Joint branding alliance & communication : their effect on the central brand image and the transfer of brand and category related associations

(1)

Master Thesis

Joint Branding Alliance & Communication: Their

effect on the General Brand Image and the transfer

of Brand and Category related Associations

Name: Axel Stuiver

Student number: 10730745 Thesis supervisor: Dr. Ed Peelen Date: 05-08-2018

(2)

Statement of originality

This document is written by Axel Stuiver who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

This study aimed to examine the effects of key-value overlap elements in joint branding expressions in the form of advertisements. To check whether a brand with a bad image or an image it tries to (re)define is able to partner up with a brand that has a more positive image. An image that is in line with the image the other brand is trying to (re)define. This study examines if such a partnership and the use of key-value overlap elements in the expressions (advertisements) can explain the fit, lower suspicion and examine if this could boost the negative brand image of the one brand by the transfer of positive associations from the other brand without having a negative impact on the partnering brand. The latter is based on conclusions of previous research regarding Corporate Social Responsibility and the Elaboration Likelihood Model. Furthermore this study uses the model of Keller as base to define brand image and the (measurement of) constructs of brand image.

This study is done by the use of an experimental design, where two questionnaires were used to collect data from 395 respondents. Results showed that there is impact of adding key-value overlap elements to a communication expression on the general brand image and on the (transfer of) associations. Though this effect was not found to be significant. Furthermore this does not lower suspicion people have towards certain partnerships. This study did found a trend where more key-value overlap elements improve the brand image, though furthermore found that this case study used brands that are to controversial to find significant effects. This study contributes to the literature regarding partnerships, joint branding alliances

between two stand-alone brands, brand image and (the transfer of) not only brand associations but also category associations.

(4)

1. Introduction

Branding has been a hot topic of research for a long period of time. The past years the

environment of, specifically the marketing communications, has changed a lot. The latter has

been a reason for Kellerto come up with the customer-based brand equity model. A model

that “emphasizes the importance of understanding consumer brand knowledge structures” (K. L. Keller, 2009). Which can be used as a means to “track how marketing communications involves mixing and matching different communication options to establish the desired awareness and image in the minds of consumers”.

In the current time we live in, that (marketing) communication becomes more and more challenging. Consumers get more critical; they have (social) concerns and expect companies to tackle these concerns (Yoon et al, 2006). Therefore not all companies have a good

(brand)image. A lot of companies furthermore operate in a (core) line of business that has a bad image. From a company's point of view they would like to create a “favourable corporate (brand) image and develop a positive relationship with consumers and other stakeholders” (Yoon et al, 2006).

This brand image is defined by Keller as: “perceptions about a brand as reflected by the brand associations held in consumer memory” (K.L. Keller, 1993). Whereas Keller explains the mentioned associations as: “The other informational nodes linked to the brand node in

memory and contain the meaning of the brand for consumers”. So brand image can be seen as node in the network memory (model) of the consumers, where the associations form the links to that node and in that way a (memory) network is created.

What can be concluded from the theory of Keller is that companies with a bad image deal with the fact that consumers, in that case, have more negative than positive associations with the brand. A lot of companies nowadays try to tackle this, by trying to redefine / boost their image in a positive way. They try to do this for example by communicating to people what the company or brand stands for: what the core (key) values of the companies are and how this links to their (core line of) business.

A good example of a company that tries to do this is McDonald’s, which has been trying to improve its image since 2007. They have been active in a series of environmental and social initiatives, which have been designed to prove they care. For example health has become an important part of their new menu, some McDonald’s restaurants were opened that do not even sell hamburgers, but focus on the concept of a McDonald’s salad bar. The current literature

(7)

states though (for example Ӧberseder et al, 2012) that consumers are very sceptical about companies that pursue such activities / changes that are in line with their core business. Especially when the core line of business has a bad image. Meaning that companies like McDonald’s have a big challenge in trying to redefine the brand image in the minds of consumers.

A lot of research has been done (for example: Yoon et al, 2006 , Einwiller et al, 2006, Becker-Olsen et al 2006) on companies with a bad image trying to improve this image. This research has mainly focused on the topic of Corporate Social Responsibility (CSR). That research states that consumers are very sceptical about companies that pursue CSR activities that are in line with their core business (for example Yoon et al, 2006). Especially when the core line of business has a bad image. Furthermore related literature (for example: Öberseder et al, 2012) showed that a company should definitely not advertise for such activities. So how should companies like McDonald’s redefine their new (intended) image in the mind on consumers? Research gap

One might say that CSR is a more and more out-dated term, though some interesting facts have been concluded by the CSR research that could be useful for future research. One of those conclusions is that communicating about partnerships or CSR activities might raise suspicion and have a negative effect. One could say, in line with the model of Keller, that this has a negative effect on the brand image. Though on the other hand Pappu et al (2014) did found that: “Any sponsorship (type of Joint Branding) that has the potential to raise suspicion would likely benefit from a more detailed examination of fit and similarity”.

Further research has focused more on this detailed examination of fit and similarity. Based on the Elaboration Likelihood Model (ELM), “the probability of message or issue-relevant though occurring” (Petty et al, 1983), past research has shown that:

Explaining how the key-values of two companies (brands) overlap with each other and / or overlap with the values a consumer holds (further referred to as ‘key-value overlap’), can cause people to more centrally process the message the advertiser is trying to bring forward. With above in mind and looking at the fact that nowadays companies try to redefine their (brand) image by partnering up with other brands:

Could this mean, taking Keller's model as a base, that explanation of the link between the two partnering brands might positively affect the associations people have towards the brand and

(8)

the company? Could the positive associations be transferred from one brand to the other? Because there has actually been a lot of research that claims that associations can be

transferred (for example: E. Lebar et al, 2005; J. Boisvert et al, 2011 ; R. Yuan et al, 2016). Though most of this research has focused on the process of brand extensions (for example Bridges et al, 2000) and not two stand-alone brands partnering up.

Past research furthermore gives pointers to the fact that the associations people have towards the company can be (positively) influenced by certain factors. So could and should companies trying to boost or redefine their bad image communicate these factors?

Since CSR is somewhat out-dated and might raise suspicion, this research will take a look if the communication of above can be done in joint branding alliances expressions.

Joint branding can take all types and shapes (E. Lebar et al, 2005). It would be interesting to see, with above in mind, how a joint branding alliance in the form of an advertising campaign (communication) might influence the image of a company / brand that has a negative image. A famous and practical example of this is Coca-Cola sponsoring the Olympic games. By explaining the link between the core-values of the brand and the core values of the Olympic games; they generated a more positive attitude towards the company / brand, the brand and the link between the brands (Pappu et al, 2014). With this in mind, could the key-value overlap be a key element in communications of a joint branding alliance that enables positive associations to be transferred from one brand to another? According to the model of Keller, this would also positively affect the brand image.

It would be interesting to see whether a joint branding alliance between two stand-alone brands would be beneficial for both brands and if the positive associations from one brand can be transferred to another brand. Could this be a solution for companies with a bad (brand) image to redefine their image? How can companies overcome this challenge in a world where the marketing communications is still drastically changing? Could this be done by partnering up with another brand?

This study aims to prove, using conclusions of existing researches as a base, that: two brands can partner-up in joint branding alliance to positively boost their image. This is done by answering the following research question:

Can a company pursue a joint branding alliance with a company that has a strong brand equity or which is more positively known and use key-value overlap elements in their communication to give their own brand image a positive boost?

(9)

2. Literature review & Hypotheses

To answer the research question an experimental design is chosen as the research design. Meaning that an experiment will take place. More information about this research design can be found in chapter 3 and 4.

First all the several elements and constructs of the research questions and experimental design will be further explained below bases on the current literature. After this is done, it is

explained how the parts relate to each other in the experimental design. Followed by hypotheses that are drawn from this.

2.1 Brand equity

This study is mainly based on brand equity and the brand equity model of Kevin Lane Keller. Therefore several articles of mainly Keller will be used to explain the concept of brand equity, customer based brand equity, brand knowledge as an important part of that and all related terms.

Like summarized by Keller, brand equity (in a general sense) is “defined in terms of the marketing effects uniquely attributable to the brand: (K. L. Keller, 1993). The meaning of this is explained more, when looking at the two general motivations of studying brand equity: One being financial motivations, to estimate the actual account value of a brand. And the second being a strategy-based motivation, to improve marketing productivity.

This study will focus on the latter. Marketers continually need a more thorough understanding of consumer behaviour for strategic marketing purposes. Knowledge about the image that has been created in the mind of the consumers is one of the most valuable assets in this case. Nowadays “researchers widely acknowledge brand equity as a key marketing performance indicator, a source of competitive advantage, and a vital component of business success” (K. Chatzipanagiotou et al, 2016).

A lot of research has been done on this topic and “although brand equity may derive from various stakeholders, consumer-based brand equity dominates marketing research” (K. Chatzipanagiotou et al, 2016). Keller, long time ago, conceptualized brand equity from the perspective of the individual consumer. Known as customer-based brand equity. He provides a conceptual framework of what “consumers know about brands and what such knowledge implies for marketing strategies” (K.L. Keller, 1993).

(10)

2.1.2 Customer-based Brand equity

Keller conceptualizes brand equity from the perspective of the individual consumer, known as Customer-based brand equity (CBBE). Which Keller defines as: “The differential effect of brand knowledge on consumer response to the marketing of the brand” (K.L. Keller, 1993). In modern literature, other researchers define CBBE for example as: “a set of perceptions, attitudes, knowledge, and behaviours on the part of consumers that results in increased utility and allows a brand to earn greater volume or greater margins than it could without the brand name” (Christodoulides & de Chernatony, 2010). Meaning that due to a mix of marketing elements consumers have certain associations specifically towards the brand (and not the same products or services from another brand), which they can recall from their memory when they are confronted with the brand or a similar product / service.

The above described occurs when “the consumer is familiar with the brand and holds some favourable, strong and unique brand associations in memory” (K.L. Keller, 1993). Keller explains this all in his model, which is based on even earlier research about types of

associative model formulation. An example where Keller refers to is the associative network model. This model “views semantic memory or knowledge as consisting of a set of nodes and links. Nodes are stored information connected by links that vary in strength” (K.L. Keller, 2009). It can be seen as a network, with nodes that can potentially be linked. This linking process may happen, as explained by Keller, when external information is there (which activates the node to become linked to other nodes) or when internal information is retrieved from long-term memory. This latter process makes the network stronger and stronger, since the nodes are getting connected.

An important conclusion Keller derives from this, is that “marketers must realize that the long-term success of all future marketing programs for a brand is greatly affected by the knowledge about the brand in memory that has been established by the firm’s short-term marketing efforts” (K.L. Keller, 2009). The meaning of this brand knowledge (nodes in the network), a key part in this study, will be further explained in the next part.

2.2 Brand Knowledge

So brand knowledge is, based on previous mentioned, conceptualized by Keller as “consisting of a brand node in memory to which a variety of associations are linked” (K. L. Keller, 2009). It is important to understand content and structure of this brand knowledge according to Keller, since it “influences what comes to mind when a consumer thinks about a brand” (in response to a certain marketing activity for that specific brand). Keller explains this brand

(11)

knowledge as a node in the network, to which certain associations are linked. This brand knowledge and the affect of consumer response are determined by the awareness of the brand and the favourability, strength and uniqueness of the brand associations in the memory of the consumer.

So according to Keller, two dimensions distinguish brand knowledge: one being brand awareness, which is related to the strength of the brand node or trace in memory (recall and recognition). And the other one being brand image. Which Keller defines as: “Perceptions about a brand as reflected by the brand associations held in consumers memory” (K. L. Keller, 2009). These dimensions and mentioned terms will be further explained next. Where brand image is the most important one for this study.

2.2.1 Brand awareness

As mentioned above, “brand awareness is related to the strength of the brand node of trace in memory” (K.L. Keller, 1993). How well established are the brand identities? Can consumers identify the brand in different conditions? Meaning the likelihood and the ease that brand comes to mind for consumers in certain situations.

Brand awareness consists of brand recognition and brand recall performance. Brand

recognition meaning that consumers are actually able to confirm that they have seen the brand before and know the brand. Whereas brand recall performance is related to how well a

consumer can identify or retrieve the brand in their memory when given a certain cue (for example the product category).

According to Keller, there are three major reasons why brand awareness plays an important role in the decision making process of consumers:

• It is important that consumers think of the brand when they think about the product category.

• Brand awareness can affect decisions about brands in the consideration set.

Which is based on the Elaboration Likelihood Model (explained more in further in this chapter): when the involvement is low people make choices based on simple on

simple decision rules. Brand awareness may in that case affect the decision.

• Brand awareness affects consumers’ decision-making by influencing the formation and strength of brand associations in the brand image.

A brand node is a crucial part of the brand image, a node that has been established in the memory. The nature and strength of that brand node, according to Keller, “affects

(12)

how easily different kinds of information can be become attached to the brand in memory” (K.L. Keller, 1993).

2.2.2 Brand image

Keller defines brand image as: “Perceptions about a brand as reflected by the brand

associations held in consumer memory”. Furthermore for example James explains that brand image “consists of the attributes and associations that consumers connect to a brand” (D. James, 2005). Where, according to Keller, the brand associations are “the other informational nodes linked to the brand node in memory and contain the meaning of the brand for

consumers” (K.L. Keller, 1993). And for example Boisvert states that brand associations “represent knowledge linked to a brand in a consumer’s memory” (Lynch and Srull, 1982 as quoted by Boisvert et al, 2011).

2.2.2.1 Brand Associations

These above mentioned associations differ in favourability, strength and uniqueness. These dimensions distinguish the brand image and this plays a role in the brand equity process: how the consumers respond to the brand under certain conditions.

• Favourability

Does the brand have attributes and benefits that satisfy the needs and wants of the consumer?

• Strength

The strength of the connection to the brand node. Depends on the amount of information that enters the memory of the consumer and how that information is related to the brand image.

• Uniqueness

Whether brand associations are shared or not with other brands. Is the association uniquely attributed to the certain brand?

Keller explains that the favourability, strength and uniqueness are affected by the level of abstraction and qualitative nature of the brand associations. Abstract associations have a more evaluative nature, and therefore are more durable and assessable in the mind of the

(13)

Keller furthermore explains that since a brand is linked to a product category, some of the product category associations might also become linked to the brand. According to Keller, product category associations are important in the determination process of the consumer. On how the consumer responds to a product or service related to a certain category. For example if a consumer thinks that fast food restaurant are always bad and unhealthy, he or she might have similar unfavourable thoughts and associations to any particular restaurants in that category. Just because the restaurant operates in that certain category.

Keller furthermore shows, together with Burke and Srull (1988) that the number of competing brands advertising in a certain product category can cause disturbance in the mind of the consumer. Negatively affecting the ability to recall certain communication effects of a specific brand in that category. Though they have also shown that this negative effect can be overcome by using ad retrieval cues, showing certain ad information when a consumer makes a brand evaluation. Which gives reason for this study to take place.

Furthermore other related research also talks about these associations, for example Einwiller talks about corporate associations. Which he defines as: “Corporate associations encompasses all the information and beliefs that a person holds in reference to a particular company” (Brown and Dacin, 1997 as quoted in Einwiller et all, 2006).

So the associations a consumers has in mind determines a major part of the brand image. Literature states that an important part for companies to boost the brand image is corporate social responsibility, which will be explained next.

2.3 Corporate Social Responsibility

Research has been done before to investigate how companies with a bad image could boost their image, this research focused a lot on Corporate Social Responsibility (CSR). CSR activities can be seen as activities a company does because they have a responsibility to society, from the point of view of companies this might be used to boost the image of company / brand.

The current literate though furthermore states that consumers are very sceptical about

companies pursuing CSR activities that are in line with the companies core (line of) business, especially when that line of business has a bad image (Y. Yoon et al, 2006).

Furthermore for example Mazutis et al (2015) refer to the term green washing. They refer to the fact that CSR is increasingly seen as a tool to cover up firms societal harms. The current literature furthermore showed that a company should definitely not advertise for CSR in that case (for example Öberseder et al, 2012). Though this research has not focused on how

(14)

companies with a bad image, or operating in a line of business with a bad image, should boost this image. So what can be learned from the conclusions drawn from CSR research, and how can these be helpful for future research? One could debate that CSR is a more out-dated term and research shown how this might raise suspicion. Therefore this study will look at the interesting conclusions that are drawn from CSR and that could be helpful for future research and looks at ways companies an brands (with a bad image) could boost this image.

As Öberseder et al, 2012 concluded, communicating about CSR raises suspicion and has a negative effect. There are several forms of communication strategies, for example pro-active and re-active communication strategies (Wagner et al, 2009). According to Keller, marketers should ‘mix and match’ these communications options and strategies, to build brand equity. Or Like stated by Bhattacharya et al (2004): “In devising effective communication strategies, marketers need to make both source and message decisions that minimize unfavourable attributions”. Bhattacharya et al (2004) furthermore look specifically at companies that operate in a core line of business with a bad image. They state that belonging to a certain industry (e.g. oil, tobacco, and alcohol) is likely to dampen the effect of CSR initiatives due largely to the unfavourable often cynical attributions that consumers are likely to make in these instances”. Though for example Yoon et al (2006) state, when talking about

communication, that “if there is other contextual information available, consumers may process it systematically to determine the company's true motives”. And they furthermore state that “information that is at odds with perceivers' expectations about the actor, e.g. because it contradicts what is known about the actor's previous behaviour, further increases the likelihood of extensive attributional processing”.

Pappu et al (2014) found that: “Any sponsorship (type of joint branding) that has the potential to raise suspicion would likely benefit from a more detailed examination of fit and similarity”. Which is also an interesting finding. Further research has focused more on this detailed

examination of fit and similarity. Based on the Elaboration Likelihood Model (ELM), explained in more detail next, past research has shown that:

Explaining how the key-values of two companies (brands) overlap with each other and / or overlap with the values a consumer holds can cause people to more centrally process the message the advertiser is trying to bring forward. Though no research has been done how the transfer of associations, specifically with unfavourable associations, is influenced.

(15)

2.4 Elaboration Likelihood Model (ELM)

Like mentioned by Keller, the elaboration likelihood model (Petty and Cacioppo, 1986) suggest that consumers may base their choice on brand awareness considerations when they have low involvement. The effect that is aimed for, is that people actually process the message the brand(s) try to bring forward and base their decision on that.

Petty and Cacioppo et al (1981) have shown that there are “two distinct routes to attitude change”. Which route one takes depends on the elaboration likelihood of the communication situation. The elaboration likelihood means “the probability of message or issue-relevant though occurring” (Petty et al, 1983). When the likelihood is high, they state that the central route is the most effective one.

So according to Petty et al, there are two routes to attitude change. The first one is the peripheral route. Attitude changes that occur via this route do not occur because the person has systematically processed the message or considered the pros and cons of the message. In this case the attitude changes is caused by positive or negative cues in the object (e.g.

advertising image) or because the person makes a simple decision based on elements of the persuasion context, e.g. the advertising image (Petty et al, 1983). When the likelihood of issue-relevant thoughts occurring is low, Petty et al (1983) state that the peripheral route is more effective.

The central route “views attitude change as resulting from a person’s diligent consideration of information that he/she feels is central to the true merits of a particular attitudinal position”. (Petty et al, 1983). One could also state that people taking this route are actually processing the message the advertiser is trying to bring forward. The person will personally consider the pros and cons of the message. The message gives people information to think about. Petty et al (1983) furthermore found that “if under scrutiny the message arguments are found to be cogent and compelling, favourable thoughts will be elicited that will lead to attitude change in the direction of the advertiser”.

Regarding communication, what factors of the communication (e.g. advertisement) would influence people taking a certain route? For example Yoon et al (2006) state: "If there is other contextual information available, consumers may process it systematically to determine the company's true motives”.

The latter is the effect a company that wants to boost the brand image is looking for. But how could this be done? Previous research gives pointers that this can be achieved by making use of key-value overlap elements in communications.

(16)

Combining the above fact with the conclusions drawn from CSR literature, as could be concluded partnerships and specially communicating about these is often seen as suspicious. But the CSR literature furthermore concluded that a detailed examination of fit and similarity in that case might help. The Elaboration Likelihood Model provides insights into the how. A famous example of a brand that successfully did the above is Coca-Cola. Coca-Cola

sponsoring the Olympic games is an example where the fit and specially explanation of the fit caused people to centrally process (ELM) the message. Or in that sense at least think about why the brand partnered up (sponsored) the Olympic games. By explaining the link between the core-values of the brand and the core values of the Olympic games; they generated a more positive attitude towards the brand and the link (Coca-Cola – Olympic games). Current literature provides more information about this key-value overlap.

2.5 Key – value overlap

Sen et al (2001) and Einwiller et al (2006) concluded that the effect of initiatives such as CSR activities on consumers' company evaluations is mediated by “self-company congruence” and “consumer-company identification”. Translated by Einwiller et al (2006) as:

“The extent to which aspects of an organization's identity overlap with consumers' identities”. Einwiller furthermore talks about the term organization identification. Which he translates as “the degree to which consumers feel a sense of connection to a company and the degree to which aspects of the perceived organizational identity are self-referential and self-defining for them”. Furthermore Einwiller et al (2006) state: “If a consumer identifies with a company, he or she is likely to have positive thoughts and feelings (associations!) about it. This implies that the identified consumer’s corporate associations are favourable. They continue by stating: “When strongly identified consumers’ positive beliefs about a company are challenged by negative information, they are likely to protect and preserve those beliefs”.

This goes further, as can be seen in the example of Coca-Cola sponsoring the Olympic games. By not only making the link between the core values of the brand and the personal values a consumer’s beholds, but also explaining the link between the two partner brands and explain how the key-values overlap. Which gives reason for this study.

Furthermore previous literature (for example Einwiller et al, 2006) has also looked at the fit between, in the case of CSR, the link between the company business line and the cause’s mission and what this does with the attributions.

(17)

Though one might conclude that CSR is becoming somewhat of an out-dated term, a lot of research about this topic is dated. Furthermore previous research shows that consumers are very critical and suspicious about brands with a bad image partnering-up or engaging in activities to boost their image.

Therefore this study will take a look at Joint Branding, which is hot topic of research further explained in the next chapter. In the lights of CSR, partnerships and key-value overlap joint brandings is an interesting way to look at the challenges brands with a bad image face nowadays.

Could above example from Coca-Cola also be applied nowadays, not by the means of CSR but by a joint branding alliance between two companies? Where brands partner up in a joint branding alliance to boost the image of the brand, instead of performing CSR. Could the brands explain the fit? The link between the key values of one brand and the key values of another brand? Show people that they are sincere, for example in the practical example that was mentioned in the introduction: McDonald’s.

Could the brand in such a way show people that they are sincere and in that way positively boost their image? What if McDonald’s, partners up with another brand, and show people how the core values of their new green and healthy image overlap with another brands related to that new image. For example sports and a healthy lifestyle.

First of all, some importing constructs named above will be explained more in detail.

2.4 Joint Branding (alliance)

“Joint branding occurs when two brands enter an agreement to market some product or service in tandem” (E. Lebar et al, 2005). The reason that brands partner up is, that nowadays brands need to keep differentiating to beat the competition. So actually in order to be unique and stay ahead of the competition.

This joint branding can take many shapes, with one of them being in the form of an

advertising campaign, which is the one this study will take a closer look at. Brands partner up together for a variety of reasons, but like stated by E. Lebar: “Often an alliance is an

opportunity for one brand to build equity by leveraging characteristics or attributes of the other, partner brand” (E. Lebar et al, 2005). Furthermore Lebar et al explain that alliances work best when both of the involved parties gain equity from the partnership. Meaning that both parties benefit from the marketing effects of for example a joint branding alliance in the form of an advertisement campaign.

(18)

It would be interesting to see, in the light of the model of Keller when he talk about the constructs of brand image, if the positive associations a consumer has of a certain brand can be transferred to the brand they have partnered up with. With both of these brands benefitting from this joint branding alliance, both gaining equity. But how should the brands

communicate, in the shape of advertisements, to gain this effect?

Coming back to the theory of Keller, there are three ways associations are created: • On the base of direct experience.

• Information about the product or service communicated by the company / brand and other commercial resources, world of mouth.

• On the base of inferences from some existing brand associations.

Keller continues with another important base of this study, when he talks about secondary links that can be created in the minds of consumers. He states that another type of inferred association occurs when “the brand (association) itself is linked to other information in memory that is not directly related to the product or service” (K. L. Keller, 1993).

He furthermore states that if the brand becomes identified with other entities, consumers may see that this brand shares associations, producing secondary links. These secondary links, according to Keller, may lead to transfer of (global) associations such as attitude or creditability.

So could joint branding, in form of advertisements, make above happen? Could partnering-up with another brand that is more positively known, or where consumers behold more

associations the own brand is aiming for help? Showing people how the brands overlap, cause people to transfer the associations? Could the above named example McDonald’s, partner-up with another brand that is more related to their new image (they strive for) and cause people to find this more creditable? And how could the partnering brands show people the overlap? What key-value overlap elements in the communication should be used in what way to make this happen?

2.5 (marketing) Communication(s)

According to Keller, marketing communications are “the means by which firms attempt to inform, persuade and remind consumers… about the products and brands they sell” (K.L. Keller, 2009). Keller translates this furthermore as the “voice” of the company and the brand(s) and explains that marketing communications can be used as means to “establish a dialogue and build relationships with and among consumers” (K.L. Keller, 2009). This

(19)

marketing communication can be used to show a consumer who makes the products and more specifically what the company and brand stands for. According to Keller, “marketing

communications allow companies to link their brands to other people, places, events, brands, experiences, feelings and things”. Which is actually what this study is focused at, using marketing communication to link brands together. Show people how they link and show people what the brand stands for. And therefore see if people transfer associations related to experiences, feelings etc. from one brand to the other.

This all can contribute to the brand equity, by establishing the brand in the memory of the consumer and create, or in the case of this study, redefine the brand image (Luo and Donthu, 2006 as quoted by K.L. Keller, 2009).

Keller talks about eight major communication types, whereas the most interesting ones related to this study will be summed-up below:

• Advertising

Any paid form on non-personal presentation and promotion of ideas, goods or services • Public relations and publicity

A variety of programs, designed to promote or protect a company’s image or its individual products

• Interactive marketing

On-line activities and programs designed to engage customers or prospects and directly or indirectly raise awareness, improve image or elicit sales of products and services

There are five more major forms of communication types, but above are most interesting for this study. Since this study aims at a joint branding alliance between two stand-alone brands with the main goal to improve the (brand) image.

According to Keller, marketing communications have an effect on the brand equity. “By creating awareness of the brand; linking the right associations to the brand image in consumers’ memory; eliciting positive brand judgments or feelings; and/or facilitating a stronger consumer-brand connection”. (K.L. Keller, 2009). Regarding this study, the

questions remains whether this effect can be transferred from one brand to the other, could the link between the two brands be made?

Marketers must consider a lot of factors when they are developing their communication mix. According to Keller, anything that causes consumers to pay attention to the brand, for example sponsorships or in this case a joint branding alliance can increase brand awareness.

(20)

But Keller furthermore states to enhance the recall of the brand; more elaboration and processing may be needed. Because in that case, stronger brand links to the product category or consumer needs are established. Which improves the memory performance.

The latter is where the previous research, about CSR and the ELM theory, has given us interesting conclusions.

Summary & framework

So in general previous research, about for example CSR and communication, found that a lot of companies used CSR as a way to boost the brand image. Partnering up with certain other brands or causes, though this might raise suspicion as previous research also showed. Companies engaging in CSR activities that are in line with their core business, might cause consumers to have a negative attitude towards the company (Yoon et al, 2006). But Yoon furthermore states that people are willing to make inferences if the companies’ motives are not solely ulterior. But to conclude the latter, people must process the information

systematically to see the key-value overlap (Mazutis et al, 2014, Öberseder et al, 2012). This is where the ELM theory provides interesting conclusions.

Continuing on the brand equity mentioned before, Keller (2009) states that building a strong brand is a benefit for the effectiveness of communications. “As a result of the strength and equity of the advertised brand, consumers may be more willing to attend to additional communications for a brand, process these communications favourably and have a greater ability to later recall the communications…” (K.L. Keller, 2009). So if a certain brand has strong brand equity, could those consumers be more willing to attend to additional

communications for that brand if this is related to a joint branding alliance?

Could the brands use certain elements in their communication to show people the overlap between the brands (key-values) and cause people to see why the two brands partner up? Could this, with Keller’s theory as a base, cause people to transfer the positive associations a consumer’ has of the one brand and product category, to be transferred to the brand / category that is more negatively known? In that way, boosting their image. It would be interesting to see if the brand image consumers have, could be redefined by this.

(21)

This study will try to answer, taking the brand equity theory of Keller as a base, how

companies with a bad (brand) image can boost / redefine this image. The following research questions derives from this:

Can a company pursue a joint branding alliance with a company that has a strong brand equity or which is more positively known and use key-value overlap elements in their communication to give their own brand image a positive boost?

The part of the framework from Keller this study is based on, can be found on the next page.

Figure 1 – Brand knowledge explained by Keller (K.L. Keller, 1993)

Looking at the theory of Keller, companies who want to boost their (more negative) image have a bad image due to the fact that people have more negatively associations than positive associations (on certain elements). Though, like can be concluded from the theoretical part, a lot of research has already discovered that associations can be transferred.

This study will try to look at how a brand with a more negative image, which it tries to boost, could enter a joint branding alliance with another brand that is more positively known and that has a strong brand equity. Using the conclusions previous research has provided, as a base and input on the how.

(22)

The following hypothesis have been formulated based on the theory provided above and will be used to answer the research question:

H1: When people are faced with a joint branding expression that elaborates on how the core-values of the brands overlap, people are more triggered to think about the connection

between the brands, which will have a positive effect on the general brand image of both brands.

This hypothesis is based mainly on the theory of David Aaker (1996b), who claims that the measurement of associations and differentiation can be structured around three perspectives on the brand: the brand as a product (value), the brand as person (brand personality) and the brand as organization (organizational associations). Furthermore a lot of research refers to this technique of measuring, whereas on the one side questions of Aaker were used to measure the general brand image (GBI) and secondly (in those researches) product brand image (PBI) was measured. Since this is a case study for McDonalds and both brands operate in completely different categories, a decision was made to use the general brand image questions derived from Aaker (the ones that were applicable for this study), since the scale work across different product categories, and secondly measure the associations.

When the overlap between the brands is explained this study first of all expect a positive influence on the general brand image, since it is expected to let people think (based on ELM) about the connection and take away the suspicious factors.

H2a: Key-value overlap elements in a joint branding expression have a positive effect on the associations of a brand. The more and better the key-values and the overlap between them is explained in an advertisement, the higher the chance that people associate more with

favourable brand associations and less with unfavourable brand associations.

This hypothesis and hypothesis 3A are based on theory and research of mainly Keller using the ELM theory as a base. A positive effect is expected on the brand associations people already had towards the brand (H2a), improving favourable brand associations and having less negative brand associations. Secondly a positive effect is expected on the transfer of brand associations people did not had with the brand before (coming from the other brand in partners up with) (H3a)

H2b: Key-value overlap elements in a joint branding expression have a positive effect on the associations of the related category. The more and better the key-values and the overlap

(23)

between them is explained in an advertisement, the higher the chance that people associate more with favourable category associations and less with unfavourable category

associations.

This hypothesis and hypothesis H3b are also based mainly on the theory of Keller. A lot of research has focused on (the transfer of) brand associations, but not a lot of those researches also focused on the category associations. While Keller states, when he talks about brand awareness that it is important that people think about the related category. Keller furthermore explains that since a brand is linked to a product category, some of the product category associations might also become linked to the brand. Which improves the network and thus the brand image. According to Keller, product category associations are important in the

determination process of the consumer. Furthermore one could state that brand associations are less embedded than category associations, since a lot of brands operate in the same category. It would be interesting to see if an effect could also be found in the (transfer of) the more embedded category associations.

A positive effect is expected on the category associations people already had towards the related category (H2a), improving favourable category associations and having less negative category associations. Secondly a positive effect is expected on the transfer of category related associations people did not had with the category before (coming from the other category related to the brand it partners up with) (H3b).

H3a: Key-value overlap elements in a joint branding expression have a positive effect on the transfer of brand associations from one brand to the other. The more and better the key-values and the overlap between them is explained, the higher the chance that favourable brand associations are transferred from one brand to the other.

H3b: Key-value overlap elements in a joint branding expression have a positive effect on the transfer of category associations from one category to the other. The more and better the key-values and the overlap between them is explained, the higher the chance that favourable category associations are transferred from one category to the other.

To answer the research question:

H4: When key-value overlap elements in a joint branding expression are used and the partnership is explained, this will have a positive effect on the total brand image of both partnering brands.

(24)

3. Data and Methods

This section describes the performed study. What, how and why this is done. A complete description is given on the way the data is obtained. How the different constructs of the research were operationalized and the types of analyses that were used will be described.

3.1 Research design

This study is a quantitative research. A deductive approach is taken. Which is an approach where the literature is used to help identify theories and ideas that will be tested using data. ‘Developing a theory and hypothesis (or hypotheses) and design a research strategy to test the hypothesis’ (Saunders, Lewis & Thornhill, 2009).

An experimental design is chosen as a strategy to perform this study and answer the research question. The design, groups and measurements are mainly based on the brand knowledge part of the theory of Keller, which can be found in chapter 1.

With a quantitative research method, the relationships between variables are examined. The relationships are measured numerically and a range of statistical techniques are used to analyse the relationships. The data in this study was collected with questionnaires. Since a questionnaire ‘allows the collection of a large amount of data from a sizeable population in a highly economical way’ (Saunders, Lewis & Thornhill, 2009). The design contained one control group and three experimental groups. Which are based on three different forms of advertisement (communication). More about information about these advertisement types can be found in chapter 3.3.3.

3.2 Procedure and sample

Case study

To make things practical, this study focused on a well-known company who is currently already trying to boost their image by tackling the negative elements of their brand image. Which was, like mentioned in the introduction, the company McDonald’s. Meaning that this research is a case study.

A joint branding alliance (partnership) with Basic-Fit was chosen for the experimental part of this study. Since this study focuses on the Dutch market, a well-known fitness brand in the Dutch marketing is chosen. Operational in a category (fitness) that resolves around the healthy lifestyle McDonald’s is trying to bring forward.

(25)

This study will try to see if this joint branding alliance can help McDonald’s boost (certain elements of) their image.

McDonald’s being a brand that (potentially, assumption that will be verified) has a more bad image and operates in a core line of business (product category) with (potentially) a more negative image (assumption that will be verified). And a brand that (potentially) has a more positive image (assumption that will be verified) and operates in a core line of business (product category) that is focused on a healthy life style and being active and sporty. Which therefore also might have a more positive image (assumption that will be verified as well). This study tries to see if these both brands can partner up in a joint branding alliance. To see if McDonald’s could for example state and show people that they want to motivate people to go out and sport, at more healthy and have a healthy lifestyle. To show people that McDonald’s follows a strategy that is not solely focused on providing people with fast food, but they also have a healthy menu. Focus on showing people that it is fine to enjoy a hamburger every now and then, as long as you sport next to it or have a healthy lifestyle. Showing people, in the end, that they are sincere in the image they bring forward.

This study tries to see if McDonald’s could show this core values and mission within their communication and trigger people to think about this, when partnering up with a brand that is operational in a category that evolves around the image McDonald’s is trying to bring forward as part of their image. Adding key-value overlap (explaining) elements in the advertisements. This study tried to show that this might cause people (based on the ELM theory) to be more triggered to think about the partnership.

The partnership was, next to the advertisement, explained in the questionnaire (in end, as an introduction to the last questions). The idea of this partnership was that you could save for a discount at Basic-Fit when choosing something healthy and responsible from the menu of the McDonald’s. Commercial (advertisements) of only healthy products of the McDonald’s would be shown in the Basic-Fit. This fictional partnership is based on the “enjoy a

hamburger every now and then, as long as you have a healthy lifestyle”. Which is something that can be spotted in almost every canteen of a sport club (in the Netherlands at least). Selling French fries and other kind of snacks, since it is “fine to enjoy a snack after some sports”.

The benefit for Basic-Fit was to show and motivate people in their gym to eat healthy. Furthermore Basic-Fit would potentially benefit by getting more customer base, due to the

(26)

discount savings at McDonald’s when choosing healthy menu items. All of this is based on the practical example given before, Coca-Cola sponsoring the Olympic games where both brands gained brand equity.

Procedure

The questionnaires (one for the control group and one for the experimental groups) were distributed among my own network, furthermore making use of social networks (Facebook, Linked-In) and by making use of fellow students. Furthermore the professional network of a marketing professional was used to distribute the questionnaire. The questionnaire was made in Dutch, since this case study was about two Dutch brands and aimed at the Dutch market. Most of the questions used were derived from articles published in English journals, more detail about the questions asked can be found in the next chapter. Questions used were translated into Dutch and the translation was verified with two advanced English-speaking people active in research. To ensure a good response rate, the number of questions was limited and a control has been done (by tests) that filling in the questionnaire did not take (much) longer than 10 minutes. Each question was clearly explained and terms were explained as well.

Control questions were asked in the beginning of the questionnaire to test how familiar respondents were with the brand McDonald’s (M = 1.89, SD = .790) and the Basic-Fit (M = 2.55, SD = 1.039), to ensure that the data was useful. For these questions a 5-point likert-scale was used ranging from ‘Extreem bekend’ (1, extremely familiar) until ‘Extreem onbekend’ (5, extremely unfamiliar). Where a lower score indicates more agreement.

The research aimed for N = 100 respondents for each group of the questionnaires (one control group, three experimental groups) to eventually do useful analyse. After correction of the outliers a total of 370 respondents completed the questionnaire.

One of the questionnaires contained some design flaws and these were corrected for. In the control group questionnaire respondents were able to check multiple answers for the scale questions. So two corrections needed to be done:

• First all the respondents who checked more than one answer on a single question were deleted in total in the questionnaire (all results of the specific respondent)

o After this was done, the questionnaire was distributed again to some people, to gain extra respondents and compensate for the loss of data.

(27)

• Second of all, a syntax is SPSS was used to transform the multiple answers to a scale answered question (like the other likert scales question where one answer could be given).

There was a second design flaw in the questionnaire. Since this is an experimental design, respondents were randomly assigned in one out of three experimental groups. Each group saw the same control questions (age, gender, etc.). Than each respondent was randomly assigned to a group. Each group saw a different advertisement followed by questions that needed to be answered with the seen advertisement in the back of the respondents mind. More details about these constructs can be found in the next chapter.

Though since for every group a different ‘block’ was created in the questionnaire, all the questions that followed after seeing the ad were stored separately. Meaning that every

question was found times three in the final data. The same questions for respondents who saw advertisement 1, the same for advertisement 2 etc. Due to this design, the data was hard to interpreter. And analyses like ANOVA were impossible to do due to this set-up. This was corrected by combining all data and using one variable to indicate which advertisement respondents saw.

3.3 Measurement of variables

Before the study took place, some pre-tests were done. The first pre-test was done by interviewing some (random) people about the study that will take place. This pre-test was mainly used to test whether the overall idea and concept of this study made sense to people. Whether those people would confirm the effect that is expected to be found. Or to be more specific: do, at least, the different types of advertisements express the things they should express. For more details about this communication expressions see chapter 3.3.3. The next chapters will explain the different constructs and variables used in this study.

3.3.1 General brand image

Based on several researches (for example that of E. Martinez et al, 2007), brand image is measured in two ways. According to these researches this is measured based on first of all the General Brand Image (GBI), where they refer to the article of D. Aaker (1996b), and they furthermore measure Product Brand Image (PBI). Though since this study evaluates a joint branding alliance between two brands that operate in completely different categories and one is even service related. In stead associations will be measured, bases on several studies explained in the next chapter, next to the general brand image.

(28)

David Aaker (1996b) claims that the measurement of associations and differentiation can be structured around three perspectives on the brand: the brand as a product (value), the brand as person (brand personality) and the brand as organization (organizational associations). Six general brand image statements were derived from Aaker, the ones that were applicable for this study. Since the scale work across different product categories.

These questions were answered using a 5-point likert scale ranging from ‘Helemaal mee eens’ (1, strongly agree) until ‘Helemaal mee oneens’ (5, strongly disagree). Where a lower score indicates more agreement with the named statement.

3.3.2 Associations

A second pre-test was conducted where people were asked about the abstract associations they have towards both brand and related category’s. This is based on several previous studies that did the same (George S. Low et al, 2000, Margaret K. Hogg, 1998, E. Martinez et al, 2007). A questionnaire was used for doing this, where each of the respondents was furthermore verbally informed.

People were explained the definition of an (abstract) association and were asked to write down at least three and a maximum of five brand associations and five category associations for both brands. They were furthermore asked whether the by them named associations were strong associations and whether they were favourable (positive, neutral of negative). The latter is based on Keller (K.L. Keller, 1993) since Keller claims that associations can very according to their strength, favourability and uniqueness. The results of the named

associations can be found in appendix A.

The last two questions were knowingly asked to the respondents themselves, so there is no influence of external people coding the associations in the category’s negative, neutral or positive. Uniqueness was not asked about, since it is hard to measure or ask whether people only have these specific associations with the named brand.

Furthermore respondents were told about the study that will take place and they were confronted with concepts of the different types of communication. They were asked what comes to mind when seeing this type of advertisement, in order to verify the (desired) result of the communication types (further explained next). A total of 15 people were asked about the abstract associations that come to mind when thinking about both brands and related category’s, generating 188 different associations.

(29)

The results of the named associations were clustered (grouped). Several people verified this. For each brand and each product category the most named associations were eventually used in the questionnaire where each brand and category had two positive, two negative and one neutral association. Also equal amounts of strong associations were used.

The above associations were than used as input for questions in the questionnaire. Respondents were asked how much they agree with each association after seeing an advertisement (randomly assigned to one of the three advertisements); five for the brand McDonalds, plus five associations named for Basic-Fit asked at the brand McDonalds (to check for the transfer of associations) and vice versa for the brand Basic-Fit. The statement (‘I associate the brand / category X with association x’) were all answered using a 5-point likert scale ranging from ‘Helemaal mee eens’ (1, strongly agree) until ‘Helemaal mee oneens’ (5, strongly disagree). Where a lower score indicated more agreement with the named

associations at the brand or category.

3.3.3 Different type of advertisements

This study focused on a joint branding alliance between two brands that partner-up by engaging in an advertisement campaign together. A little (pre)-test was performed where people where exposed to two different types of advertisements, one simple advertisement and two advertisement that elaborated more on the core-vale overlap. One had more key-value overlap elements than the other. This was done to test which types of advertisement should be used. The later was done for the brand McDonald’s and the brand Basic-Fit combined. One advertisement just had two logos combined in a frame (simple advertisement), the logo of McDonald’s and the logo of Basic-Fit. The other advertisements had more key-value overlap (explaining) elements in the advertisements. All advertisement where made based on input from a marketing professional and other correspondents who helped designing the

advertisements. Some people (N=10) where asked which of the advertisements caused them to have the most issue relevant thoughts, without mentioning more information about the study. All explained that the key-value overlap explaining elements in the ad caused them to have more thoughts about the linkage between the two brands.

The outcome of this pre-test somewhat confirmed that showing people advertisements with more value-overlapping elements caused them to think more about the linkage. Therefore these types of advertisement types will be used during the experimental design. Based on the input of this pre-test, previous research, the input of an marketing professional and elements

(30)

McDonald’s is already using (green colour instead a red one) the final design of the advertisements made:

• A simple advertisement; showing just the two logos in one frame.

• An second advertisement that contained more key-value overlapping elements, using visual elements (for example colour) and funny elements.

• A third advertisement that showed even more key-value overlapping elements and also elaborated on the value-overlap between the brands (textual).

3.4 Control variables

The following control variables were used for this questionnaire:

• Age.

• Family situation. • Level of education.

• Whether the respondent does sport / fitness (goes to a gym) and how often.

• Whether the respondent eats (and how often) fast food or goes to fast food restaurant.

These control variables were used to check whether the groups are comparable (to check if each group represented by the same age categories, family situations etc.). This will be analysed and explained in detail in the result section. Furthermore these control variable might have an influence or explain the effect on the relation between the seen advertisement and the scores on general brand image and associations.

3.5 Additional questions

Additional questions were added to the questionnaire. One questions at the beginning of the questionnaire asking the respondents whether they though the named brands were strong brands. This questions was asked to check if the strength of the brand might have a

moderating effect on the relationship between the seen advertisement and the general brand image / associations.

Furthermore respondents were asked in the beginning of the questionnaire what their opinion regarding the both category’s were, to check whether this might influence the results.

Furthermore additional question were asked and the end of the questionnaire to get more of an idea of how the respondents think about the partnerships. These questions might be useful to check if they explain effects between the seen advertisement and the general brand imago /

(31)

Respondents were, at the end of the questionnaire, asked whether they though the partnership was logical (from the point of view of both brands). The statements were answered using a 5-point likert-scale. Ranging from ‘Helemaal mee eens’ (1, strongly agree) until ‘Helemaal mee oneens’ (5, strongly disagree). Where a lower score indicates more agreement with the

statement and thus finding the partnership more logical.

Furthermore a question was asked what the partnership did with their attitude towards the brand. The latter questions was asked using a 7 point semantic differential (rating) scale ranging from:

• ‘Sterk negatief ten opzichte van hiervoor’ (strongly negative compared to before) • Until: ‘Sterk positief ten opzichte van hiervoor’ (strongly positive compared to before)

With a middle option of ‘Zelfde als hiervoor’ (same as before). Indicating whether the partnership had a negative, neutral or positive effect for both brands looking at the attitude towards the brand after the partnership was explained.

3.6 Verifying assumptions using the control group

The control group was used to verify the named assumptions in this study about the image of McDonalds and Basic-Fit before this study. The assumption that the brand McDonald’s and category fast food had a more negative image, more negative associations than positive ones. And the other way around, verify that the brand Basic-Fit and the category fitness had a more positive image, more positive associations than negative ones. Furthermore the control group in this way of measurement provided a baseline.

If the control group would show that the assumptions could not be verified, the control group purely acts as a baseline of the current image for both brands. Focusing on the differential effect of the different types of communication. Claiming that a certain advertisement / form of communication, due to adding specific value-overlapping elements, has a more positive effect on the brand image (via associations) than the other.

(32)

4. Results

This section describes the collected data of this study. How the data was analysed and in this chapter the named the named hypothesis will be tested. (Extended) conclusion will be provided in the discussion chapter (chapter 5).

4.1 Comparability of groups

First of all a check was done to see if the groups are comparable based on the control

questions. Whether each group was represented by the same age categories, family situations etc. First of all the descriptive statistics were analysed and the means were compared.

Figure 2 – Descriptive statistics control variables across groups

Above descriptive statistics show that the mean scores of the control variables among the groups were pretty much equal. Though what stands out if that the control variables for the age (‘Wat is uw leeftijd?’) and family-situation (‘Wat is uw gezinssituatie?’) have higher mean scores when looking at the control group who saw no advertisement (‘geen

advertentie’). This was further analysed.

Since most of the control variables are nominal or categorical (and thus not normally distributed), next to the standard descriptive statistics, a Kruskal-Wallis was used to test whether there were any significant differences between the control variables between all the groups. This is non-parametric test used compare groups, based on rank scores. This test is used when assumptions of normality or violated. Which was the case, looking at the type of variables that were compared. Defining the hypotheses for the Kruskal-Wallis test:

H0: Median Control group = Median Advertisement 1 = Median Advertisement 2 = Median Advertisement 3

(33)

So the Kruskal-Wallis was performed to check if the samples came from populations with the same distribution, or in other words: there are no significant differences in the control

variables between the groups.

Performing the Kruskal-Wallis test in SPSS gave the following results:

Figure 3 – Kruskal-Wallis test results control variables

All the control variables were set as test variables, and the variable that indicated which advertisement the respondent saw was used as the grouping variable. The test shows that the control variables ‘age’ ((H3) = 23.012, p = .000) and ‘family-situation’ ((H3) = 11.599, p = .009) were significantly different across the groups. The p value is < 0.05 and therefore the null hypothesis was rejected and hypothesis 1 was assumed.

The first thing that stood out is that these two variables might be related, family-situation, in most cases is (logically reasoned) related to the age. The second thing that stood out was that the average mean rank scores of the control group differed from the other groups when looking specifically at the two mentioned control variables.

(34)

When a custom table was made in SPSS, to zoom-in in more on the data, this could also be concluded:

Figure 5 – Custom table 1 - significant different control variables

The control group, which did not see an advertisement, were mostly in the age category 25-35 years old (41%). But in the all the experimental groups, most respondents were in the age category < 25 years. Based on this data, the two control variables also seemed to have an effect on each other. Respondents from the control group, with a higher age, in general mostly live with a partner (with or without a child). While respondents from the experimental groups, which are of a younger age (in general), live on their own (without any children). Which makes sense, knowing that most of the respondents from the younger age groups who filled in the questionnaire were students.

With this in mind, a Kruskal-Wallis test was performed between only the experimental groups, no control variables showed significant differences and were thus comparable:

Figure 6 – Kruskal-Wallis between experimental groups

So there seemed to be a difference in the control group for the variables ‘age’ and ‘family-situation’. A next important question was then whether these two variables had a significant influence on the dependent variable. In this case the dependent variable was general brand

Joint branding alliance & communication : their effect on the central brand image and the transfer of brand and category related associations

Master Thesis