all the constructions are based on variables and rating scales that closely coincide with those used in previous studies. Thus, the success of the brand extension in this research will also be based on different researches, that of Völckner and Sattler (2006), Aaker and Keller (1990) and Bottomley and Doyle (1996).
4.2 Data collection
Information about the respondents will be asked in the questionnaire as well. These questions will be asked at the end of the questionnaire and are about the age, gender and the highest education that the respondents have finished or are following at the moment.
4.3 Plan of Analysis
A total of 97 respondents completed the questionnaire at the website ‘www.thesistools.com’, this questionnaire can also be found in appendix 3. Each respondent answered the questions for three imaginary brand extensions, so a total of 291 cases were obtained. The program ‘SPSS’ will be used to obtain the analytical results. The questions with a 7-point Likert Scale are all ordinal scales, this can be seen as interval scales as well. The reason for this is that each variable is measured with some ordinal scaled questions with a relative small interval. These mingled constructs can be seen as interval scaled variables. In earlier research, like that one of Völckner and Sattler (2006), the constructs are used like this, so in this research this approach will be used as well. The questions about gender and education are nominal scales and the question about age is a ratio one. With these variables different tests can be done to measure how one variable influence the other. The results will be presented in the next chapter.
First, an examination of the reliability and validity of the measures will be done, by computing the Chronbach’s alpha for all scales. After that, the influence of the different variables will be measured. As previously mentioned, the primary purpose of this research is to refine earlier research. In those researches, the direct, mediating and interaction effects on the success of brand extensions were measured empirically for FMCG’s. Even though the emphasis of this research will be on durable, luxury goods, the model specification of earlier research are usable, because of the same construct and questions used in earlier research, like that of Völckner and Sattler (2006).
Because the aim of this research is to refine earlier research, the identified effects will be integrated into alternative models according to Völckner and Sattler (2006). The first is an effects-only model (M1). Model M1 contains the direct effects of success factors on the extension’s success that have been proposed in earlier research. Second, the mediating effects are included by measuring the correlations between the different variables, so the structural relationships between determinants can be identified (M2). At last, the moderator effects will be introduced by using a step-by-step procedure (M3a and M3b).
5. Results and conclusions
In this chapter the main results and conclusions based on the results of the data collected through the questionnaire will be given. First, the internal reliability and validity of the used measures will be discussed. Second the results of an analysis of the direct, mediating and moderating effects that influences the success of a brand extension will be discussed, and after that the results of the two moderating effects will be presented. In the next chapter, there will be a discussion, in this chapter the differences between this and earlier research will be explained more widely as well. After that some recommendations for further research will be given, and a reflection of this research.
All the respondents are high educated people between the ages of 18 and 27. 42,3% of the respondents is male, the other 57,7% is female. This correlates with the population of high-educated people between the ages of 18 and 27.
5.1 Internal Consistency
TABLE 1 Internal Correlations
Variables Number of Chronbach’s
questions Alpha - Quality of the parent brand 2 ,744 - History of previous brand extensions 2 ,771 - Parent brand conviction 2 ,793
- Parent brand experience 2 ,807
- Marketing support 2 ,835
- Retailer acceptance --
- Fit 3 ,882
- Linkage of the utility of the parent brand 1 to specific product attributes
- Perceived risk of unknown brands 2 ,648
- Consumer innovativeness 3 ,930
- Success of the brand extensions 4 ,877
5.2 Direct effects
In earlier research, like the one of Völckner and Sattler (2006), a regression analysis is used to measure the direct effects, therefore, this will be done in this case as well. The results are shown in table 2, model 1 (M1), and also in appendix 4.
Only the history of previous brand extensions, the fit between the parent brand and the brand extension, the linkage of the utility of the parent brand to products attributes of the original product category, and the marketing support have a direct effect on the dependent variable, the success of the brand extension. However, the history of previous brand extensions and the linkage of the utility of the parent brand to products attributes of the original product category do not show the expected sign, but the opposite one. This will be discussed further in the next chapter.
5.3 Mediating effects
5.4 Moderating or interaction effects
To measure the moderating effects the variables that interact with each other are multiplied (like Aaker and Keller (1990) did). So, besides the mentioned moderating effects in the literature, a new variable is made by multiplying the innovativeness and the quality of the parent brand (in model M2a), and another one by multiplying the innovativeness and the marketing support (in model M2b). With these new variables, a new regression analysis is done, by adding a new interaction effect step-by-step with the direct effects (also available in appendix 4).
All the interaction effects mentioned in earlier research (H16 – H20) are not significant.
Consumer innovativeness and the quality of the parent brand
The interaction between the innovativeness and the quality of the parent brand has no significant effect on the success of the brand extension (table 2, M2a). By including this interaction effect, the other directs effect did not change a lot in comparison to the situation in M1.
Consumer innovativeness and the marketing support
The interaction between the innovativeness and the quality of the parent brand has no significant effect on the success of the brand extension (table 2, M2b). However, the marketing support as a direct effect on the success of the brand extension gives a greater effect on the success of a brand extension, while the other direct effects are nearly the same.
5.5 Conclusion
The result of this research is that an answer to the problem statement can be formulated. There are four direct effects that influence the success of brand extensions, H5 andH6 are supported, H2 and H7 are of significant influence, but not in the expected way. There is no evidence for the direct effects of H1, H3 ,H4, H8 and H9, so these are rejected.
In the next chapter some of the differences will be discussed further. In that chapter some recommendations for further research will be given and a reflection on this research as well.
TABLE 2
Structural Parameter Estimates
Estimates
Variables M1 M2 M3a M3b
H1 Quality of the parent brand n.s. n.s n.s. n.s.
H2 History of previous brand extensions -,145 *
-,145* -,145* -,145*
H3 Parent brand conviction n.s n.s. n.s. n.s.
H4 Parent brand experience n.s. n.s. n.s. n.s.
H5 Marketing support ,242 ** ,242** ,240** ,328** H6 Fit ,494 ** ,494** ,496** ,502** H7 Linkage of the utility of the parent brand ,224
**
,224** ,223** ,220** to specific product attributes
H8 Perceived risk of unknown brands n.s. n.s. n.s. n.s.
H9 Consumer innovativeness n.s. n.s. n.s. n.s.
H10 History -> Parent brand conviction -- ,480 **
,480** ,480** H11 Parent brand experience -> PB conviction -- ,541
**``
,541** ,541**
H12 Marketing support -> Fit -- ,341
**
,341** ,341** H13 Parent brand conviction -> Quality -- ,684** ,684** ,684** H14 History -> Quality of the parent brand -- ,481
**
,481** ,481** H15 Perceived risk -> Consumer innovativeness -- ,364
**
,364** ,364**
H21 Consumer innovativeness x Quality -- -- n.s. --
H22 Consumer innovativeness x Marketing support -- -- -- n.s. *
significant at the p < ,05 level n = 291 , adjusted R square = .65 **
significant at the p < ,01 level n.s. = not significant
6. Discussion and recommendations
The aim of this research is to refine earlier research, and to add some new components. In this paragraph the results of this research will be discussed in comparison to earlier research. First the direct effects will be discussed, followed by and after that the mediating and moderating effects. The differences with earlier research on FMCG’s will be discussed, and after that the measures of fit and the managerial implication will be discussed. At the end of this chapter some recommendations for further research, and a short reflection on this research will be given.
6.1 Differences with earlier research on FMCG’s
This research has the aim to refine earlier research, and focus on durable goods in stead of FMCG’s. In the research of FMCG’s, Völckner and Sattler (2006) did a covering research on different effects mentioned in the marketing literature. In this paragraph the differences of earlier research in comparison to this research will be discussed.
6.1.1 Direct effects
The conclusion of Völckner and Sattler (2006) is that there are five important determinants for managers that directly effect the success of a brand extension. These five can be ordered, from the most important to the less important one, as followed: the fit between the parent brand and an extension product, followed by marketing support, parent brand conviction, retailer acceptance and parent brand experience.
In this research about the success of durable brand extensions four determinants of success of a brand extension were found. These can be ordered, from the most important to the less important one, as followed: the fit between the parent brand and an extension product, the marketing support, the linkage of the utility of the parent brand to specific product attributes and the history of previous brand extensions.
So the fit between the parent brand and an extension product and the marketing support are important for the success of brand extensions, no matter if these are fast mover consumer or durable goods.
The parent brand experience and conviction are of less importance to durable brand extensions. Perhaps this can be explained with the expectation that consumers have less experience with durable goods then FMCG’s, commonly they buy more FMCG’s, with the consequence that consumers have less parent brand knowledge and associations. Yeung and Wyer (2005) found that the evaluation of an extension would be more dependent on the similarity to the core brand when a core brand did not elicit effect; consumers have no subjective feelings that they experience when they encounter a brand. This is a possible subject for further research.
For durable brand extensions, the linkage of the utility of the parent brand to specific product attributes and the history of previous brand extensions are important determinants of the success of a brand extension to managers.
For the linkage of the utility of the parent brand to specific product attributes Rangaswamy, Burke and Oliva (1993) found a negative influence, so the less the utility of the parent brand is linked to product attributes of the original product category, the more successful the extension. This is in contradiction with the results of this research about the success of durable brand extensions; the more closely the associations of the parent brand are tied to the attributes of the original product category, the more successful the brand extension. This difference can be explained by the theory of Meyvis and Janiszewski (2004). They propose that brand extension success depends on the accessibility of the benefit associations, and that the accessibility depends on the amount of interdependence by competing brand associations. An implication of this is that broad brands tend to have more accessibility benefit associations than narrow brands, and in conclusion broad brands can engage in more successful extensions than narrow brands. In this research three brands (Sony, Philips and Canon) that can be seen as broad brands were introduced. These brands can have more accessibility benefit associations, and can be more successful with brand extensions.
brand extensions (Dacin and Smith, 1994) and when the products affiliated with the brand are successful (Aaker and Keller, 1990).
However, this effect is not found in this research about the success of durable brand extensions, but quite the opposite; the higher the number of previous brand extensions, the more success of products affiliated with the brand, the more successful the durable brand extension. So, in contradiction with earlier research, this effect is a negative one. Meyvis and Janiszewski (2004) have predicted that this could happen. According to them a diverse portfolio of brands can have advantages and disadvantages. On the one hand, the diversity of past extensions creates a more diffuse brand image. On the other hand, it makes the brand more acceptable in other distinct categories. To make it clear, an example; a brand may be known for a benefit association, its durability. All of the products of this brand are kitchen appliances, which is a non-benefit association. These different associations are not equally accessible. Due to interference effects in memory, the accessibility of a brand’s association, and also the strength of the positioning, is influenced by the accessibility of the category association.
A direct effect of the quality of the parent brand on the success of a durable brand extension is not found. Earlier research, focused on FMCG’s, were in disagreement about the direct effect of the quality of the parent brand, which is mentioned in paragraph 3.1.1. So, the conclusion of Smith and Park (1992), that there is a positive relation between the quality of the parent brand and the success of a brand extension is not supported in this research. A possible explanation for this is that in this research three imaginary brand extensions were used, and only 2,8% of all the brand extensions is presupposed as low qualitative by the respondents. In further research, this can be explored more widely.
The last direct effects that are not significant in this research are the extension’s product category characteristics , the perceived risk and the consumer innovativeness. A possible explanation for this is that the target group in this research, high-educated young people, is less influenced by the perceived risk and the consumer innovativeness. Even though these effects were significant in the research of Völckner and Sattler (2006), their conclusion also was that these determinants are no important drivers of the brand extension success.
6.1.2 Mediating effects
(Völckner and Sattler, 2006). The outcome of this research is that the higher the perceived risk, the higher the innovativeness. An explanation for this is that Völckner and Sattler (2006) did door-to-door interviews, through which the population was quite different than the high-educated young people. Further research, with a more versatile population can be done to test this possible explanation.
6.1.3 Moderating effects
None of the moderating effects have a significant effect on the success of a brand extension. Völckner and Sattler (2006) found that two interaction effects (fit x quality ; fit x parent brand conviction) were significant, but the contribution of the moderating effects were small in comparison to the direct and the mediating effects. So, in this research the moderating effects play a relative minor role, and the incremental variance explained by the moderating effects is small and not of significant influence. The conclusion is that managers have to focus on the direct and mediating effects.
6.2 Measures of fit
In previous research a lot of research is done about the variable of fit, as the most important effect on the success of a brand extension. In these researches different results were found. One of the issues that previous researchers could not agree on is the relative importance of the different measures of fit (mentioned in paragraph 3.1.3). These will be discussed further in this paragraph.
Sunde and Brodie (1993) did a replica of the research of Aaker and Keller (1990), with the three measures of fit as well. Their conclusion is, in contradiction with Aaker and Keller, that there is a relationship between the positive quality image of the parent brand and the evaluation of a brand extension. And this is not only strong when there is a basis of fit between the two product classes. Concerning the measures of fit their conclusion was the same as Aaker and Keller, ‘transfer’ and ‘complement’ are the most important measures.
However, in the replicas of Bottomley and Doyle (1996) and Bottomley and Holden (2001) the conclusion was that all three measures of fit are important, and significant predictors of how consumers evaluate a brand extension. In these two researches the residual-centering method of Lance (1988) was used, the reason for this was to counteract the problems around the multi-collinearity. Another result in these last two researches was that the interaction between the quality of the parent brand and the fit only created synergetic effects in the evaluation of the extension in the measurements of fit ‘transfer’ and ‘complement’.
However, in line with Smith and Park (1992), Völckner and Sattler (2006) found that the quality of the parent brand has a significant direct effect on the extensions success (with an estimated structural parameter of + .10).
In the questionnaire all three items were asked for.
Complement : There is a global similarity between the (parent brand) and the (extension product) Substitute : The specific associations with (the parent brand) are also relevant in (the extension category)
Transfer : The people, facilities, and skills used in making (the original product category) are helpful if the manufacturer of (the parent brand) were to make a (extension product)
