FA C IN G
IN NOVA TIO N
WI TH
BRAND
EX TEN SIO N
Autonomous Cars Brand Extension Drivers
by
Ronny Giovanni De Salvador Ortega
FA C IN G
IN NOVA TIO N
WI TH
BRAND
EX TEN SIO N
Autonomous Cars Brand Extension Drivers
by
Ronny Giovanni De Salvador Ortega
Faculty of Economics and Business Marketing Department
Marketing Intelligence Thesis
[16 of January 2017 ]
Address: Winschoterdiep 46 D107, 9723AC Groningen Phone number 31 6 38 71 76 78
Email aljro78@hotmail.com Student number S 2970589 Supervisor Felix Eggers
Preface
In a world where markets are saturated with products, innovation is the only way for companies to stay competitive. Brand extensions have been for many years used on innovative products, however most of the literature had focused on incremental innovation and neglected the disruptive innovation. This type of innovations are not very frequent but they are the ones that generate the deepest changes in the long term. One of my interests is to study this type of innovation and understand the different business strategies that facilitate their integration into the company's vision.
TABLE OF CONTENTS
1 INTRODUCTION ... 6
2 THEORETICAL FRAMEWORK ... 9
2.1 From attitudes toward brand extension to brand extension preferences ... 12
2.2 Determinants of Brand Extension Success ... 12
2.2.1 Fit ... 12
2.2.2 Brand Parent Characteristics ... 13
2.2.3 Brand Extensions Product Category Characteristics ... 14
2.2.4 Marketing support ... 15
2.2.5 Moderating effects ... 15
2.2.6 Control variables ... 18
2.3 Conceptual Model ... 19
3 DATA AND METHODOLOGY ... 20
3.1 Choice based Conjoint analysis ... 20
3.1.1 Absolute and relative attribute importance ... 21
3.2 Experimental design ... 22
3.3 Measures ... 23
4 RESULTS ... 25
4.1 Description of data ... 25
4.1.1 Sample Characteristics ... 25
4.1.2 Level of involvement and consumer innovativeness indexes. ... 26
4.2 Choice Based Conjoint analysis ... 26
4.2.1 Assessment of Model fit ... 28
4.2.2 Assessment of Brand extension success factors and other attributes on autonomous cars 29 5 DISCUSSION ... 33
6 SUMMARY IMPLICATIONS AND LIMITATIONS ... 38
REFERENCES ... 41
Table of figures
Figure 1 Conceptual model ... 19
Figure 2 Exemplary choice set ... 23
Figure 3 Example of 0/1 coded responses (check=1 and no check=0) ... 23
Tables Table 1 Brand extension studies from 2006-2014 ... 10
Table 2 Levels and attributes of the model ... 22
Table 3 Success Factor Measures ... 24
Table 4Percentage of respondents who perceive that brand extension factor on the brand ... 25
Table 5 Estimates of conjoint models and relative attribute importance ... 27
Table 6 Goodness of fit. Likelihood ratio test ... 28
Table 7 Information criteria ... 29
Table 8 Likelihood ratio between models ... 29
Table 9 ... 31
Table 10 Maximum utility by brand ... 32
1 INTRODUCTION
The strategy of using an established brand name to launch new products or to reach new markets -Brand extensions- has been recognized as important issue among companies1. Therefore, it has
been an essential topic in marketing research during the past two decades (Sattler,Völckner,Riediger, and Ringle., 2010; Kim, Park & Kim 2014). For companies, a beneficial brand extension strategy can reduce new products introduction costs (Collins-Dodd and Louviere 1999; Henseler ,Horvath, Sarstedt & Zimmermann, 2010 ;Smith and Park,1992; Tauber 1988 ; Volcker & Sattler,2006), mitigate risk in new products category (DelVecchio, 2000; Milberg, Sinn & Goodstein, 2010) and increase parent brand image (Balachander and Ghose, 2003).
Despite of the advantages and the huge popularity of brand extensions, several studies have shown that there are determinants that can derail this strategy2, In this regard, marketing research has
focused on identifying and measuring these success factors. .
There are a considerable number of studies focused in identifying and analyzing the success factors. Starting with the quality of the parent brand and the fit between the parent brand and the extended product category (Aacker & Keller, 1990); consumer characteristics as beliefs and emotions (Pina, Riley & Lomax, 2013); extended category attributes, market structure and brand extension marketing context (Kapoor & Heslop, 2009; Volcker & Sattler, 2007; Sattler,Völckner,Riediger, and Ringle., 2010).
However, according to the literature review of this study, it was found that no research has yet analyzed brand extension drivers for innovative products3. At the same time few studies includes in
a model4, the competitive effects of the brand extension market on the preference of brand
extensions. (Kapoor and Heslop, 2009; Milberg, Sinn & Goodstein, 2010).
Due to the limitations founded in previous studies,the aim of this study is to extend the current framework of brand extension success drivers to innovative products and also to model the effects of competition in the preferences of brand extension. So the study focuses in the following research question: what are the drivers of brand extension success for autonomous cars?.
1 Simms (2005) identified 82 per cent of new product introductions as brand extensions.
2 80% on brand extension Failure rates in FMCG product categories Ernst & Young and ACNielsen 1999; Volcker & Sattler, 2006)
3 Most of the previous studies have developed models to identify success factors in FMCG products category.
In order to answer the research question, This study use Choice based conjoint analysis that will expand the field of knowledge by including the effects of consumer innovativeness, level of involvement and competition in brand extensions and also by determining the importance and relevance of the brand extension success factor previously founded in the theory. The use of conjoint analysis in this study has an advantage over previous studies that is the simultaneous assessment of more than two brand extension success factors in a competitive setting for the evaluation of brand extensions.
The managerial implications of this study are related with the marketing practice in the brand extension context. Which are the most important success factors? and how managers should distribute the resources of the company according to these factors in an innovative and competitive environment.
The category of autonomous cars was chosen because it allows to evaluate the Brand extension success factors in a scenario where the attributes of the product are not yet defined. By doing this, it is possible to validate and generalize the results of previous studies that were based on consumer well known products.
With regard to autonomous cars technology product category, over the past few years, many advances in technology has been made to allow cars drive themselves using the existing road infrastructure and facing different environmental contexts. Assuming that these technologies become successful in a near future , autonomous cars could become a possible effective response for several urban mobility challenges, mostly given an increasing number of overpopulated cities. (Scott Le Vine, Alireza Zolfaghari, John Polak,2015).
This technology also has the potential to reduce crashes, energy consumption and pollution. At the same time, it could improve urban land use and increase the time for performing work or leisure activities . However, like other new technologies, self driven cars could also have negative consequences like increases in the number of car accidents if the system fails. (Rand.org, 2014). In this context, several renowned companies 5 from different sectors, had understood the
self-driven cars technology as an unique opportunity to increase their profits. It will improve their long term strategic position in the mobility sector, but also it has considerable financial and legal,risks among others. (Fortune, 2016)
5 Google, Apple, Uber, Tesla Motors, General Motors,Ford, Toyota, Nissan, Volkswagen,
Other aspect included in this study is the consumer perception regarding this category product, Is it this product a natural development of the automobile industry? Or is it a revolution of the technology and services industry?. This is important because it will affect the perceived fit between parent brand and the extension product.
2 THEORETICAL FRAMEWORK
As in many previous researches on Brand extension, this study uses a systematic literature review to identify which determinants of extension success had been proved significantly important across studies. In this case, the literature review used twenty four studies (see table 1 rows) on brand extension evaluation for years 2006 - 2014 and twelve suggested variables (see table 1 columns) from the conceptual model of Volcker & Sattler, (2006), the results of its statistical models were checked in order to classify the variables depending on its significance level (significant, partially significant, not significant, no tested).
In brand extension literature, a starting point is the work of Aaker and Keller (1990), where they found in general three aspects on the success of brand extension products: First, the success of the extension depends on an extension product that is closely tied with the offerings of products and services of the brand or it converges to image of the brand. Second, the capacity to transfer the skills, people, assets and resources related with making the original product to the extension product and Third, the quality of the Brand. Fit or similarity could be summarized as the union of the first two aspects and the third aspect is itself another independent success factor. Following studies had showed that this two factors are still important and are the basis of the following evolution stages of brand extension modelling. Volcker & Sattler 2006 made an extensive analysis of the literature using studies from 1985 to 2001 and identified twelve determinants of brand extension success (including fit and the quality of the brand) that had been relevant on those studies. In our case, the Volcker and Sattler study allows to extend the time frame of the analysis on brand extension (twenty four years of research effort) and it facilitated the inclusion of new developments by employing a common recognized language (convergence in the variables definition) that helps to bridge the previous theoretical framework. The results of this study literature review validated the significance of several of Volcker and Sattler identified factors and also found new variables that could be included in the model of brand extension. To summarize, it was found that Fit, Quality strength of the parent brand, Parent Brand conviction, Parent Brand experience, Perceived Risk, Marketing support are significant for explaining brand extension success (see table 1).
related with the parent brand and also with the extension product category are important issues on the cognitive process of associations between parent brand and the extension product. Moreover, consumer’s characteristics like the level of involvement have begun to be taken into account because it could influence the weights that costumers gives to the different success factors. This study systematic review also found that there are very few studies that simultaneously assess the product category competition settings and level of consumer involvement in the brand extension evaluation. Furthermore, many of these brand extension studies were based solely on FMCG products limiting the generalizability of the results to other product categories. Therefore, this study extent the scope of the analysis by testing the generalizability of the success factors using autonomous cars as a new type of extension category in a competitive context.
Table 1 Brand extension studies from 2006-2014
√ —> tested and significant
(√) —> tested but only partially significant (not for all categories that were included) n.s. —> tested but not significant
In the literature review (table 1), the most used variable in brand extension success factor studies is Fit with twenty two studies; in nineteen of these studies this variable was tested and is statistically significant and in three this variable was tested but is only partially statistically significant. Quality (strength) of the parent Brand is the second most used variable with nine studies, in all of them this variable is statistically significant. This results confirms both variables as the basis of brand extension modelling
In third position is Parent brand conviction with nine studies, in four studies this variable is statistically significant, two only partially statistically significant and three studies is not statistically significant. In the studies were this variable appear, fit also plays an important role as a predictor variable of brand extension.
Perceived risk ranked four with five studies, in four studies this variable is statistically significant and one only partially statistically significant. In these studies, this extension product category characteristic variable is usually accompanied by Fit and quality of the brand.
Fifth place goes to Marketing support with only three studies, in two studies this variable is statistically significant and in one only partially statistically significant. In these cases brand extension success depends also on fit, quality of the brand and perceived risk.
Parent brand experience has two studies and both are statistically significant. It is important to highlight that this variable had be used with fit and perceived risk not only to measure attitude towards brand extension but also purchase intentions.
Lastly there is Brand extension market characteristics with two studies, in one of these studies this variable was tested and is statistically significant and in the other study this variable was tested but is only partially statistically significant.
2.1 From attitudes toward brand extension to brand extension preferences
As it will be seen later, the methodology used in this study allows us to go beyond describing attitudes to describe intentions or preferences related to brand extension products. The Theory of planned behavior postulates that individual behavior including adoption intentions is determined by attitudes toward behavior, subjective norms and perceived behavioral control (Ajzen, 1991; Taylor & Todd, 1995). In this context, the previous results which apply in most of the cases towards brand extension attitudes could be extended to describe the consumer preferences towards brand extension. In our case, autonomous cars use intention (brand extension preferences) is explained by brand extension success factors along with other product attributes (level of autonomy, price). So the brand extension factor will influence how often an autonomous car alternative is used. 2.2 Determinants of Brand Extension Success
2.2.1 Fit
As it was mentioned before, this variable represents one of the main mental structures of brand extension evaluation. In our evaluation of brand extension on autonomous cars, Fit comprises high global similarity and consistency between the parent brand and the autonomous cars (Aaker and Keller ,1990; Sattler,Völckner,Riediger, and Ringle, 2010), high ability to produce the autonomous car by the company owner of the parent brand (Aaker and Keller ,1990; Volckner and Sattler, 1996), high capacity to transfer brand associations and attributes to autonomous cars (Broniarczyk and Alba ,1994; Seltene and Brunel, 2008; Volckner and Sattler, 1996). In previous studies where fit was significant (see table1), positive fit drove to positive brand extension attitudes; if there is a level of match between the perceived properties of the brand and the perceived properties of the extension product then it will generates a positive attitude towards brand extension (Volckner and Sattler, 1996). In other words, the brand fit could be seen as an attribute of the brand extension product that helps the customer to make a decision.
2.2.2
Brand Parent Characteristics2.2.2.1 Quality strength of the parent brand
This variable had been usually used since the work of Aaker and Keller (1990) on evaluation of brand extension and as it has been shown from the literature review that this variable still important to explain a good performance of the extension product (services and FMCG products). According to Keller & Aaker (1992), high quality brands will perform better than average quality brands on brand extension. In addition to this, other studies6 recall quality strength of the parent brand as a relevant
predictor of consumers brand extension evaluation (Bottomley, & Holden 2001; Sichtmann & Diamantopoulos, 2013). In autonomous cars, a special situation emerge and this is because autonomous cars is a new complex product with many unknown attributes in a new market. So in this case the quality of the brand is a signal to infer the quality of the extension product, by three ways (DelVecchio, 2000): First is that the brand becomes a real guarantee to the consumer, if this product (autonomous cars) fails to deliver completely satisfaction, the brand loses value and hence also the company loses equity value. Second, a good quality strength brand communicates the high level of skills and capacity that the company has to produce the autonomous cars. Third, the quality of the brand allows the costumer to compare the autonomous cars of the parent brand with the autonomous cars of other brands and resolve his/her preferences for the extension product. So the higher the quality strength of the brand, the higher the guarantee and the perceived level of skills and capacity of the company and also higher the competitive advantage that translates in a positive evaluation of the extension product.
H2. Perceived parent brand quality has a positive effect on consumers’ attitudes toward autonomous cars
2.2.2.2 Parent Brand conviction
Defined as greater liking and trust in the brand name (Kirmani, Sood, and Bridges ,1999); Volcker & Sattler,2006; Klink, & Smith, 2001), this variable includes all the emotions, beliefs and affections for the parent brand. The relevance of this variable is because it takes into account part of the cognitive process that is often not explained by brand quality and brand experience, for example a consumer could score a brand extension from a new company higher compared with a more traditional and
experienced brand because it could accommodate better the customer self-concept or serve as a mean to achieve his/her goals (Spiggle, Nguyen, and Caravella ,2012). In Fedorikhin, Park, Thomson, (2008), the greater liking increase the emotional bond between the customer and the brand and therefore increase attachment to the brand extension. This attachment drives finally positive behavioral resources towards the brand extension product. In the review, for all the studies were some level of significance was found in Parent brand conviction, a positive effect of this variable was found on attitudes toward brand extension product (Seltene and Brunel, 2008; Fedorikhin, Park, Thomson, 2008; Milberg, Sinn & Goodstein,2010; Pina, Riley & Lomax, 2013; Kim,Park,, Kim, J., 2014; Batra, Lenk, and Wedel ,2010). In general, people with more favorable attitudes toward the parent brand will likely to respond in a positive way to brand extensions (Kirmani, Sood, and Bridges ,1999). For our study, we expect a similar positive effect of conviction in brand extension on autonomous cars category.
H3. Parent-brand conviction has a positive effect on consumers ‘attitudes toward autonomous cars. 2.2.3 Brand Extensions Product Category Characteristics
2.2.3.1 Perceived Risk of unknown brands
The perceived risk of unknown brands in the brand extension product category plays an important role in the evaluation of brand extension. When the level of perceived risk in the extension product category increases, customers will diminish their disposition for trying the new product (Volcker & Sattler,2006). A possible cause of high level of perceived risk is consumers lack of experience in the extension product category (DelVecchio,2000;Volcker & Sattler,2006). Under these circumstance, the use of the brand name in the new product will reduce the perceived risk in the extension product category and will increase the willingness to try this product. So in scenarios where the level of perceived risk is high but below the consumers maximum affordable perceived risk (willingness to take risk), the brand extension product will be his/her most preferred choice. On the contrary, if the level of information and experience on the extension product category increases, they could prefer other brands . (Klink & Athaide, 2010; Hem, L.E.; Iversen, N. M.; Olsen, L. E., 2013)
H5. Higher perceived risk in the extension category compared to the brand extension product will has a positive significant effect on consumers’ attitude toward the brand extension.
2.2.4 Marketing support
Brand extension product with a properly marketing support is a successful new product introduction (Klink & Smith, 2001; Sattler,Völckner,Riediger, and Ringle , 2010). In a brand extension context, potential costumers must be aware of the brand extension product and must know where they can find it. A brand with a good marketing strategy could even persuade a prospective consumer to try out the brand extension product. In the autonomous cars category is important that the customer knows the existence and attributes of the product and also which needs could be satisfied with it and where and when the self-driven vehicles can be used. If the company fails to provide an appropriate marketing (advertising) support to the brand extension product, the costumer will consider other brands or even not consider the product at all.
H6. The perceived availability of the brand extension product has a positive effect on consumers’ attitudes toward the brand extension.
2.2.5
Moderating effects2.2.5.1 Fit x Quality strength of the parent brand
The success of brand extension depends not only on a good transfer of the consumer brand associations to the brand extension product but also the valence of those associations. When those brand associations are positive (good quality strength brand), consumers are less strict about accept them as attributes on the brand extension product. Therefore, this brand associations becomes new product attributes, an a higher fit will result. Finally, this higher fit will allow a higher degree to which new brand associations are transferred to the extension (Sattler,Völckner,Riediger, and Ringle , 2010; Oakley, Duhachek, Balachander, & Sriram,2008). In a similar way, Batra, Lenk, and Wedel (2010) revealed that fit is the result of prior attitudes from the parent brand. In autonomous cars, it is expected also a positive interaction between fit and quality strength of the parent brand but with the particularity that the brand is the main information source about the brand extension product
2.2.5.2 Fit x Parent brand conviction
A brand which are based on more abstract concepts can stretches more than a brand positioned on functional attributes . (Alokparna and Deborah , (2010) ; Albrecht, Backhaus, Gurzki, and Woisetschläger 2013). In general, depending of the parent brand characteristics, Consumers will construct different attribute relationship (networks) that will connect or not with attributes of the extended product. When the parent brand is more abstract than functional, is easier to link those attributes with the attributes of the extended product, making fit significant in the analysis of the Consumer. In Sichtmann& Diamantopoulos (2013) , brand origin is used to associate feelings and perceptions with the extended product category.
In Fedorikhin, Park, Thomson, (2008) and Kim,Park,, Kim, J., (2014), when the fit is moderated or high, parent brand conviction for the brand becomes a significant brand extension predictor. Normally, high parent brand conviction generates positive emotions that will be used to retrieve the consumer cognitions about the brand but, if the fit is low, those consumer cognitions couldn’t be employed to evaluate the convenience of brand extension.
Similar to the quality strength of the parent brand, this factor increases fit of the autonomous cars, and because the product category will depend in the early stages of the market on abstract concepts rather than specific functional attributes, the result is that emotions and beliefs becomes an important cue for consumers evaluation
H8. An interaction between fit and parent brand conviction has a positive effect on consumers’ attitudes toward the brand extension.
2.2.5.3 Fit x Consumer characteristics.
Both level of involvement and consumer innovativeness are personality traits that remains constant between alternatives when the respondent make his/her choice and because of that, their effects on brand extension preferences cannot be measured via conjoint analisys directly. Therefore, they should be explained by a moderating effects. (Bearden, Netemeyer & Teel, 1989)
2.2.5.3.1 Fit x Level of involvement
The level of involvement affects consumers capacity to process information about the brand extension (Ahluwalia, 2008; Alokparna and John, 2007; Torelli and Ahluwalia 2012). In Ahluwalia (2008), with moderate fit, a brand extension has a reasonable chance of success if the extension market tends to include highly relational thinkers with high motivation to elaborate. Likewise, Alokparna and John (2007),found that fit is an important variable to explain extension success but at the same time they observed that this variable would have a magnified effect if consumers were allowed to think holistically , in particular for functional brands.
H9. Consumer high level of involvement in the extension category has a positive effect on fit toward the brand extension.
2.2.5.3.2 Fit x Consumer innovativeness
In (Klink, & Smith, 2001) there are two effects of Consumer innovativeness on Brand Extension, one direct and the other indirect. The direct effect is due to the innovators and early adopters who are more willing to try new goods and services and as a consequence they will react more positively to a new product on brand extension than followers or late adopters. The indirect effect is the result of consumer innovativeness negative impact on fit. In the latter case, low risk aversion consumers (high consumer innovativeness) accept more easily low fit brand extension so in these cases fit diminishes its impact on brand extension evaluation.
This study assumes that the direct effect will be through perceived risk of unknown brands because when new products come about, they are only adopted by people who are comfortable taking the high risk always associated with new products (Hirschman, 1980; Klink, & Smith, 2001). But for the indirect effect, consumer innovativeness is used a moderator of fit where higher consumer innovativeness will diminishes the fit between the parent brand and autonomous cars and on the contrary less innovative consumer will require a higher fit so its importance increase on brand extension success models.
2.2.6 Control variables
2.2.6.1 Autonomous cars non-brand attributes
Besides the Brand, the choice model also uses price and level of autonomy as autonomous cars attributes in the choice sets to show scenarios close to reality with which the consumer has to deal with.
2.2.6.2 Brand Familiarity
Familiarity with the brands was included in the model to control the validity of the results. In Volcker & Sattler (2006), before running their brand extension model, they made a pretest that help them to select only well-established brands which they used to properly account for the total brand effect in the model.
2.2.6.3 Brand extension Market (level of competition)
3
DATA AND METHODOLOGY 3.1 Choice based Conjoint analysisThis study employs Choice-Based Conjoint analysis to test consumer brand extension preferences choices using success factors considered in the theory. This technique allows to uncover respondent preferences by choosing his/her most preferred option from a set of options similar to people every days situations. In this descompositional approach a subset of alternatives7 or combinations of
attributes (each having different levels) is presented to the respondents and they are asked to select their most preferred option. (Louviere et al. 2000; Louviere and Woodworth 1983)
According with random utility theory, the overall utility U of an individual i for an object j is defined by
Equation 1
= +
is a systematic component or systematic utility and is an error component, that includes all the effects that are not accounted for (Manski, 1977; Eggers, Eggers & Kraus, 2014).
The systematic utility of the alternatives is usually specified as a linear combination of observed attributes and can be estimated using a multinomial logit model where the dependent variable is the probability that respondent i chooses a from a choices set with J alternatives (Lausen, Krolak-Schwerdt, & Böhmer,2015)
Equation 2
( |J) = exp( ) ∑ exp ( )
It is assumed that the brand extension preference for autonomous cars is a function of Fit (FIT), Parent Brand Familiarity (BFAM), Quality strength of the parent brand (BQUALITY), Parent brand conviction (BC), Parent brand experience (BEXPERIENCE), Perceived Risk of unknown brands (BEPRISK), Brand extension Marketing Support (BEMS), Consumer
innovativeness(INNOVATIVENESS), Consumer level of involvement (INVOLVEMENT), Level of autonomy (LA), Brand residual and Price per hour of autonomous car use (PRICE) .
Equation 3 = + + + + + + + + + + + , ∗ + , ∗ + , ∗ + , ∗
3.1.1 Absolute and relative attribute importance
The absolute importance shows the maximum effect of each attribute on consumer utility. It is defined as (Vermunt & Magidson, 2005; Eggers, Eggers & Kraus, 2014):
Equation 4
= (max | − min | )
Where the utility associated with level a and attribute p | is equal to:
| =
=numeric score of level a for attribute p
= effect of attribute p
= effect of attribute p on level a
And the relative attribute importance measures the importance of an attribute compared with the total importance across attributes in the utility function. In this case, Relative attribute importance is calculated by taking the absolute importance and divided by the sum of ranges across all
attributes. This measure is rescale to sum 1. Equation 5
The relative attribute importance calculated in this study doesn’t include the none option. 3.2 Experimental design
In this study design (see table 2), each autonomous car stimulus was described by three attributes: brand, price and level of autonomy. Both, brand and price attributes used six varying levels and for Autonomy only two levels were considered. For brand, five brands with autonomous cars projects were chosen from the automobile and software & technology sectors and one additional unknown brand was included as a benchmark8. The price levels were selected looking for an
approximate representation of the car renting price per hour in the market and for the autonomy attribute, two levels were used according to the currently available products and research
projects.
Table 2 Levels and attributes of the model
To complete the experimental design, it was used a total of eleven choice sets that presented a dual response choice design in which respondents first select their preferred alternative from among the three autonomous car options offered to use in the choice set and then choose among their preferred alternative and a no use choice option9 (see figure 2 for an example). The stimuli was
controlled for the efficient criteria of level balance, orthogonality and minimal overlap (Eggers, Eggers & Kraus, 2014). The missing observations were controlled by using only mandatory questions in the online survey.
8 The use of a unknown brand as a benchmark allows to include the effect of brand equity for each brand in the model. In Keller (1993), Brand equity from costumer perspective is defined as the different consumer response derived from the brand knowledge.
9 According to Wlömert & Eggers (2016), there is three possible advantages using this format: First, it increases the salience of the use decision. Second, it increases the cognitive effort used by respondents by making relevant the choice of the alternatives and third, it gives always preference information independent of the use decision.
Brand Autonomy Price
1. Tesla Motors 1. Partially
self-driving 1. €10
2. Apple 2. Fully self-driving 2. €12
Figure 2 Exemplary choice set
3.3 Measures
To measure the effect of all success factors (except consumer characteristics) in brand extension preference using choice based conjoint, it was designed a 0/1 coded responses in the questionnaire where the respondents decide to assign or not a success factor to each brand (see fig 3 for an example). For consumer characteristics, it was built eight questions from which four questions were to measure the level of involvement and the other four were to measure consumer innovativeness. The level of involvement is composed by: I enjoy driving a car on my own (7-point rating scale), I enjoy being a passenger in a car (7-point rating scale), I use a car very often (7-point rating scale), I am very interested in cars (7-point rating scale). The items used to construct consumer innovativeness are: Overall I enjoy buying the latest products (7-point rating scale), I like to buy new products before others (7-point rating scale), If I choose self-driving cars, I would feel very uncertain of the level of quality that I am getting (7-point rating scale), I prefer choosing a well-known self-driving car because I need the reassurance of an established brand offering this product (7-point rating scale).
In table 3 is a summary of the different success factors in brand extension used in the study with their respective measure based in Volckner and Sattler (1996) and the literature review.
4 RESULTS
4.1 Description of data 4.1.1 Sample Characteristics
To collect the data an online survey was conducted in english for participants from around the world during two weeks (22 of November to 5 of December of 2016). A total of 339 respondents completed the survey where 48,1 % were males and 51,9% were females. Their age ranges were: 18 to 25 years (65,49%), 26 to 35 years (26,55%), 36 to 45 years (4,42%), older than 45 years (3,54%). The countries with more respondents come from Netherlands (n=182; 53,7%), Colombia (n=30; 8,8%) and China(n=28; 8,3%) which all three accounted for 70,8% of the observations.
Table 4Percentage of respondents who perceive that brand extension factor on the brand
Having in mind that Robocab was included to check the validity of the observations , It can be observed (see table 4) that all the brands are familiar to the respondents (>=84%) and most of them have good quality strength of the brand(Tesla, Apple, Google and Mercedes Benz).
The majority of the respondents also trust in all 5 brands (>50%). When the respondents were asked about the experience on the brand, only Google has been tested by the majority of the respondents (BEXPERIENCE =87%). Most of the respondents expect Tesla and Mercedes Benz will deliver attractive autonomous car products. Ford scores low in both image fit and category fit and apple has a low score in Category fit. Apple is considered the brand with the highest risk for autonomous cars followed by Ford and Google. Only Tesla and Mercedes have a good percentage of the respondents that feel confident in, when buying a self-driving car.
Brand extension success factors Tesla Motors Apple Google Mercedes Benz Ford Robocab
4.1.2 Level of involvement and consumer innovativeness indexes.
For a good estimation of the different models and a correct test of the hypotheses, the correlations of both the four items for level of involvement and the four items for consumer innovativeness were checked. In the case of consumer innovativeness, it was found that “Overall. I enjoy buying the latest products” and “ I like to buy new products before others” are interrelated (α=0,801) so a new variable that is the average of these two items was constructed10. The remaining items were used
separately in the models because both factor and reliability analysis indicates a very low interrelation (α=0,398). (See appendix 1). For level of involvement a index was constructed using the factor score determined by the factor analysis from the items “I use a car very often”, “I enjoy driving a car on my own” and “I am very interested in cars” (KMO=0,656 , Bartlett sig= 0, α=0,674)11
(See appendix 2). The remaining item “I enjoy being a passenger in a car” was used separately in the models.
4.2 Choice Based Conjoint analysis
Several multinomial logit models were estimated on an aggregate level using maximum likelihood procedure. The levels of Brand, Level of autonomy and Price per hour are defined as nominal attributes (part-worth model for each of these attributes) in the model. For all the brand extension success factors a vector specification (linear) is used for each of these attributes12. Finally, the none
option is included as numeric (linear model) in the models.
The idea behind the estimation of several models is to evaluate the effect of the brand extension factors and moderators on consumer brand extension preferences. First, it is presented a benchmark model that consists only of the estimates for Brand, Level of autonomy and Price per hour (Model 1). Then model 1 is compared with the extended models Model 2 (in which it is included all success factors considered in the conceptual model), model 3 (Model 2 plus all Moderating effects) and model 4 (model 2 plus only significant moderating effects) . See table 5
10 Factor analysis could not be performed because KMO was lower than 0,6 (see appendix 1) 11 Factor analysis results in a dimension reduction from 3 to 1 variables (eigenvalue =1,819 >1 and
a 60,63% of the cumulative variance explained) . In addition , Factor score were obtained using varimax rotation allowing no correlation between factors.
12 Because it was used 0/1 coded responses for the brand extension success factors (except
4.2.1 Assessment of Model fit
From table 5 , all models have Pseudo-R2 adjusted larger than 0,2 and can be considered acceptable and Model 4 has the larger Pseudo-R2 adjusted from all the estimated models (0,294) and it is followed by Model 3 (0,293).
Also a likelihood ratio test is used to check if these models are better than a random selected choice (null model) for describing the consumer preferences about autonomous cars. The conclusion is that all the estimated models are better than the null model ( each estimated model has parameters significantly different from zero) . see table 6
Table 6 Goodness of fit. Likelihood ratio test
To test the predictive validity of the models, the hit rate and the Mean Absolute error were calculated for each model. For this, one the eleven choice sets available per respondent was not used for the estimation. Instead, the observed values of this holdout choice set are compared with the predicted values of the models to see how close they are. In table 5, Model 3 and Model 4 have the lower Mean absolute error (6,3%) and these two models have a similar an acceptable hit rate (69,2%) and are only surpassed by Model 2 (69,3%).
Table 7 Information criteria
Table 8 Likelihood ratio between models
(a) (b) (c)
From the assessment of the models fit, the inclusion of brand extension success factors significantly improves the preference and helps to explain the consumer cognitive process behind consumer choice in autonomous cars.
4.2.2 Assessment of Brand extension success factors and other attributes on autonomous cars For a more exhaustive analysis of the models and the variables that compose them, The estimates of conjoint models and the relative attribute importance values are used together (see table 5). The estimated partworth utilities (price, brand and level of autonomy) and linear parameters (brand extension and moderators) are depicted and all are effect coded (in the partworth part ,the sum across the betas for each attribute is zero). Also In Table5, these partworth utilities are transformed so it can be identified which attribute is most influential in affecting consumers preferences (Relative attribute importance)13.
The betas for Level of autonomy and price show face validity across all models as increasing the level of autonomy (price) yields higher (lower) utilities. In addition, in Model 1 (without brand extension success factors and moderators), Tesla Motors is the most preferred brand for autonomous cars followed by Mercedes Benz , Google, Apple, Ford and Robocab (the unknown brand). In the extended models (Model 2, 3 and 4), these brand effects decomposes into brand
extension drivers, moderators and brand residuals. A significant negative residual remains for both Apple and Google which may indicate variables other than Brand extension factors to explain the brand's effects on consumer preferences for these technological companies.
The no-choice option has negative partwoth utility in model 1 so choosing one of the autonomous cars is preferred than choosing none but this changes for Model 2, 3 and 4 where the partwoth utility is positive so in those models the customer could consider this option depending of the other attributes utility values.
Across all models, the γ estimate for Parent brand experience is not significant so H4 is rejected. Perceived Risk of unknown brands is only significant when the model doesn’t include moderators. On the contrary, Brand extension Marketing support is only significant when the model includes moderators. Finally, one of the dimensions of fit (CATEGORY.FIT) is only significant when the model excludes not significant moderators. Model 4 (best fit model according previous section) directly support hypothesis H1 (IMAGE.FIT + CATEGORY.FIT = 0,3), H2 (0,3), H3 (BCTRUST + BCEXPECT +BCCONFIDENCE =1,28), H6 (0,13).
In model 2 (see table 5), the inclusion of the brand extension success factors increase the relative importance of brand from 53,23% to 66.6% and it also allow to identify the brand residual (9,08%) or the part unexplained by the brand extension factors that affect consumer preferences. As we mention before this model performs better than model 1. Going From model 1 to Model 4, the relative importance of both price (from 41,03% in model 1 to 14,85% in model 4) and level of autonomy (from 5,74 in model 1 to 1,78% in model 4)is diminished; Therefore, this means that the brand and brand extension factors plays an important role in consumer decision making when the consumer faces a new and completely unknown environment Then ignoring the Brand extension factors can lead to an overestimation of the non-brand parameters.
Table 9 Total Brand effect
Hence, price (see table 5) has the highest impact on consumers preferences followed by the moderating effect of consumer risk profile on Fit (CATEGORY.FITxreassurance =12,76%, IMAGE.FITtxfeel uncertain =10,24%), the moderating effect of consumer innovativeness on Image fit (IMAGE.FITxinnovativeness =7,92%) and the moderating effect of consumer´s involvement on category fit (CATEGORY.FITxinvolvement =6,32%).
Analyzing the main effects of the brand extension factors on consumer preferences, image fit and category fit together account (without moderators) for 7,82%, brand confidence (5,95%) , brand expectations (4,81%) , brand trust ( 4,79%) and brand quality (3,61%). The remaining significant brand extension success factors ( Brand Familiarity, Brand extension Marketing Support, Perceived risk of unknown brands and brand experience) together accounts 6,45% of the relative attribute importance.
In the case of innovativeness (the two items that were grouped together) , it had a positive and significant moderating effect on fit (IMAGE.FITxinnovativeness γ= 0,11 and pvalue <0,05); for “If I choose self-driving cars, I would feel very uncertain of the level of quality that I am getting” , it had a negative and significant effect on fit (IMAGE.FITxfeel uncertain γ=- 0,14 and pvalue <0,05); and for “I prefer choosing a well-known self-driving car because I need the reassurance of an established brand offering this product” , it had a positive and significant effect on fit (CATEGORY .FITx reassurance γ=0,18 and pvalue <0,05) and with respect to their Relative Attribute Importance, the results were 7,92%, 10,24% and 12,76% respectively.
From the CBC analysis, it can also be obtained along with the characteristics of the brand, the product that maximizes consumer utility. In model 4, the bundle of autonomous car attributes, that maximizes the customer utility are :
Equation 6
Jmax={Fully self-driving, € 10, high brand familiarity, high brand quality, high brand trust, high brand extension preference, high brand extension marketing, high brand image fit, none or very low category fit, none or very low brand extension perceived risk, high brand confidence, high consumer innovativeness, none or very low consumer uncertain of the level of quality of the autonomous car, high reassurance of an established brand, low involvement , car driver}
We proceeded to calculate the maximum utility (model 4) for each brand using the largest positive coefficients of each one of the attributes. In the case of level of autonomy, price a brand residual, the maximum coefficient is used . For Brand extension and Moderators other than level of involvement and consumer innovativeness, the coefficients were multiplied by the percentage of respondents who perceive that brand extension factor on the brand. And, in the case of level of involvement and consumer innovativeness moderators, each coefficient was multiplied by the average of the corresponding variable (these are rating variables).Finally, all the coefficients are summed to obtain the maximum Brand extension utility . In Appendix 3, the values of each attribute used to calculate the maximum utility are shown. Table 10 shows the Maximum utility by brand Vmax by brand where both Tesla Motors and Mercedes Benz have the greatest consumer utility when
all the attributes of the autonomous car are optimized.
5 DISCUSSION
The results of the four CBC models show that the success brand extension factors are important to explain the brand's effect on brand extension preferences using autonomous car. But before estimating models with brand extension success factors and moderating effects , it was found that the brand is the most important attribute when a customer consider to use an autonomous car (attribute importance 53,23%) followed by Price (41,03%). The low attribute importance of level of autonomy indicates that consumers react very little to changes in the levels proposed in the study but always increasing the costumer’s utility when the level of autonomy is increased.
One of the advantages of the methodology used was to consider separately the two measures of fit specified in the literature because after the estimation of the models, it was possible to identify that Category fit had an opposite effect to that predicted in the theory. In other words, brand extensions of brands that currently offer products and/or services that are closely tied to self-driving cars are less effective because the customer begins to depend less on the brand and more on the physical and non-brand attributes of the product to make their decisions. However, the net effect of fit on Brand extension is positive since the customer continues to consider the brand as an efficient tool that identifies a new product in a new market where there is not much information.
The importance of Quality strength of the parent brand in Brand extension was also verified. A strong brand will have easier access to the autonomous cars market where high quality standards are required because the brand will have all the means at its disposal to make more effective the communication of its message to the consumer. Also as it was mentioned above, the customer sees the quality of the brand as a signal to infer the quality of the extension product and also acts as a “collateral” (DelVecchio, 2000) for the use of autonomous cars.
Although the theory indicates that the experience in the brand is important, the results indicate that it is difficult for the customer to relate past experiences on the brand with the valuation and selection of the new product. Within the mass consumption brands considered in the study as Apple, Google and Ford there was no transfer of brand experiences (whether positive or negative) to the new product. According to these results, positive brand experiences (pcs and smartphones software and hardware, web browser and cars to name a few) doesn’t allow to create associated concepts that will help to identify the preferred autonomous car product.
The Higher perceived risk of unknown brands in the extension category was not significant. Under these circumstance, the use of the brand name in a product like autonomous cars doesn’t reduce the perceived risk in the extension product category. As it was mentioned earlier in this study, an unknown brand was included in the model to take into account all possible alternatives and that the estimate was not biased towards known brands. Despite this, the results reveal that for the customer, there is no risk differentiation between the different brands in the autonomous cars category that allows him/her to manage the risk of the category through the use of the brand. Following Volcker & Sattler (2006), Familiarity with the brands was included in the model to control the validity of the results because the successful use of brand extension strategy depends on well-established brands in the market place. The results (BFAM γ= 0,21 and pvalue <0,05)show that most of the brands used in this study are familiar to the consumer and that the model appropriately includes the effect of the brand on consumer preferences towards autonomous cars.
Moderating effects
The interaction between fit and quality strength of the brand has an effect contrary to that predicted in the theory (IMAGE.FITxBQUALITY γ= -0,25 and pvalue <0,05). Positive brand associations related with the quality strength of the brand don’t increase brand extension preference through fit because they can’t be transformed into skills and facilities for creating self-driving cars. On the contrary, these attributes end up decreasing the degree of image fit. In this case, what is happening is that the entry of a strong brand into a risky market such as that of autonomous cars is seen by the customer as out of the range of the brand and negatively evaluates the brand extension. Moreover, a not so well recognized brand can take advantage of the fact that the customer will be less strict when evaluating the fit between brand image and the extension product. Nevertheless its low relative attribute importance (2,99% ) limit its effect on Fit.
In both model 3 and 4, there was no evidence to confirm the interaction between Fit (brand image fit and brand category fit) and parent brand conviction (Brand Trust, Expectations about brand offerings and Consumers brand confidence). Therefore, a brand cannot stretches more in the autonomous car category based on beliefs and emotions to get a higher fit.
Finally, the inclusion of the characteristics of the consumer (Consumer innovativeness and level of involvement ) in the model allowed to identify interactions between these characteristics and Fit. Contrary to what was expected on Klink, & Smith (2001), Consumer innovativeness interacts positively with Fit. This indicates that early adopters and Innovators require a better Fit to prefer one option over another. This may be due to the fact that these types of clients must use more of what they infer from the brand than what they can obtain from the attributes of the product to draw their conclusions and even though they have low risk aversion, their great motivation, knowledge and curiosity makes them more demanding to accept products that do not match well with the brand. In general, this segment of consumers will have a greater utility of the product compared to other consumers if the brand has good Fit with the autonomous cars and, on the other hand if there is no good Fit between the brand and the brand extension product, this type of consumers will have a utility much lower than other consumer segments.
the abilities that the brand has to create Autonomous car will be more negative and this will consequently decrease the Brand extension preferences
The positive effect between “need the reassurance of an established brand” and category fit indicates that if the brand produces products that are closely tied to self-driving cars, this will allow the consumer easy relate the brand with the Autonomous Cars and prefer this brand extension. This effect could disappear when the consumer acquires new information about the product because the consumer will start to depend more on this new information than on the one obtained through the brand.
In Ahluwalia (2008) and Alokparna and John (2007) it was shown that the interaction between Level of involvement and Fit has a significant effect. Even Ahluwalia (2008) goes further, claiming that this relationship is positive for both consumers who are high interdependent self-construal and consumers who are Low interdependent self-construal 14. According to the results this statement is
totally rejected. In this case, as the level of consumer motivation increases, the products and services of the brand are perceived as distant from the category of Autonomous cars. What may be happening is that the consumer with high level of involvement can be looking for associations that allow him to accept the fit between the brand and the product but at the same time negative associations are accumulated that diminishes the effect fit of the brand on the autonomous cars category. Also, as many of the attributes of the autonomous cars are unknown, the starting point (pivot point or reference point) may be an idea that after several experiences can change radically. This process can be accelerated when the person is motivated and as result it will generates constant changes on Fit.
Managerial implications
Apart from validating the different brand extension success factors observed in the literature, this study also found which is the ideal product and how could be optimized the advertising in a scenario where there is no market yet. (see equation 6)
14 Interdependent self construal is related with persons that constructs their self view based in relationship
From the brands considered in the study (see table 10 and appendix 3), Tesla motors and MB are the brands that maximize the utility of the consumer. These brands have in common that they are recognized and with a high position in the market. Moreover these brands generate confidence and in the mind of the consumer they can make attractive products. Although they depend on early adopters and innovators in the early stages of the product life cycle, they can take advantage of the lack of product information to stretch the brand and adjust the product to the stronger brand attributes. Also, It is important that the characteristics of the product are the result of the transference of beliefs in the brand and not the opposite. The latter can be seen as a result of the negative effect of Category Fit on Brand extension preferences and that contrasts with previous studies where Fit had a positive effect.
With regard to the question that at some point was asked about whether consumers perceive autonomous cars as a natural development of the automobile industry or on the contrary as an extension of the technology and services industry, the results are inconclusive.
In appendix 3, it can be seen that Google has a higher brand image fit compared with traditional car brands such as Mercedes Benz or Ford, but is punished for not having a finished product that resembles the autonomous cars, however after including this effect is still above Ford in the consumer preferences.
In addition, having products close to the autonomous cars is not guaranteed to increase consumer preference, one example is Tesla Motors , which despite being one of the brands with significant advances in autonomous cars technology is punished more than the rest of the brands by this concept (-0.13).
Marketing support
Although the effect of advertising on consumer preferences is small. This may be because the question refers solely to the advertising of the product. According to what is observed in the other Brand extension success factors, increasing the brand awareness through messages that are coupled to the consumer's belief system and positively affect the consumers emotions, can have a positive and powerful effect on brand extension preferences.
Messages that not only describe the brand extension product, but also connect it to more ambitious and general goals and place the consumer as the protagonist of a significative social change can generate recognition, loyalty and trust in the brand and this will later help in the evaluation of the brand extension.
Also this messages should communicate trust, confidence and good expectations and in this way, transfer these positive associations to the brand extension product.
6 SUMMARY IMPLICATIONS AND LIMITATIONS
The main objective of this work was to identify the Brand extension factors for a new product whose market does not yet exist. For this, first a bibliographical review of the papers on Brand extension published in the main Journal of marketing was carried out. This allowed to identify the most significant factors and limitations that accompany these studies. Within these limitations, it was found that many studies were based on FMCG and also did not analyze all factors at once. Moreover, a good number of them were made without taking into account the competition and the characteristics of the consumers.
characteristics suggested in the theory. The design of the survey took into account the efficient criteria of level balance, orthogonality and minimal overlap to be able to realize a Choice based Conjoint analysis. Finally, a sample of 339 people from different countries was obtained with an average age of 26,14 years.
Third, the Choice based Conjoint analysis methodology was used to analyze the survey. But before doing this, both the information obtained from the Choice Sets and the information obtained from the other types of question (the 0/1 code and ranking type questions) were prepared. The treatment of 0/1 code type questions was to include them into the Choice sets conditioned to the brands that appeared for each alternative. And the ranking type questions were included in Choice sets conditioned to the respondent that appeared for each choice set.
Factor analysis and reliability analysis were applied to obtain both the Consumer innovativeness and level of involvement indexes that were include in the choice sets as a ranking type question. Four models were estimated and for each one of them the goodness of fit was evaluated . After this, an analysis of the results was conducted to respond the different Hypothesis formulated in the conceptual part of the thesis. It was also studied the cases for which the hypothesis were rejected and possible responses to these cases were proposed.
This study wants to expand knowledge about Brand extensions by using a completely new product on which there is still no market but where different brands have focused their efforts in the long term. As a novelty, a recognized methodology (Choice based Conjoint) was used to study all the brand extension factors together and their effect on brand extension preferences and include both the competition between brands and a more real consumer behavior. In this methodology setting, the customer had to choose from different product options and then he had to choose whether to use or not the product he chose and in this way reveal his true preferences.
One of the findings in this study was that for a new product, brand experience and perceived risk in Brand extension category are not significant and that brand image fit should be considered instead of Brand extension category fit. In addition the connection of positive emotions is essential for the success of the brand.
marketing department can optimize its advertising campaigns and focus its resources and efforts on the most important Brand extension factors.
Within the limitations of the study it is found that in spite of having an international sample, a good percentage of the study participants are from the Netherlands so there may be cultural and socioeconomic effects that may affect the generalization of the results to other countries different from developed countries. Due to the methodology used, it was not possible to obtain the main effects for the variables related to the consumer characteristics and they could only be included as moderating effects in the model. And finally there is a difference between Intention and observed behavior (use) that is described in theory but it can not be quantified because the product is not available in the market. This may lead to consumers' preferences not corresponding to their observed behavior when the service is available in the market.
However one of the advantages of the model is that it can be applied once the product is in the market and the difference between intentions and use can be trimmed by using incentive alignment in the survey (1 free month of self driven car rental). Also this model and its methodology
For future studies it would be important to replicate the Brand extension model to other new products and verify the results. It would also be important to compare the effectiveness of the brand extension with other strategies to launch new products (New brands , Co-branding, etc.). This is because for autonomous cars, the complexity and risk of the project has forced several brands to seek a strategic partner.
REFERENCES
Aaker, D. A., & Keller, K. L. (1990). Consumer evaluations of brand extensions. Journal of Marketing, 54(1), 27−41.
Ahluwalia, R. (2008). How far can a brand stretch? Understanding the role of self-construal. Journal of Marketing Research, 45(3), 337–350.
Ajzen, I. (1991). The Theory of Planned Behavior, Organizational Behavior and Human Decision Processes, 50,179-211.
Albrecht, C.-M., Backhaus, C., Gurzki, H. and Woisetschläger, D. M. (2013), Drivers of Brand Extension Success: What Really Matters for Luxury Brands. Psychol. Mark., 30: 647–659. doi:10.1002/mar.20635
Alokparna, B, M., and John, D,R., (2010),What Makes Brands Elastic? The Influence of Brand Concept and Styles of Thinking on Brand Extension Evaluation, Journal of Marketing vol 74 (3), 80-92.
Balachander , S . and Ghose , S . ( 2003 ) Reciprocal spillover effects: A strategic benefit of brand extensions .Journal of Marketing 67 (1), 4 – 13 .
Barone, M, J.,(2005) “The Interactive Effects of Mood and Involvement on Brand Extension Evaluations,” Journal of Consumer Psychology 15, no. 3, pp. 263–270.
Batra, R., Lenk, P., & Wedel, M. (2010). Brand extension strategy planning: empirical estimation of brand-category personality fit and atypicality. Journal of marketing research, 47(2), 335-347. Bearden, W., Netemeyer, R., & Teel, J. (1989). Measurement of Consumer Susceptibility to Interpersonal Influence. Journal of Consumer Research, 15(4), 473-481.
Bottomley, P. A., & Holden, S. J. S. (2001). Do we really know how consumers evaluate brand extensions? Empirical generalizations based on secondary analysis of eight studies. Journal of Marketing Research, 38(4), 494–500.
Broniarczyk, S, M., and Alba, J,W., (1994), "The Importance of the Brand in Brand Extension,' Journal of Marketing Research, 31(May), 214-28.
DelVecchio, Devon (2000), "Moving Beyond Fit: The Role of Brand Portfolio Characteristics in Consumer Evaluations of Brand Reliability," Journal of Product & Brand Management, 9 (7), 457-71 Eggers, F., Eggers, F., & Kraus, S. (2015s). Entrepreneurial branding: Measuring consumer preferences through choice-based conjoint analysis. International Entrepreneurship and Management Journal. http://dx.doi.org/10.1007/s11365-014-0344-1 (in press)
Fedorikhin, A.; Park, C. W.; Thomson, M., (2008). “Beyond fit and attitude: The effect of emotional attachment on consumer responses to brand extensions”. Journal of Consumer Psychology, 18,281-291.
Fortune. (2016). Tesla Versus GM: Which Self-Driving Car Roadmap Will Win? Available at:
http://fortune.com/tag/autonomous-cars/ [Accessed 25 Oct. 2016]
Gierl, H., Huettl, V., (2011). “A closer look at similarity: The effects of perceived similarity and conjunctive cues on brand extension evaluation”. International Journal of Research in Marketing,28, 120-133.
Hem, L,E., Iversen, N,M. , Olsen, L, E,. (2014) Category characteristics' effects on brand extension attitudes: A research note, Journal of Business Research, Volume 67, Issue 8, , Pages 1589-1594, Henseler, J., Horváth,C., Sarstedt,M. and Zimmermann, L. “A Cross Cultural Comparison of Brand Extension Success Factors: A Meta-Study.” Journal of Brand Management 18, no. 1 (2010): 5–20. doi:10.1057/bm.2010.25.
Hirschman,E.C.(1980). “Innovativeness, Novelty Seeking, and Consumer Creativity. Journal of
Consumer Research , 7, 289-295.
Kapoor, H., Heslop, L. A., (2009). “Brand positivity and competitive effects on the evaluation of brand extensions”. International Journal of Research in Marketing, 26, 228-237.
Keller, K. (1993). Conceptualizing, Measuring, and Managing Customer-Based Brand Equity. Journal of Marketing, 57(1), 1-22. doi:10.2307/1252054
Kim,J., John, D ,R. (2008). Consumer response to brand extensions: Construal level as a moderator of the importance of perceived fi t . Journal of ConsumerPsychology 18 (2) : 116 – 126 .
Kirmani, A., Sood, S., & Bridges, S. (1999). The Ownership Effect in Consumer Responses to Brand Line Stretches. Journal of Marketing, 63(1), 88-101.
Klink, R. R. and Athaide, G. A. (2010), Consumer Innovativeness and the Use of New versus Extended Brand Names for New Products*. Journal of Product Innovation Management, 27: 23–32. doi:10.1111/j.1540-5885.2009.00697.x
Klink, R. R., & Smith, D. C. (2001). Threats to the external validity of brand extension research. Journal of Marketing Research, 38(3), 326–335.
Lausen, B., Krolak-Schwerdt, S., & Böhmer, M. (2015). Data science, learning by latent structures, and knowledge discovery(Studies in classification, data analysis, and knowledge organization; Studies in classification, data analysis, and knowledge organization). Heidelberg: Springer. doi:10.1007/978-3-662-44983-7
Louviere, J. J., & Woodworth, G. (1983). Design and analysis of simulated consumer choice or allocation experiments—an approach based on aggregate data. Journal of Marketing Research, 20(4), 350–367.
Louviere, J. J., Hensher, D. A., & Swait, J. D. (2000). Stated choice methods. Cambridge: Cambridge University Press.
Meyvis, T., Goldsmith, K., & Dhar, R. (2012). The importance of the context in brand extension: How pictures and comparisons shift consumers’ focus from fit to quality.Journal of Marketing Research, 49(2), 206–217. doi:10.1509/jmr/08.0060
Milberg, S. J., Sinn, F., & Goodstein, R. C. (2010). Consumer reactions to brand extensions in a competitive context: Does fit still matter? Journal of Consumer Research, 37(3), 543–553.
Monga, A., John, D., & Dawn Iacobucci served as editor and Laura Peracchio served as associate editor for this article. (2007). Cultural Differences in Brand Extension Evaluation: The Influence of Analytic versus Holistic Thinking. Journal of Consumer Research, 33(4), 529-536. doi:10.1086/510227
