Understanding Viral Marketing: The Role of Brand Awareness and Association, Slides of Marketing

Download Understanding Viral Marketing: The Role of Brand Awareness and Association and more Slides Marketing in PDF only on Docsity! BACHELOR THESIS Fall 2009 Kristianstad University college International Business and Economics Program The Viral makes you aware -How is brand association affected by viral marketing through individual networks? Writers Ida Nilsson Magnus Svensson Supervisor Christer Ekelund Examiner Timurs Umans www.hkr.se Abstract Nowadays the internet has become more and more important in today’s technology addicted society. Consumers of today use the internet as a tool to communicate, do shopping and search for information. Hence, the internet has become a convenient way for organizations to reach out to consumers in an easy and cheap way, specially through the tool viral marketing. In this paper we research the phenomenon brand awareness and brand association affected by viral marketing. This is done through two research questions: How does a commercial affect consumers to create a network Viral marketing campaign? And Why do consumers purchase a brand when they are reacting negatively on a commercial?. To answer the research questions we chose a well discussed Swedish commercial from Apoliva as an example, especially since the commercial has received many negative reactions. Based on the commercial, we made a survey on the internet community site Facebook to measure how and why users reacted to the commercial and spread the reactions further; if and how their opinions changed toward the commercial and if and why consumers buy from the brand. The conclusions we draw from the research questions are that consumers tend to buy a brand although they do not like the commercial related to the brand, since the commercial itself is not of importance and in most cases the consumers do not relate the commercial with the brand. Furthermore, consumers tend to discuss a commercial if it is not a mainstream commercial. Although consumers discuss with their network on the internet, most of the discussion tend to be mouth-to- mouth and the spreading of opinion on the internet is mostly to vent their feelings. Keywords: Viral marketing, Brand association, Brand awareness 4.4.3 Hypothesis 3: Viral marketing transmitter spread the word to vent their feelings. ..............................................................................................................37 4.4.4 Hypothesis 4: If networks have a negative opinion, then the individual might change his/hers opinion, due to the related image. ..............................................38 4.4.5 Hypothesis 5: A memorable commercial creates more brand awareness than a mainstream commercial. .....................................................................................39 4.4.6 Hypothesis 6: Previous positive relations to a product make it difficult to change to a negative opinion. .............................................................................40 4.5 RELIABILITY ..................................................................................................40 4.6 VALIDITY ......................................................................................................41 4.7 GENERALISABILITY ........................................................................................43 4.8 RESPONSE RATE .............................................................................................43 5. RESULTS AND ANALYSIS ........................................................................45 5.1 STATISTICAL ANALYSIS AND RESULTS.............................................................45 5.1.1 Hypothesis 1 ..............................................................................................46 5.1.2 Hypothesis 2 ..............................................................................................47 5.1.3 Hypothesis 3 ..............................................................................................52 5.1.4 Hypothesis 4 ..............................................................................................56 5.1.5 Hypothesis 5 ..............................................................................................63 5.1.6 Hypothesis 6 ..............................................................................................66 5.2 SUMMARY .....................................................................................................69 6. CONCLUSIONS ...........................................................................................70 6.1 SUMMARY OF DISSERTATION ..........................................................................70 6.2 CONCLUSIONS................................................................................................71 6.3 CRITICAL REFLECTIONS ..................................................................................74 6.4 FUTURE RESEARCH ........................................................................................75 6.5CONTRIBUTIONS .............................................................................................75 6.5.1 Practical Contributions ..............................................................................75 6.5.2 Theoretical Contributions ..........................................................................75 REFERENCES .................................................................................................76 APPENDICES APPENDIX 1A QUESTIONNAIRE SWEDISH ..............................................................78 APPENDIX 1B QUESTIONNAIRE ENGLISH ..............................................................81 APPENDIX 2 CALCULATIONS, ANALYSIS ...............................................................84 LIST OF TABLES TABLE5. 1 ANOVA ARE YOU ACTIVE IN FORUMS/HOW MANY HOURS DO YOU SPEND ON INTERNET .......................................................................................................46 TABLE 5. 2 FREQUENCY WHERE HAVE YOU SEEN THE COMMERCIAL .....................48 TABLE5. 3 ANOVA HAVE YOU DISCUSSED THE COMMERCIAL WITH YOUR NETWORK/WHERE HAVE YOU SEEN THE COMMERCIAL ..........................................49 TABLE5. 4 ANOVA HAVE YOU DISCUSSED THE COMMERCIAL WITH YOUR NETWORK/WHAT DO YOU THINK OF WHEN YOU HEAR APOLIVA ............................50 TABLE5. 5 ANOVA HAVE YOU DISCUSSED THE COMMERCIAL WITH YOUR NETWORK/ WHAT DO YOU THINK ABOUT THE APOLIVA COMMERCIAL ...................51 TABLE5. 6 CHI-SQUARE ARE YOU ACTIVE IN FORUMS/HAVE YOU DISCUSSED THE COMMERCIAL WITH YOUR NETWORK ....................................................................53 TABLE5. 7 CHI-SQUARE HAVE YOU TAKEN STAND AGAINST THE APOLIVA COMMERCIAL ON SOCIAL NETWORKS/ARE YOU ACTIVE IN FORUMS ON SOCIAL NETWORKS ..........................................................................................................54 TABLE5. 8 CHI-SQUARE HAVE YOU DISCUSSED THE COMMERCIAL WITH YOUR NETWORK/HAVE YOU TAKEN A STAND AGAINST THE APOLIVA COMMERCIAL ON SOCIAL NETWORKS ..............................................................................................55 TABLE5. 9 FREQUENCY DO YOU LISTEN TO ADVICE FROM YOUR NETWORK ...........57 TABLE5. 10 FREQUENCY HAVE YOUR OPINION TURNED BETTER OR WORSE AFTER THE COMMERCIAL ................................................................................................57 TABLE5. 11 FREQUENCY HAVE YOU BOUGHT THE PRODUCT .................................58 TABLE5. 12 ANOVA HAVE YOU BOUGHT THE PRODUCT/WHAT DO YOU THINK ABOUT THE COMMERCIAL ....................................................................................58 TABLE5. 13 CHI-SQUARE HAVE YOUR OPINION TURNED TO BETTER OR WORSE/HAVE YOU BOUGHT THE PRODUCT .................................................................................59 TABLE 5.14 FREQUENCY WHEN DID YOU START USING THE PRODUCT ...................60 TABLE5. 15 FREQUENCY HAVE YOU SEEN THE COMMERCIAL ...............................61 TABLE5. 16 ANOVA WHAT IS YOUR OPINION ABOUT THE COMMERCIAL/WOULD YOU BUY THE PRODUCT .......................................................................................62 TABLE5. 17 FREQUENCY WHAT BRAND DO YOU THINK OF WHEN IT COMES TO SKIN- CARE PRODUCTS ..................................................................................................63 TABLE5. 18 FREQUENCY DO YOU KNOW THE BRAND APOLIVA..............................64 TABLE5. 19FREQUENCY HAVE YOU DISCUSSED THE COMMERCIAL WITH YOUR NETWORK ............................................................................................................64 TABLE5. 20ANOVA HAVE YOU DISCUSSED THE COMMERCIAL WITH YOUR NETWORK/WHAT IS YOUR OPINION ABOUT THE COMMERCIAL ...............................65 TABLE 21ANOVA WHEN DID YOU START USING THE PRODUCT/WOULD YOU BUY THE PRODUCT ......................................................................................................67 TABLE5. 22 CHI-SQUARE HAVE YOU DISCUSSED THE COMMERCIAL WITH YOUR NETWORK/WHEN DID YOU START USING THE PRODUCT .........................................68 LIST OF FIGURES FIG. 2.1 THE RESEARCH ONION............................................................................. 7 FIG. 3.1 CUSTOMER BASED BRAND EQUITY ...........................................................14 FIG. 3.2 BRAND ASSOCIATION .............................................................................16 FIG. 3.3 VMC EVALUATION MODEL ....................................................................22 FIG. 3.4BRAND ASSOCIATION THROUGH THE VIRAL MARKETING MODEL (BAVM)28 FIG. 6.1BRAND ASSOCIATION THROUGH THE VIRAL MARKETING MODEL (BAVM)72 1 1. Introduction This first part contains information about the background to our topic which is viral marketing and brand awareness/association. Further we discuss the problems, purpose and limitations of this study. We also present our research question. 1.1 Background Marketing of today is a science, where knowledge of consumer behaviour is becoming more essential when it comes to reaching people, together with the ability to make your brand noticed among the masses of messages (Keller, 1993). Meanwhile, knowledge of marketing channels is even more complex, from the first rumour spreading to newspaper, radio and television. Even the newest channel of marketing, the internet, is evolving into new channels within itself (Rowley, 2004). Among these internet channels we find new names and words like Facebook, Twitter, blogging and Youtube. These are all names of different social-communities or networks where people meet, exchange information and receive marketed messages every day. Marketers now learn how to use new trends in marketing such as viral marketing also known as buzz marketing and word-of-mouse to mention some. This is far from a recent phenomenon, as Ferguson (2008) says; ―before the advent of the printing press, broadcast media and the internet, word-of-mouth was the only way to market your goods.‖ (p.180). It is possible to spread the word of a new product or service at light speed, generate buzz at the approximate level of a Hollywood premiere and leverage brand evangelists to encourage trial and activation (Ferguson, 2008). Since the invention of the internet, people around the world have come closer to each other more rapidly than ever before and Marshall McLuhan´s expression the global village from 1964 is now a reality. This closeness has given an excellent chance to leave the ordinary world and form communities on the internet where friends keep in touch with writing on each others walls, sending emails or discuss 4 1.4 Research question For a guideline throughout this paper we have chosen to work from the phenomenon brand awareness and brand association affected by viral marketing. The phenomenon is studied through the research questions; A: How does a commercial affect consumers to create a network Viral marketing campaign? B: Why do consumers purchase a brand when they are reacting negatively on a commercial? To research these questions we look at the connections between different networks and how people spread the word, especially over the internet, thus affecting the consumers´ brand awareness/association. The research questions turns to organizations, what they can have in mind when creating a viral marketing campaign or a commercial/ad that could turn into a viral marketing campaign, due to attention from the consumers. 1.5 Limitation After reviewing literature and articles in the field of viral marketing and brand awareness/association, we are working after two specific models. The models that we mainly work after are Viral Marketing Communication Evaluation Framework by Cruz and Fill, 2008, that is explaining how viral marketing works. The second model that we work after is Brand Association by Blythe, 2008. After reviewing those two models we created our own model that combines both viral marketing and brand awareness/association, but for our hypotheses we also look at the rest of the theories in chapter 3 in our paper. We have limited our self to do our questionnaire about one specific skin-care product brand called Apoliva. The commercial about Apoliva will aid us in the research as an example of a commercial turning into viral marketing. Because we are doing our survey based on one specific product, it can not be applicable on other type of products, but hopefully give a guideline in how the consumer becomes affected of commercials. 5 1.6 Outline This paper has the following outline. Chapter 1- Introduction: This first part contains information about the background to our topic which is viral marketing and brand awareness/ brand equity. Further we discuss the problems, purpose and limitations of this study. We also present our research question. Chapter 2- Research method: In this chapter we discuss our research approaches we use throughout our dissertation. We are going to explain if it is a positivism, interpretivism or realism research philosophy and also if it is a deductive or inductive approach. We also explain the meaning of these different philosophies and approaches. Chapter 3- Theory: The purpose of this chapter is to explain theories in brand awareness and brand association together with the expression viral marketing: what it is, does it work and how to use it. We also explain the connection between these two important factors in marketing, viral marketing and brand awareness/association. Chapter 4- Empirical method: In this chapter we present our empirical method. We explain the research design and strategy together with time horizon, credibility and operationalism. Chapter 5- Results and analysis: This chapter presents the results of the research. The results of the six hypotheses will be brought forward and analysed and, with the support from SPSS, either accepted or rejected. Chapter 6- Conclusion: In this chapter we start with a summary of our paper, further we present the findings of our analysis and critical reflections. We also suggest future research within our chosen topic and how the research could be used in reality. 6 2. Method In this chapter we discuss our research approaches we use throughout our dissertation. We are going to explain if it is a positivism, interpretivism or realism research philosophy and also if it is a deductive or inductive approach. We also explain the meaning of these different philosophies and approaches. 2.1 Choice of methodology The purpose with this paper is to see how much viral marketing affects the consumer’s purchasing decision, in combination with brand awareness and brand association. We have chosen to focus on the question why the consumer still buys the product even if he/she does not like the ad or commercial and how much the network around the consumer affects the decision in purchase or not. Therefore, we start by studying the already existing literature and articles. After that we continue to create our own model to work after and then a survey. A survey is the best choice because we want to reach as many as possible in the medium we focus on, the internet. This questionnaire is e-mailed out on a social community all at one time with the possibility reply within a week. 2.2 Research philosophy Research philosophy is important for the researcher in order to know in which direction and from which point of view the research is conducted. The researcher should be aware of the choices he/she make and in which mindset the direction should take. This so he/she can defend the research in the end and so there is a connection throughout the parts in the paper. 9 questions to obtain a deeper insight into the subject, hence, we rule out the other alternatives and use the positive approach. 2.3 Research approach When it comes to research approach there are two different approaches, deductive and inductive research. Deductive research is called a scientific research and represents the ordinary look on the connection between research and theory. The researcher uses what is already known from previous research and theories to create a hypothesis that needs to be subjected to empirical testing. The testing is made by collecting data and then comparing the findings with the hypothesis and thus confirming or rejecting it. To finalise the deductive research, the researcher implicate the findings for the theory that is used as foundation for in the study and might then complement the theory with his/hers conclusion (Bryman & Bell, 2007). While deductive research use a theory or model as base for the observations, the inductive research start with observations which will lead to an existing theory or developing a new one (ibid.). The inductive approach is more open for suggestions and new theories and do not lock itself into only one option. When having an inductive approach it is more common to use a small sample and have qualitative data instead of quantitative. If the researcher is interested in why something is happening more than what is happening, an inductive research is better (Saunders et al., 2009). The inductive approach is often characterised with qualitative studies and normally considered a strong approach when it comes to developing new theories. However, according to Bryman & Bell, (2007), while inductive researchers create informative and helpful findings it is often without a clear theoretical importance, ―They provide insightful empirical generalizations, but little theory‖ (p.14). In this paper we use a deductive research approach since we use existing theories and models as foundation to the hypotheses we develop. The hypotheses are empirically tested and data is collected through the use of a questionnaire. From the tests we link back to the hypotheses and theoretically based research questions. 10 2.4 Choice of theory We reviewed different scientific articles and literature from the areas marketing and viral marketing together with brand awareness and brand association. All the chosen articles come from well known business researchers and lecturers of the subject. Highlighted authors are Gezelius & Wildenstam in marketing, De Bruyn & Lilien, De Valck and Dobele in viral marketing, Keller in brand awareness and Keller and Leone et al. in brand association. From the theories found we have chosen one model from each subject and created our own model which our research is based upon. From the model and theories six hypotheses where created to aid in answering our research question. 11 3. Theory The purpose of this chapter is to explain theories in brand awareness and brand association together with the expression viral marketing: what it is, does it work and how to use it. We also explain the connection between these three important factors in marketing, viral marketing and brand awareness and brand association. 3.1 Marketing Marketing is a necessity for all organizations. Marketing is important to reach the future customer and make him/her aware that products or services do exist. Today it is hard to capture the customers and to reach him/her through the buzz that exists, because there are many fighting for the interest and shares on the market (Gezelius & Wildenstam, 2007). There are many different things that are included into the term marketing, for example commercials, ads, sponsorships, events, personalized mail and personal selling. All this comes together to create a need for the consumer. This need might already exist and if it does not then it is marketers’ duty to convince the future consumer that life will be better with the product or service (ibid.). Marketing can be very expensive because marketers are trying to push for their product and to make a product visible at all situations can be costly. Due to the fact that all companies are trying to be cost efficient there are new trends and new ways to reach out to customers created all the time. That makes internet a, relatively, new way for marketers to use and especially a fast way to spread information. The so called viral marketing or word-of-mouth is one way to reach out. This means that the clients spread the message voluntarily to its network. It is also an efficient way to spread the message about a new brand or brand association (ibid.). 3.2 Brands Branding is the way to make yourself known. By using a brand a wide range of advantages will be found, as awareness, association and identification together with the probability of brand loyalty (Hollensen, 2007). However, the main reason 14 Fig. 2.1 Customer based brand equity (Note: from Hollensen , 2007, p.440) In this paper Brand Quality, Brand Loyalty and Propriatry Assets are not intended to be brought up further. According to Anderson et al., (2009), brand equity is reflected in ―actions or responses of present or prospective customers‖ (p.140). When they are willing to try what is offered; when the consumer is more unwilling to try the competitors offerings; if they are likely to purchase what is offered; if they are willing to pay the higher price than offered by competition; if the customer is unwilling to switch to competitors when price rises; if they reduce thinking time in closing a sale of the offered product; if the consumer is willing to buy a larger share of what is required. Anderson et al., (2009), continues that through the offerings presented and through the special quality, functionality and performance the brand equity is built, but, furthermore, through the specified branding strategy the company use. But still, to build brand equity, the brand itself must become known, a fact which according to Hoeffler and Keller (2002) makes brand awareness the most important first step when building brand equity, followed by brand association. Brand Awareness Brand Association Brand Quality Brand Loyalty Propriatory assets 15 3.2.2 Brand Awareness None of the categories or effects mentioned above can be used without first having brand awareness, no one buy from a brand if they are not aware of it. (MacDonald & Sharp, 2003; Keller, 1993) According to Keller, (1993), brand awareness is the combination of brand recognition and brand recalling. Brand recognition can be defined as if a customer have heard or seen a brand and then see or hear it again, the customer will remember the brand and distinct it from competing brands. Brand recalling is when recognising a product or service category, the customer’s mind retrieves the brand in relation to the product or service (Keller, 1993). However, according to Macdonald and Sharp, (2003), though the two concepts of brand recognition and brand recalling are well known in the marketing industry, the role they play when consumers make a decision is less well known. The authors continue that when a consumer is aware of a brand he/she will most unlikely look for another brand in the same category, but if the consumer relate a product to two or more brands, the brand that the consumer have been the most exposed to will be the one the he/she considers first, since he/she recognises that product first and most likely associates the brand with the product (MacDonald & Sharp, 2003; Keller, 1993). According to Hoeffler & Keller, (2002), brand awareness should also have depth and width. Depth being how easy a customer recalls and recognizes the brand while the width is how much the consumer think of the brand when buying and consuming. The authors continue that all brands should have both depth and width in the brand awareness since depth will give increased thinking of the brand when in need and width will increase the thinking of the brand linked to situation when consuming. 3.2.3 Brand Association As a consumer becomes aware of a brand and starts recognising and recalling it from memory, he/she most certainly also associates the brand with a feeling or an image that will help the memory of the brand (Keller, 1993; Leone et al., 2006). Furthermore, in most cases the brand holders want the consumer to associate the brand with a specific picture or situation, which is projected when the customers experience the brand. This makes the brand association related to the brand image (Leone et al., 2006). According to Keller, (1993), the brand image is how the 16 Direct experience of the product Information from trusted sources Brand exposure to specific product Word of mouth Marketing communications Development of brand association Acceptance of brand applied to different product: association of similar characteristics consumer reflects the brand from memory, while brand association is what the consumer likes about the brand; what about the brand that bring meaning to the consumer. Or as Leone et al., (2006), explains it: ―Brand image is defined as customer perceptions of and preferences for a brand, as reflected by the various types of brand associations held in customers’ memory.‖ (p.126). The image and association of the brand need to be together since the image will symbolize the qualities of the association. This is reflected in the consumer’s mind, since in all, the image refers to the combination of a real world entity and a mind entity that the consumer associates with the real world, or refers to an advert. In the end this brings an identity about the brand to the consumer (Keller, 1993; Stern, Zhinkan & Jaju, 2001; Gezellius & Wildenstam, 2007). An example is when most men hear the name Marlboro. They relate to a cool cowboy, the Marlboro Man, ―who is perceived as independent, mature, macho, and tough‖ (Jensen-Arnett, 2001, p.427). The picture of the cowboy becomes the image of the brand, so when you smoke the Marlboro cigarette the consumer will see themselves as this cool cowboy; therefore, the Marlboro cigarettes will be associated with a feeling of coolness. Fig. 3.2 Brand Association (Note: from Blythe, 2008, p.128) 19 experience and the consumer is a first time user of the product or service. If it is a more costly product or service the first time users will believe the negative word- of-mouth easier as well because then the risk is higher to lose money. Chevalier and Mayzlin as well as Dellarocas et al., (De Valck et al. 2009), done research on this phenomenon and compared book sales. They found that negative reviews have a more powerful impact than positive ones (De Valck et al., 2009). A negative word-of mouth is spread more easily than a positive one (Kiss & Bichler, 2008). For an example the movie industry is one of the industries that have noticed that. The new generation is always connected to the rest of the world, through emails, cell phones, social community sites and so on. Therefore, if someone out of the new generation sees a movie and does not like it he/she can tell the friends (network) immediately with a text message and the negative viral marketing has begun. This in return can affect a movie release to the extent that ticket sales can decrease in only one day. Hence, it is important for the movie industry to have a plan how to turn negative viral marketing to a positive one (Dobele et al., 2005). It is also very sensitive how the ad is created, if it is based on a sensitive hot topic that might be offensive to people or not. For example, Volkswagen Polo had an ad with a suicide bomber in a car which did not give a positive response and damaged the brand image (Porter & Golan, 2006). 3.3.3 Viral marketing strategies As mentioned before, to create a successful viral marketing campaign is difficult, but one company, Honda, managed well in April 2003. They understood the importance of creating both a nice looking advertisement and that would make people talk about it. Honda in the UK created a commercial that lasted for two minutes and they aired it for the first time at the Brazilian Formula 1 Grand Prix. This way they reached their target group, people with interest of cars, all at once. The people that were at the Formula 1 Grand Prix considered this ad to be very interesting and started to talk about it and also download it from the internet to pass it on to their network with the same interest. The brand awareness for Honda increased and also their sales the first quarter of 2004. Because it was a viral marketing ad, which could be seen over the internet, they reached not only the UK 20 but also the US. Honda in the US also noticed a positive difference in purchases and visitors to the Honda stores (Dobele et al., 2005). Five important key elements for the ad that was used by Honda and which are also important for viral marketing campaigns in general are described below. The ad should (Dobele et al., 2005, p. 146): 1. Be fun/interesting. The ad should be fun and interesting and capture the viewer or reader so that people want to talk about it. 2. Be well targeted. The Honda ad had the perfect timing for the right people with a broad leverage of consumers. 3. Combine technologies. Honda used different technologies to spread the word, both television and internet messaging. This particular ad was also discussed in television shows. 4. Have credible sources. The Grand Prix Formula 1 people are professionals, therefore, they are believable also when the word that was spread was voluntary word-of-mouth (WOM). The WOM made the formula 1 network spread the word and look the ad up on the internet. 5. Show attachment to the product that is easy to use or highly visible. The ties between the real world and the product were visible. This made the brand message likeable. Other brands that been successful in viral marketing are Burger King with the expression ―have it your way‖. This is a campaign on the internet where a man in a chicken suit follows orders from whomever with internet access. Also BMW launched a successful Hollywood film series campaign with Hollywood directors that produced the films (Porter & Golan, 2006). According to Cruz and Fill’s, (2008), model (fig. 3.3) a person or company chooses to make a viral marketing communication (VMC) campaign with the goal to reach and create awareness of the brand or product, to count downloads and rate how much is transferred and shared, to increase revenues of interest (ROI) and brand equity through the campaign. There are three types of goals with a marketing campaign: the cognitive goal is that the marketer wants to make people 21 aware of the product or brand; the behavioural goal shows how popular the specific link, ad etc has been and the more people talk about it the more popular it is; the financial goal shows if the ad, commercial etc is working and increasing the revenues (Cruz & Fill, 2008). The first step of the campaign is to find the target audience and where they reside on the internet. The second step is to find the message that attracts attention, either as a personal message or as open or hidden commercial. Step three is to use the correct media, email, mobile-phone or website, to reach the intended respondent. The final and fourth step of the VMC campaign preparations is to find the appropriate tool to link the message to the respondent; it can be done through an advert game, picture, text or video. When the preparations are made, the campaign is launched. After a while measurements of the results are made and evaluated against the objectives intended with the campaign, if it was intended to create awareness, downloads or ROI and brand equity. After evaluating, the decision to abandon, contain or reignite the campaign is made. 24 2. The Conversationalist visit the website 3-4 times a week for about half an hour and is there to discuss information. He/She is socially involved and a mature member. 3. The Informationalist visit the website 3-4 times a week for about half an hour. He/She visit the website both to retrieve and give out information. The informationalist is socially involved and a mature member. 4. The Hobbyist focus on personal webpages, guest books and profiles. He/She is socially involved and the educational level is relatively low. 5. The Functionalist visit the website for half an hour once a week and then go there to retrieve information. He/She is not socially involved and of the different member types this one has the highest education level. 6. The Opportunist visit the website less than once a week and then only for about 15 minutes to retrieve information. He/She is not socially involved in the community. There might be other types of members if the researcher would look into a different type of social community site with other interests or size. The main thing is that there are different types of members that have different importance for a social community site. For a marketer this is of importance so he/she can target the right way. To create a successful viral marketing it is necessary to know who the organization is targeting and that is when these member types comes of use. Knowing the purpose of the viral marketing determines who will spread the word. Keller and Berry sum it up in a good way: To succeed today, you need to connect with people who are at the center of the conversation… Specifically, you should make sure you are reaching the decision makers who are influential in others' decisions. Influentials are well connected, they have ties to a significantly larger number of groups than the average American (Kiss & Bichler, 2008, p. 234). In this case with the culinary website, it would be best to create a viral marketing campaign that reaches the conversationalist. This because the conversationalist is highly socially involved and he/she has an interest in the product or service that is discussed. He/She is a member that likes to share information and is well equipped to start a ―buzz‖ for the marketers. If the marketers approach an 25 opportunist, for example, the viral marketing campaign will never start because they are not socially involved at the social community site. Nevertheless, one important thing that should not be underestimated is the offline viral marketing that the opportunist contributes to because he/she still read the experts piece of advice and talk to the offline network. According to Ahuja, Michels, Walker and Weissbuch (2007) the offline conversation is still about 80% while the online are the leftover 20%. This means that the major part of viral marketing is taking place offline and not in front of a computer. Furthermore, according to the authors Ahuja et al., (2007), “teens are heavy users of both offline and online media, spending nearly 1.5 hours per day on the computer social networking.‖(p. 151-152). The offline behavior how the members act when they are not online on a social community or a website is not visible. This makes it hard to measure the offline behavior when it sometimes has a higher group influence than an online dialogue. The core member is important for the construction of a social community network, he/she is the expert at the subject at the site and other members listen to them. That makes the core member the creator of the site and in which direction the web site is taking depends on the core members’ knowledge. People are influenced by others in social communities, but the extent they become influenced has a strong connection if the community has a high heterogeneity or not. If the interests are the same and members are alike, the influence is higher than from other communities that has not a high rate of members with the same interests. Even if the members never meet in real life they listen to each other on community sites. Studies have shown that a person is more influenced on social community websites than if they visit the corporate webpage for a product or service (De Valck et al., 2009). How much a network will influence a member of a network such as a social community site depends on the structure and how the members interact in the network. The structural characteristics include the size, number of connections between individual and other members and the numbers of actual relationships in relation to how many members there are at the network. Interaction characteristics depend on how many members have the same interests in the network and also how strong the bonds are between the members. According to Henning Thurau et 26 al. in the article of De Valck et al., (2009, p. 187), there are eight different motivations for word-of-mouse participation: 1. Venting negative feelings 2. Concern for other consumers 3. Self- enhancement 4. Advice seeking 5. Social benefits 6. Economic incentives 7. Platform assistance 8. Helping the company Out of these eight it is the social benefit that is the strongest motivation to join a network and to have the highest rates of participating in the discussion forums. People join and visit social community sites for the same reason, social benefits (De Valck et al., 2009). 3.4 Connection brand awareness and viral marketing Since the early 1960’s marketers has been interested in how to spread information about products among the consumers and also how they can position their brand and change the image (Ahuja et al., 2007; Dobele et al., 2005). This leads to that viral marketing and brand awareness are connected considering viral marketing is a more efficient way create brand awareness in different markets than traditional advertising. The reason is because the information comes from a reliable source, the individual’s network (Cruz & Fill, 2008; Porter & Golan, 2006). The use of a viral marketing campaign as a strategy is mainly to increase the popularity of a brand, product or a company (De Valck et al., 2009). To be able to increase the brand support or awareness it is not only important to reach one individual but also to get the message to self-replicate through that individual’s network (work colleagues, family and friends). Today that occurs a lot through electronic communication, among others, to trigger the brand message (Dobele et al., 2005; Kiss & Bichler, 2008). 29 ‖To succeed today, you need to connect with people who are at the center of the conversation… Specifically, you should make sure you are reaching the decision makers who are influential in others' decisions.‖ (p. 234). We moreover support the hypothesis with Blythe´s Brand Association Model (fig. 3.2) and the theory of six different member types from De Valck et al., (2009). 2. Non Viral Marketing Campaign commercial can become a Viral Marketing Campaign. We believe that a commercial, not intended as viral marketing, that is spread and discussed on networks, become a viral marketing campaign. One example is the previous mentioned Honda commercial, which managed to create an attention factor that generated discussions in several networks. The hypothesis is aimed to aid in answering research question A, How does a commercial affect consumers to create a network Viral marketing campaign?. The hypothesis has its foundation in step one of the BAVM (fig. 3.4) and is also based on the Five Key Elements by Dobele et al., (2005). Moreover, we support the hypothesis according to the discussion from De Bruyn and Lilien, (2008, p. 151): ―The key driver in viral marketing is the effectiveness of unsolicited, electronic referrals to create awareness, trigger interest, and generate sales or product adoption.‖ 3. Viral marketing transmitters spread the word to vent their feelings. We believe that consumers who spread the commercial or communicated message, spread it to vent feelings they have associated with the commercial or message. One example is the Apoliva commercial, which made viewers start special forums and discussion groups on Facebook, among others. The hypothesis is aimed to aid in answering research question A, How does a commercial affect consumers to create a network Viral marketing campaign?. The hypothesis has its foundation in step two and four of BAVM (fig. 3.4) and furthermore based on the Eight Different Motivations for word-of-mouse Participation by De Valck et al., (2009). The Hypothesis is moreover supported by Hoeffler & Keller’s, (2002), theory about depth and width of the brand. We find further support from Leone et al.’s, (2006), theory about the minds of the consumer affecting actions and behavior toward the brand. 30 4. If networks have a negative opinion, then the individual might change his/hers opinion, due to the related image. We believe that when networks surrounding an individual consumer have a negative opinion associated with a commercial or brand, the individual is more liable to change his/hers opinion. This might result that the individual consumer will not buy the brand, since the networks negative opinion is associated with an image. One example is the previous mentioned Volkswagen Polo commercial with the suicide bomb, which damaged the brand image (Porter & Golan, (2006). The hypothesis is aimed to aid in answering research question B, Why do consumers purchase a brand when they are reacting negatively on a commercial?. The hypothesis has its foundation in step two and three in the BAVM (fig. 3.4) and is furthermore based on the Brand Association Model by Blythe, (2008), (fig. 3.2). The hypothesis and model is supported by the discussion by De Valck et al., (2009), about negative viral marketing which is easier to believe than positive viral marketing. Further support is from Dobele et al.’s, (2005), theory about sensitivity in ads and Kiss & Bichler, (2008), who say that negative word of mouth is spread more easily than a positive one. Leone et al.’s, (2006), theory about the relation between brand association and image is also supporting this. 5. A memorable commercial creates more brand awareness than a mainstream commercial. We believe that a commercial that makes consumers discuss and react are more liable to create a brand association. The brand association stays in their minds more and longer than the ordinary and everyday commercial, for example, toothpaste commercial or detergent commercial. One can assume that a main stream commercial is an average everyday commercial one example of a memorable commercial is the previous mentioned association between Marlboro cigarettes and the image of a cowboy. The hypothesis is aimed to aid in answering research question A, How does a commercial affect consumers to create a network Viral marketing campaign?. The hypothesis has its foundation in step one and two in the BAVM (fig. 3.4). The hypothesis is furthermore based on Keller’s, (1993), theory about brand recognizing and recalling together with consumers 31 association and referral to an advert. We support the hypothesis and theory with discussion from Leone et al., (2006), where the ―Brand image is defined as customer perceptions of and preferences for a brand, as reflected by various types of brand associations held in customers memory‖ (p.126). 6. Previous positive relations to a product make it difficult to change to a negative opinion. We believe that if a consumer have a positive relation to a product or brand, he/she is less able to change their opinion when approached with opposite thinking. The hypothesis is aimed to aid in answering research question B, Why do consumers purchase a brand when they are reacting negatively on a commercial?. The hypothesis has its foundation in step two and three in the BAVM (fig. 3.4). The hypothesis is furthermore based on the theory about brand knowledge by Leone et al., (2006), who says that it is not the facts about the brand that give the consumer brand knowledge, it is ―all the thoughts, feelings, perceptions, images, experiences and so on that become linked to the brand in the minds of customers‖ (p. 126). We support the hypothesis and theory with the theory about brands depth and width by Hoeffler & Keller, (2002), together with the discussion about brand association, exposure and recognition by MacDonald & Sharp, (2003). 34 studied subjects, for example in a board room. The structured observation ―is systematic and has a high level of predetermined structure‖ (Saunders et al., 2009, p. 300) and is collected through observations, interviews or questionnaires, which can be self- or interviewer administered (Saunders et al., 2009). In the primary data collection for our paper we have chosen to use a self administered questionnaire on the internet. The advantages for a questionnaire compared to interviews are time and cost saving (ibid.). Furthermore, since we used the internet as delivery method, we reach a large body of participants. What the questionnaire examines is if there is any correlation between the networks influence on purchasing behaviour, purchasing behaviour if the commercial is receiving negative attention and the effect on brand association in the context commercial. 4.3 Population A population is the total amount of people in a group that researchers want to test. Usually it is associated with the entire population of a nation. Since the amount of people in a population is especially large, researchers tend to test on a sample of the population. The sample is just a small portion of the population that represent the general opinion about the question asked (Bryman & Bell, 2007). There are two types of sample techniques: the probability and non-probability sample. The probability sample is when ―…each case being selected from the population is known and is usually equal for all cases‖ (Saunders et al., 2007, p. 206) and each case is able to be used in a general character. The non-probability sample is when every case is unknown and not selected by random (Bryman & Bell, 2007). The non-probability technique is divided into five types of sampling: Quota sampling which is completely random and based on the sample being a variability of different quotas in a population; Purposive sampling is when the authors choose their cases to best meet their objectives and questions; Snowball sampling is when the population is hard to find and each case found leads the researchers to the next case; Self-selection sampling is when the case comes to the researcher to take part in the study; finally the Convenience sampling is when the researchers chose cases that are easiest to reach, for example shoppers entering a shopping mall (Saunders et al., 2007). 35 Since we are studying viral marketing through networks, we concentrated on a population that is active in using networks. We chose to conduct a survey on a sample of the population using the online network and social community site Facebook. Furthermore, we use the type of convenience sampling since the samples chosen belong to our own network and are easy to reach. Our chosen sample is in the interest of study since they probably represent different groups in society and have different backgrounds; they are all active on the same network and participate in discussions within the network. Since we aim at studying the conversations and spreading of the word, the gender or background of the respondents are non-important and mainly used as control questions to see the spread and variety of respondents. Furthermore, since we are using the example of a Swedish commercial, which has not been shown in other countries, the specified sample of the population are people within the Facebook network who are living in Sweden and speaking Swedish. 4.4 Operationalisation To reach our main reason of this study a questionnaire is sent to the sample group. Our sample group is our network at the social community site Facebook where we are able to reach different gender, age groups, professions and education level. We use question one through four (see appendix 1b) in our questionnaire as control questions to make sure that we reached a variety of people. Also we only needed people to fill out our questionnaire that are aware of the product we are using as an example. Hence, we had two control questions; if they are aware of the brand and if they have seen the commercial. If the respondents answer no on either of these questions the questionnaire automatically thank them for their cooperation and they can not continue with further questions. On some multiple answer questions we have one answer that is other, here the respondents can fill out their own opinion. This was important for us to see if we had missed any potential answer. 4.4.1 Hypothesis 1: The amount of time spent on social community sites equal higher liability to discuss in forums. In our first hypothesis we want to find out if there is a relation between times spent on social communities and if they are a spreader of messages, a 36 conversationalist, which marketers can target to spread their word. To find out if this hypothesis is correct or not we use question 5, 6 and 18 in our questionnaire. We begin with putting question 6, if the respondents are active on forums, together with question 5, how much time the respondent spend per week at social communities, in an ANOVA table to see if there is any relation between them. In question 5, we choose to rank the hours spent on internet social community sites according to Ahuja et al., (2007), who says heavy usage is 1.5 hours per day which equals about 11 hours per week. 4.4.2 Hypothesis 2: Non Viral Marketing Campaign commercial can become a Viral Marketing Campaign. In the second hypothesis we want to research the fact that even if a television- commercial is not a viral marketing campaign to start with, it is possible for the commercial to become one after a while with special attention. This because the television-commercial is spread over several different media channels since it is an interesting commercial and not mainstream. To find an answer to how mainstream the commercial is, we first look at the frequency table of question 12, how many of our total group of respondents have seen the commercial. The frequency show if the commercial is mainstream, since a mainstream is less memorable. If the frequency is high, the more respondents remember the commercial and the less mainstream is the commercial. We continue with a cross-tabulation of the questions 12, if the respondents have seen the commercial, together with question 16, if the respondent has discussed Apoliva with the network. This show if it matter where the respondents have seen the commercial and if everyone are spreaders of the message. After the cross-tabulation with a Chi-square test of questions 12 and 16, we make an ANOVA table of question 11, what the respondents associate with the brand Apoliva, and question 16. We want to know if it was of importance what the respondents think about when they hear the brand name and how much they talk 39 question17, how many that have bought the product. Then we can see if any of the respondents have changed their opinion for worse but still buy the product. To finally find the connection between the previous tables we conduct an ANOVA-test between question 14, what are the respondents’ opinions about the commercial, and question 19, if the respondents would buy the product. 4.4.5 Hypothesis 5: A memorable commercial creates more brand awareness than a mainstream commercial. Our fifth hypothesis is built upon that people become aware of the brand when the commercial is memorable in one way or another. An example would be that people do not remember toothpaste commercials and can not separate them from each other but if the commercial have an outside the box thinking, people will remember the commercial and recall it in a discussion. To research about this hypothesis we start with doing four different frequencies tables from question 9, what brand the respondents think of when they hear about skin care products, question 10, if the respondents are familiar with the brand Apoliva, question 12, if the respondents have seen the commercial about the specific brand, and question 16, if the respondents have discussed the commercial with their network. The first frequency table show if the respondents think about the brand Apoliva before we mention it in the questionnaire. After that we want to know how many know of the brand or have seen the commercial of Apoliva when we mention it in the questionnaire. The last table will show how many of the respondents have seen the commercial and remember it. We continue with an ANOVA-test of question 14, what is the respondents´ opinion about the commercial, and question 16, if the respondents have discussed the commercial with their network. The test will prove that the commercial is memorable enough, for the respondents, to discuss with their network. 40 4.4.6 Hypothesis 6: Previous positive relations to a product make it difficult to change to a negative opinion. Our sixth and last hypothesis is about if an existing positive experience of a brand or product can become affected by negative opinions. If a customer already has a positive relation to the product it is harder to change that opinion to a negative one. To research this hypothesis we start with a frequency test of question 18, when the respondents began to use the product. This to know if they had begun to use it before or after the commercial started broadcast and show if they had a previous experience. We follow the frequency test with an ANOVA table of question 18, when the respondent begun to use the product, and question 19, if the respondent would buy the product. This is to find if the respondents´ opinion has changed from positive to a negative. Furthermore, we can analyze why has the opinion changed since they bought the product the first time, if the respondents´ have changed opinion because of the commercial. To prove if the network have an effect on the opinion changing, we do a cross tabulation of question 18, when the respondents began to use the product, and question 16, if they have discussed the commercial with their network. 4.5 Reliability The techniques in collecting data need to give findings that are consistent and give comparable answers as similar studies from other researchers. The reliability depends on three major questions according to Saunders et al. (2009), if the measurements will have the same results at another occasion; if equal observations will be found by another researcher; if there is transparency in the results found by the researched data. However Saunders et al., (2007, p. 149), continues, the reliability can be threatened by four problems: 1. The participant error. It occurs when a participant might change opinion during the test and because of that answer differently between tests or questions. To 41 avoid this dilemma, it can be advised to use as neutral setting as possible for the interview or questionnaire. 2. The bias of the participant. It occurs when a participant answer questions as they think they should be answered or answer in a way that makes him/her seem better and because of that lie in the answers. This dilemma is avoided through using as high rate of anonymity as possible and in analysing data. 3. The interviewer error. This occur when the interviewer or researcher ask questions or commit the research in a way that lead the participant to a specified answer or behaviour. To have an organised plan of the interview or interviewer behaviour lessens the risk of this dilemma. 4. The interviewer bias. This occurs when the interviewer has a specific opinion before the analysis or when two or more interviewer interprets the same data in different ways. To avoid this dilemma a defined plan of interpretation of the data should be applied, or more than one interviewer should analyse the data. To make the data in this paper reliable we send the questionnaire to the respondents where they can answer in their own comfortable setting, furthermore, when the questionnaire is sent back it is unchangeable which together rules out the participant error. The questionnaire is anonymous and can not be traced back to the respondents which rules out the bias of the participant. Since the researchers are not present when the respondents answer the questionnaire and rule out the possibility of interviewer error. The answers of the questionnaire go directly to a database and statistical software which rules out the dilemma of the interviewer bias. 4.6 Validity According to Bernard, (2000), ―validity refers to the accuracy and trustworthiness of instruments, data, and findings in research.‖ (p. 46). The data collected in the research are supposed to contribute to the result and the data shall not be unconnected to the research. This means that variables shall be able to have a relation to each other and not be too far stretched, for example: desks and chairs might have a relation to each other, while desks and sheep might not. 44 Our questionnaire was sent out over email on the social community Facebook. We used own network the social community site and emailed the questionnaire to 338 persons that were eligible to answer our questionnaire. Out of those 338 we received 223 answers during the four day test period. This makes the response rate to 65, 9 percent (223/338=0,659) and the calculation is based on Neumann’s response rate calculation according to Saunders et al. (2009). The reasons why all in the target group has not replied could be because they have not seen the email or thought that it was not of importance and deleted the email. Also the respondents might not been online on Facebook during the test period or thought they could answer it later but then the time run out, due to the fact that we did not put in the final date to reply in the email (Saunders et al., 2007). 45 5. Results and analysis This chapter presents the results of the research. The results of the six hypotheses will be brought forward and analysed and, with the support from SPSS, either accepted or rejected. 5.1 Statistical analysis and results In order to be able to do this research, we used SPSS statistics software. In the program we used four different types of tables to analyse our data, Chi-square test and ANOVA test together with Descriptives table, Frequency table and Cross- tabulation table. For calculations that are not from SPSS, see appendix 2. The Chi-Square test is used to establish the confidentiality of a relationship between two variables in a population or sample. The higher the value of the Chi- Square test, the less chance of relation between the variables and there will be no significant relation between them. When we tested the Chi-Square, we had used the significance level of 0.05, meaning that tests above 5 percent do not have significant relation and are, therefore, rejected. We used the Chi-Square test in combination with Cross-tabulation test to find a relation between questions with two different answers; however with more than a total of 4 squares it is not possible to do this test (Bryman & Bell, 2007). The ANOVA table is used when the answers possibilities are above two. It is a test that sees if the means come from same population or not. One requirement to do an ANOVA test is that all the standard deviations have to be the same. We have decided that the significance level is 0.05, meaning not higher than five percent (ibid.). The Descriptive table is used together with the ANOVA test to see how many respondents who are answering yes or no, where 1 equals Yes and 2 equals No in the means. Frequency tables are used when we needed to see the ratio between the different answers in numbers and percentage. Meanwhile, the Cross-tabulation table is used 46 when two answer variables are compared in numbers and percentage (Bernard, 2000). 5.1.1 Hypothesis 1 The amount of time spent on social community sites equal higher liability to discuss in forums. In the ANOVA test in table 5.1 we try the relevance between the questions Are you active on forums? and How many hours do you spend on social community sites per week?. We can see a significant relevance of 0.000. According to the test, a large group of the respondents (48.4 percent) spend four to eleven hours on the internet, but this group is not the highest percentage when it comes to discussing on forums. Though the respondents who spend 12 hours or more on the internet are not a majority (8.5 percent), they are more active in forums. This proves that the more hours the respondents spend on the internet, the higher is the tendency to discuss in social communities and spread VMC. However, though the majority are not the biggest time spenders on internet or a high percentage on forums, they still exceed the 12-hours-or-more respondents in forums in number. Table5. 1 ANOVA are you active in forums/How many hours do you spend on internet How many hours do you spend on internet N Mean Std. Deviation Std. Error 95% Confidence Interval for Mean Minimum Maximum Lower Bound Upper Bound 3 hrs or less 96 1.83 .375 .038 1.76 1.91 1 2 4 to 7 hrs 78 1.58 .497 .056 1.46 1.69 1 2 8 to 11 hrs 30 1.60 .498 .091 1.41 1.79 1 2 12 hrs or more 19 1.32 .478 .110 1.09 1.55 1 2 Total 223 1.67 .472 .032 1.61 1.73 1 2 49 Table5. 3 ANOVA Have you discussed the commercial with your network/Where have you seen the commercial Where have you seen the commercial? N Mean Std. Deviation Std. Error 95% Confidence Interval for Mean Minimum Maximum Lower Bound Upper Bound TV-commercial 124 1.54 .500 .045 1.45 1.63 1 2 You-Tube 1 2.00 . . . . 2 2 Other 3 2.00 .000 .000 2.00 2.00 2 2 In two or more places, for example Facebook, Youtube, TV 21 1.19 .402 .088 1.01 1.37 1 2 Total 149 1.50 .502 .041 1.42 1.58 1 2 ANOVA Have you discussed the commercial with your network Sum of Squares Df Mean Square F Sig. Between Groups 3.212 3 1.071 4.561 .004 Within Groups 34.036 145 .235 Total 37.248 148 In the ANOVA test in table 5.4 created by the two questions Have you discussed the Apoliva commercial with your network? and What do you think of when you hear Apoliva?, we find no significant proof of any connection between these two questions (0.093). But we can see the tendency in the table that most people (54.4 percent) associate the brand Apoliva with the commercial, instead of the product itself. Therefore, the discussion about Apoliva is mostly about the commercial but the connection between the commercial and product is forgotten. In the table we can see that some in the population (3.3 percent) associate Apoliva to a specific ingredient in the commercial, which is the female model. 50 Table5. 4 ANOVA Have you discussed the commercial with your network/What do you think of when you hear Apoliva What do you think of when you hear Apoliva N Mean Std. Deviation Std. Error 95% Confidence Interval for Mean Minimum Maximum Lower Bound Upper Bound Skin care products 46 1.63 .488 .072 1.49 1.78 1 2 The commercial 81 1.43 .498 .055 1.32 1.54 1 2 Nordic climat 4 1.25 .500 .250 .45 2.05 1 2 Other 4 1.75 .500 .250 .95 2.55 1 2 Pharmacy 9 1.67 .500 .167 1.28 2.05 1 2 Scary woman 5 1.20 .447 .200 .64 1.76 1 2 Total 149 1.50 .502 .041 1.42 1.58 1 2 ANOVA Have you discussed the commercial with your network Sum of Squares df Mean Square F Sig. Between Groups 2.354 5 .471 1.930 .093 Within Groups 34.894 143 .244 Total 37.248 148 The ANOVA test in table 5.5 is based on the question Have you discussed the Apoliva commercial with your network? together with What is your opinion about the commercial?. We find a significant connection of these two questions (0.000). What we can see in the table is that respondents who have an unpleasant opinion about the commercial (58.4 percent), have a strong tendency to discuss the commercial and their opinion about it with their network, both online and offline. Though having a negative opinion, their discussion contributes to spreading the message and creating brand awareness. This result shows that the commercial has an ingredient that creates an opinion and is tempting for respondents to discuss. 51 Table5. 5 ANOVA Have you discussed the commercial with your network/ What do you think about the Apoliva commercial What do you think about the Apoliva commercial? N Mean Std. Deviation Std. Error 95% Confidence Interval for Mean Minimum Maximum Lower Bound Upper Bound Good- Artistic 26 1.65 .485 .095 1.46 1.85 1 2 Good- highlight the product 7 1.71 .488 .184 1.26 2.17 1 2 Good- Relates to the message 11 1.82 .405 .122 1.55 2.09 1 2 Good- Other 18 1.72 .461 .109 1.49 1.95 1 2 Bad- Scary 43 1.19 .394 .060 1.06 1.31 1 2 Bad- No message 31 1.55 .506 .091 1.36 1.73 1 2 Bad- Music not related 5 1.20 .447 .200 .64 1.76 1 2 Bad- Other 8 1.62 .518 .183 1.19 2.06 1 2 Total 149 1.50 .502 .041 1.42 1.58 1 2 ANOVA Have you discussed the commercial with your network Sum of Squares df Mean Square F Sig. Between Groups 7.824 7 1.118 5.356 .000 Within Groups 29.425 141 .209 Total 37.248 148 To summarize: there is significant proof that more often a respondent view a commercial the more likely he/she is to discuss the message with his/her network. Furthermore, we find a tendency that respondents need to have an association of the brand to be able to discuss the commercial with their network. To finalise hypothesis 2, we find significant proof of the more negative opinion the respondents have about the commercial, the higher tendency to discuss it. Together, these results approve hypothesis 2. 54 Table5. 7 Chi-square Have you taken stand against the Apoliva commercial on social networks/Are you active in forums on social networks Are you active in forums on the internet (social communities) Yes No Total Taken stand against the Apoliva's commercial on social networks Yes Count 4 12 16 Expected Count 5.2 10.8 16.0 % within Are you active in forums on the internet (social communities) 8.3% 11.9% 10.7% No Count 44 89 133 Expected Count 42.8 90.2 133.0 % within Are you active in forums on the internet (social communities) 91.7% 88.1% 89.3% Total Count 48 101 149 Expected Count 48.0 101.0 149.0 % within Are you active in forums on the internet (social communities) 100.0% 100.0% 100.0% Chi-Square Tests Value Df Asymp. Sig. (2- sided) Exact Sig. (2- sided) Exact Sig. (1- sided) Pearson Chi-Square .427a 1 .513 Continuity Correctionb .137 1 .711 Likelihood Ratio .445 1 .505 Fisher's Exact Test .585 .365 Linear-by-Linear Association .424 1 .515 N of Valid Cases 149 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.15. b. Computed only for a 2x2 table In the table 5.8 we did a Cross-tabulation in combination with a Chi-square test and used the questions Have you discussed the Commercial with your network? 55 and Have you taken a stand against Apoliva’s commercial on any social community site?. The test shows a high significance (0.000), between the two questions. The test proves that there is a connection between if the respondents have discussed the Apoliva commercial with their network and if they have taken an active stand against the commercial. Out of all the respondents that took a stand against the Apoliva commercial, 93.8 percent are also discussing the commercial. The majority of the population has not taken a stand against the commercial (89.3 percent). But we find that though many people discuss the commercial on internet forums and in that way spread their opinion about a specific subject, most do not take an active stand against it. Table5. 8 Chi-square Have you discussed the commercial with your network/Have you taken a stand against the Apoliva commercial on social networks Taken stand against the Apoliva's commercial on social networks Yes No Total Have you discussed the commercial with your network Yes Count 15 59 74 Expected Count 7.9 66.1 74.0 % within Taken stand against the Apoliva's commercial on social networks 93.8% 44.4% 49.7% No Count 1 74 75 Expected Count 8.1 66.9 75.0 % within Taken stand against the Apoliva's commercial on social networks 6.2% 55.6% 50.3% Total Count 16 133 149 Expected Count 16.0 133.0 149.0 % within Taken stand against the Apoliva's commercial on social networks 100.0% 100.0% 100.0% 56 Chi-Square Tests Value df Asymp. Sig. (2- sided) Exact Sig. (2- sided) Exact Sig. (1- sided) Pearson Chi-Square 13.936a 1 .000 Continuity Correctionb 12.030 1 .001 Likelihood Ratio 16.388 1 .000 Fisher's Exact Test .000 .000 Linear-by-Linear Association 13.842 1 .000 N of Valid Cases 149 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 7.95. b. Computed only for a 2x2 table To summarize: a majority of the respondents spend four hours or more on the internet every week. However there is no significant proof that the respondents at the same time are active discussers in social community forums. Furthermore, there is a significant relevance between people discussing in networks and venting their negative opinion about the Apoliva commercial. Respondents also tend to discuss the commercial offline than on the internet. All together this shows that venting of feelings and opinions are more important than the place of discussion. This combination of results approve hypothesis 3. 5.1.4 Hypothesis 4 If networks have a negative opinion, then the individual might change his/hers opinion, due to the related image. To see if we can reject or confirm our fourth hypothesis we start with a frequency table (5.9) of the question Do you listen to your family and friends before purchasing a skin-care product?. We found that 43.5 percent listen to their network before purchasing a skin-care product. Though we expected a higher percentage, this is still a rather high percentage of the respondents that are affected by their surroundings. 59 ANOVA Have you bought the product Sum of Squares df Mean Square F Sig. Between Groups .815 7 .116 .691 .679 Within Groups 23.736 141 .168 Total 24.550 148 In table 5.13 we did a Cross-tabulation together with a Chi-square test. We used the questions Have you bought the Apoliva product? and Have your opinion turned to the better or worse after the commercial?. We found that there is no significance between the two questions in the two sided test (1.000). However, of the small portion of the respondent (8.1 percent) that did change their opinion of Apoliva after the commercial, none had a favourable opinion or had bought the product of Apoliva. Table5. 13 Chi-square Have your opinion turned to better or worse/Have you bought the product Have you bought the product Yes No Total opinion turned better/worse after commercial Better Count 0 5 5 Expected Count .4 4.6 5.0 % within Have you bought the product .0% 45.5% 41.7% Worse Count 1 6 7 Expected Count .6 6.4 7.0 % within Have you bought the product 100.0% 54.5% 58.3% Total Count 1 11 12 Expected Count 1.0 11.0 12.0 % within Have you bought the product 100.0% 100.0% 100.0% 60 Chi-Square Tests Value df Asymp. Sig. (2- sided) Exact Sig. (2- sided) Exact Sig. (1- sided) Pearson Chi-Square .779a 1 .377 Continuity Correctionb .000 1 1.000 Likelihood Ratio 1.142 1 .285 Fisher's Exact Test 1.000 .583 Linear-by-Linear Association .714 1 .398 N of Valid Cases 12 a. 3 cells (75.0%) have expected count less than 5. The minimum expected count is .42. b. Computed only for a 2x2 table In table 5.14 we did a frequency table on the question When did you start using the product?. This test was conducted to find out if the respondents have bought and used the product before or after the commercial. Out of the total of 31 respondents that used the product the majority 74.2 percent used it before the commercial started to air. This means that most of them were not influenced by the commercial, and did not react to the negative attention the commercial received when they were in a purchasing position. Table 5.14 Frequency When did you start using the product Frequency Percent Valid Percent Cumulative Percent Valid For 6 months ago or less 8 3.6 25.8 25.8 For 7 months ago or more 23 10.3 74.2 100.0 Total 31 13.9 100.0 Missing System 192 86.1 Total 223 100.0 Table 5.15 is a frequency table of the question Have you seen the commercial? The table shows how many of the respondents that have seen the commercial, which is related to the table 15. To be able to have an opinion about the commercial you have to have seen it. Out of the total amount of respondents (223), 51 were deselected because they did not know of the brand. 23 of the 61 respondents were deselected because they had not seen the commercial. This leaves 149 eligible to answer table 17. Table5. 15 Frequency Have you seen the commercial Frequency Percent Valid Percent Cumulative Percent Valid Yes 149 66.8 86.6 86.6 No 23 10.3 13.4 100.0 Total 172 77.1 100.0 Missing System 51 22.9 Total 223 100.0 Table 5.16 is an ANOVA table based upon the questions What is your opinion about the commercial? and Would you buy the product?. In this test there is a significance of 0.001, which means that there is a high relevance between the two questions. We can see the tendency that if the respondents dislike the commercial they would not buy the product. The mean are spread between 1 and 8, where values 1-4 equals a good opinion about the product and values 5-8 equals a bad opinion about the product. The majority of the respondents would buy the product of Apoliva, though they do not like the commercial. We get a total mean of the respondents who would buy the product of 3.66. The mean shows that the negative opinions about the commercial do not affect the respondents. Furthermore, 65 percent would buy the product because of it being price- worthiness and of good quality. Moreover, the respondents that would not buy the product have a mean of negative opinion (4.475). The results show that the respondents do not care about the image that the commercial projects since they look at other components than the commercial itself. 64 Table5. 18 Frequency Do you know the brand Apoliva Frequency Percent Valid Percent Cumulative Percent Valid Yes 166 74.4 74.4 74.4 No 57 25.6 25.6 100.0 Total 223 100.0 100.0 In table 16, used in hypothesis four, we also looked at the memory of the respondents. We find that out of 172 respondents there are 149 that have viewed the commercial which is 86.6 percent of the respondents who know the brand. This was a necessity for us to ask because our research is about the commercial. The high percentage shows that it is a commercial that people remember. From table 5.19 which is also a frequency table based on the question Have you discussed the Apoliva commercial with your network both online and offline?. The result shows that 49.7 percent do discuss the commercial with their network. Table5. 19Frequency Have you discussed the commercial with your network Have you discussed the commercial with your network Frequency Percent Valid Percent Cumulative Percent Valid Yes 74 33.2 49.7 49.7 No 75 33.6 50.3 100.0 Total 149 66.8 100.0 Missing System 74 33.2 Total 223 100.0 Table 5.20 is an ANOVA table based on the questions Have you discussed the commercial with your network? and What is your opinion about the commercial?. The result show from table 20 a significant relevance (0.000). Since 1 is Yes and 2 is No in the answers, As mentioned before, 1.39 is the average mean. This result shows that if the respondents have a negative impression of the commercial, they tend to be more liable to discuss it. This is a tendency that shows that this particular Apoliva commercial has made an impact and it is not a mainstream commercial. 65 Table5. 20ANOVA Have you discussed the commercial with your network/What is your opinion about the commercial What is your opinion about the commercial? N Mean Std. Deviation Std. Error 95% Confidence Interval for Mean Minimum Maximum Lower Bound Upper Bound Good- Artistic 26 1.65 .485 .095 1.46 1.85 1 2 Good- highlight the product 7 1.71 .488 .184 1.26 2.17 1 2 Good- Relates to the message 11 1.82 .405 .122 1.55 2.09 1 2 Good- Other 18 1.72 .461 .109 1.49 1.95 1 2 Bad- Scary 43 1.19 .394 .060 1.06 1.31 1 2 Bad- No message 31 1.55 .506 .091 1.36 1.73 1 2 Bad- Music not related 5 1.20 .447 .200 .64 1.76 1 2 Bad- Other 8 1.62 .518 .183 1.19 2.06 1 2 Total 149 1.50 .502 .041 1.42 1.58 1 2 To summarize: We find that when the respondents are asked to name a skin-care brand, Apoliva is rarely not mentioned. But when asked about Apoliva, 66.8 percent of all respondents are familiar with the commercial. Out of the respondents who have seen the commercial, 49.7 percent discuss the commercial with their network. Furthermore, we find significant proof that when respondents have a negative opinion about the commercial, they have a tendency to be more open for discussion about it. Though the results show that respondents do not ANOVA Have you discussed the commercial with your network Sum of Squares Df Mean Square F Sig. Between Groups 7.824 7 1.118 5.356 .000 Within Groups 29.425 141 .209 Total 37.248 148 66 primarily recall the brand Apoliva, hypothesis 5 is still approved. Since the results show that the commercial is memorable in the way of commercial ingredients, it is a ―one of a kind‖ commercial, thus, not a mainstream commercial. 5.1.6 Hypothesis 6 Previous positive relations to a product make it difficult to change to a negative opinion. What we find from the frequency table 5.14, used in hypothesis 4, When did you start using the product?, is that 3.6 percent of the respondents began using the product six months ago or less. This percentage shows that the majority of the product users already used the product before the commercial aired. However, the table also shows that only 13.9 percent of all respondents use the product. In the ANOVA table 5.21 we found that when we used the questions, When did you start using the product? in combination with Would you buy the product?, out of the entire population of 31 respondents who are using the product only one would not buy Apoliva again. The ANOVA test shows that there is not a significant relation between the two questions (0.897), but we think that there is a tendency that shows that the respondents who are already users are positive towards purchasing the product again. The majority have a positive relation to the product in form of price worthiness/covers my needs and good quality. 69 significant proof, we find tendencies that a majority of the product users, discuss the product with their network and are long time users which make them less liable to change their opinion. According to us, though we find no significant proof, the hypothesis 6 is approved since there are so few respondents using the product. 5.2 Summary The results from the data found in the questionnaires have been analyzed with help from SPSS. Each hypothesis have been tested and are here presented and summarized. Hypothesis 1 is approved Hypothesis 2 is approved Hypothesis 3 is approved Hypothesis 4 is rejected Hypothesis 5 is approved Hypothesis 6 is approved, though no significant proof was found. 70 6. Conclusions In this chapter we start with a summary of our paper, further we present the findings of our analysis and critical reflections. We also suggest future research within our chosen topic and how this could be used in reality. 6.1 Summary of dissertation From the rise of the first market, when merchants shouted out what was offered, the first viral marketing communication was born. If a product is well worth its price, the rumour is spread fast and viral marketing is created (Ferguson, 2008). To differ from competitors and to stand out from the mass of marketing messages that are shouted out and broadcasted, organisations create brands. The main reason with a brand is to make consumers aware of your product and organisation. Therefore, brand awareness and viral marketing are two significant parts in the science of marketing. Today, commercials are found just about everywhere and messages created to steer consumers’ thinking toward a specific brand or product are broadcasted in different media. However, viral marketing, specifically word-of-mouth, is still one of the most important channels of marketing. One way to use this strategy is through the internet. Since the birth of the internet, consumers in the modern society have become more aware and critical towards products and broadcasted messages. Meanwhile, the consumers enlarge their individual network by engaging in social networks on the internet, for example on Facebook. In these networks, individuals meet and discuss everything from ordinary life to personal opinions, thus, spreading their thoughts around the world. These thoughts and opinions can, through the networks, easily be turned into small campaigns for or against specific subjects by creating specified discussion forums. Through the word-of-mouth, these campaigns become small privately conducted viral marketing campaigns. In order to research the phenomenon of viral marketing campaigns, literature on marketing, brand awareness, brand association and viral marketing was reviewed. This paper is mainly based on theories regarding brand awareness by Keller, 71 (1993), MacDonald & Sharp, (2003), and Hoeffler & Keller, (2006); brand association by Keller, (1993), and Leone et al., (2006), and viral marketing by Dobele et al., (2005), and De Valck et al., (2009). The theories resulted in six hypotheses which were tested with support from SPSS statistical software and either rejected or approved. Statistical data for the hypotheses were gathered through a questionnaire which was delivered on the internet community site Facebook and answered by respondents from our individual network. The results was then analysed in a critical manner and conclusions were drawn. Furthermore, based on the theories, VMC Evaluation Framework Model by Cruz & Fill (2008) and Brand Association Model by Blythe (2008), a model was created. The model will with the aid from the results of the six hypotheses clarify the individual brand association when a respondent is affected by viral marketing through his/her network. 6.2 Conclusions The purpose with this paper was to research the effect network conducted viral marketing has on a consumer’s brand association. Furthermore, if a commercial receives positive or negative attention, the consumer’s behaviour toward the brand might be affected. This was studied by using relevant theories and models related to the subject and then testing them on an internet based network. To be able to investigate the purpose and draw conclusions the following research questions worked as a directive: A: How does a commercial affect consumers to create a network Viral marketing campaign? B: Why do consumers purchase a brand when they are reacting negatively on a commercial? Based on the relevant theories and models, the Brand Association through the Viral Marketing Model (fig. 6.1) was created to be used as a guideline when the hypotheses were set up to test our research question. 74 about the specific brand. Additionally, respondents who are negatively affected by discussions tend to be first time users, since long time users are already familiar with the brand and products. If there is a positive opinion about the commercial, there is a bigger chance of purchasing the product, in comparison to negative opinion. To answer research question A, How does a commercial affect consumers to create a network Viral marketing campaign?: we find that when a commercial creates reactions, it has specific ingredients which make respondents more liable to discuss their opinion with the network and on forum, thus making others curious and interested. These discussions help spreading the brand name and then indirectly create a viral marketing campaign. To answer research question B, Why do consumers purchase a brand when they are reacting negatively on a commercial?: it seems like respondents do buy the product although they do not like the commercial. The results tell us that the respondents do not connect the commercial with the brand itself, but with specific parts of the commercial. That is why, when a consumer have seen a product from the specified brand, he/she relate to and remember the name, but not the commercial, thus choosing this brand before others when purchasing. 6.3 Critical reflections This paper uses the commercial of the brand Apoliva as an example. The use of a commercial for a specific line of products makes it difficult to apply the research to other types of products. Although we tried to enlarge our view so the research would be applicable on other commercials that have received either positive or negative attention, it would be difficult with different types of products to apply our hypotheses on another case. After constructing the questionnaire we realized that it would have been better to have open answers on some of the number questions and also to have investigated more what classifies as many hours spent on the internet. Furthermore, we did find support from Ahuja et al. (2007) that say in their article, that spending 1.5 hours a day on social community sites is heavy internet using. Another critical 75 reflection when we conducted the questionnaire is that we missed to put in an end date to answer the questionnaire for our respondents. 6.4 Future research Future research is necessary in this field when we are combining viral marketing and brand awareness/association. Furthermore, because we have tested our research questions and hypotheses on a commercial that is mostly famous for the negative attention, it would be interesting to do a research on a commercial with a positive attention. Finally, throughout the writing of this dissertation we found that another interesting future research can be to compare what types of commercial that create negative or positive attention in different countries and compare the results of them. 6.5 Contributions 6.5.1 Practical Contributions An understanding the functions of viral marketing in a modern society is the practical contribution from this paper. We have found that even if a commercial has a negative attitude towards it, there are still people that would buy the product, due to good quality or that the product covers the needs. This knowledge is important for marketers to remember if there is a campaign that turns negative, but there are solutions to change the brand association due to previous positive experiences from consumers. 6.5.2 Theoretical Contributions Throughout this paper we choose to combine the three factors viral marketing, brand awareness and brand association. We could not find a model that connected these three in the literature. Therefore, we choose to create a model which, hopefully, can create a bigger understanding for how the different components such as spreading the word, networking and brand association/awareness are put together. 76 References Ahuja, R., Michels, T., Walker, M., Weissbuch, M., (2007), Teen perceptions of disclosure in buzz marketing. Journal of Consumer Marketing, 24/3, pp. 151–159. Anderson, J.C., Narus, J.A., Narayandas, D, (2009), Business market management – Understanding, Creating and Delivering Value, 3 rd ed. New Jersey: Pearson Education. Apoliva (2009) The Apoliva Hygiene products commercial site. [online]. Apoteket AB. Available at: <http://apoliva.se/ [cited 2 November 2009]. Bernard, R. H., (2000), Social Research Methods -Qualitative and Quantitative Approaches. Thousands Oaks: Sage Publications, inc. Blythe, J, (2008), Consumer Behaviour, London: Thomson Learning. Bryman, A. Bell, E., (2007), Business research methods, 2 nd ed. New York: Oxford University Press Inc. Burmann, C., (2009), Towards an identity-based brand equity model, Journal of Business Research 62, pp. 390-397. Creswell, J. W., (2007), Qualitative inquiry & Research Design: Choosing Among Five Approaches. Thousand Oaks: Sage Publications, inc. Cruz, D. Fill, C., (2008), Evaluating viral marketing: isolating the key criteria. Marketing intelligence & Planning, 26 (7), pp. 743-758. De Bruyn, A. Lilien, G., (2008), A multi-stage model of word-of-mouth influence through viral marketing. Intern. J. of Research in Marketing, 25, pp. 151-163. De Valck, K., van Bruggen, G., Wierenga, B., (2009), Virtual Communities: A marketing perspective. Decision support systems, 47, pp. 185-203. Dobele, A. Toleman, D. Beverland, M., (2005), Controlled infection! Spreading the brand message through viral marketing. Business Horizons, 48, pp. 143-149. Ferguson, R., (2008), Word of mouth and viral marketing. Journal of Consumer marketing, vol. 25 (3), pp. 179-182. Gezelius, C. Wildenstam, P., (2007), Marknadsföring – Modeller och Principer, Stockolm: Bonnier Utbildning. Hoeffler, S., Keller, K.L., (2002), Building Brand Equity Through Corporate Societal Marketing, Vol. 21, pp. 78-89. 79 9.Om du tänker på en hudvårdsmärke, vilket märke tänker du först på då? --------------------- 10. Känner du till märket Apoliva? Ja Nej Ifall du svarat nej på denna frågan så tackar vi för din medverkan 11. Vad associerar du främst med Apoliva? Hudvårdsprodukter Reklamen Nordiskt klimat Läskig modell Apoteket Annat 12. Har du sett Apoliva-reklamen? Ja Nej Ifall du svarat nej på denna frågan så tackar vi för din medverkan Ifall du svarade Ja, var någonstans har du sett den då? Tv-reklam Youtube Facebook Flera utav ovanstående 13. Har du tagit ställning emot Apoliva-reklamen i någon grupp på något socialt nätverk (ex. facebook)? Ja Nej 14. Vad anser du om Apoliva-reklamen och vilken är den främsta anledningen? Bra – Konstnärlig Dålig - Läskig Bra – Framhäver produkten Dålig – Inget budskap Bra – Bra relation till budskapet Dålig – Musiken ej passande Bra – Annat ________________ Dålig – Annat _________________ 15. Har Apoliva-reklamen ändrat din uppfattning om produkten? Ja Nej 80 Ifall du svarade Ja, har din uppfattning blivit ändrad till: Det bättre Det sämre 16. Har du diskuterat Apoliva-reklamen med din omgivning på och utanför Internet? Ja Nej 17. Har du köpt/använder du Apoliva? Ja Nej Ifall du svarade Nej, gå vidare till fråga 19. 18. Ifall du använder Apoliva produkter, när började du använda dem? För 6 månader och mindre sen För 7 och fler månader sen 19. Skulle du kunna tänka dig köpa Apoliva produkter och till vilken anledning? Ja – Det är prisvärt/täcker mina behov Nej – För dyrt Ja – Bra kvalitet Nej – Dålig kvalitet Ja – Läcker reklam Nej – Gillar inte reklamen Ja – Annat__________________ Nej – Annat _________________ Vi tackar dig för din medverkan och ber dig skicka tillbaka dina svar snarast. Ida och Magnus 81 Appendix 1b Questionnaire English The answers to the survey is confidential 1. Are you a woman or a man? Woman Man 2. How old are you? 15-20 21-25 26-30 31-35 36 or older 3. What is your highest finished education? Primary school Senior High School Work preparation education Community college University 4. In what business do you work? Communal County Industry Selfemployed Commerce Hotel/ restaurant Bank/finans/Insurance Unemployed None, is student 5. How many hours do you usually spend actively on Facebook or similar social communities network a week? >3 hrs 3-6 hrs 7-11 hrs 12 hrs or more 6. Do you usually discuss actively in forums or social networks (for example Facebook or similar) on the internet? Yes No 7. Do you usually listen to advice by your surroundings before purchasing skin care products? Yes No 8. Could you buy a skin care product when the commercial is not appealing to you? Yes No