A Comparative Study on the Effects of Social Media
Strategies on Customer Brand Engagement between
MNEs and SMEs.
Master Thesis
Michiel A. Smitskamp Student Number: 10468269 March 15th 2015 – Final draft
MSc. Business Administration - Marketing track University of Amsterdam
2
Statement of Originality
This document is written by Michiel A. Smitskamp who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Abstract
The western world has been overwhelmed by the large amount of social media available today, it has changed the way marketers go about their every-day activities. This research paper makes an attempt at establishing the effectiveness of social media strategies, through the use of viral social media marketing campaigns. A comparison was made between the effective use of these social media campaigns between MNEs and SMEs to investigate whether the size of a company would have a significant effect on customer brand
engagement. The results of these tests were based on customer brand engagement, which used four building blocks to define its purpose. These building blocks include brand
experience, commitment, loyalty and satisfaction. Accordingly, variables such as willingness to act and times spent on social media were measured to identify their influence on customer brand engagement. The experiment consisted of four different social media campaigns, which form the basis of the data collection. From the four social media campaigns, each was
labelled depending on their orientation as well as their strategic intent, in order to investigate whether or not a distinction can be made between the various types of campaigns. Data was collected through an online-survey, specifically focused on Facebook users. The results of the experiment showed the importance of willingness to act on customer brand engagement through willingness to share and willingness to spend. Additionally, the results proved that the orientation of a social media campaign had a significant impact on customer brand engagement, as a positive campaign would show a positive effect on customer brand
engagement. However, the strategic intent of the campaign and the size of the company had no significant influence on the building blocks of customer brand engagement. This research paper concludes with theoretical and managerial implications as well as limitations and various ways to improve further research.
4
Table of Contents
Abstract ... 3
1. Introduction ... 6
2. Literature Review... 7
2.1 Customer Brand Engagement ... 8
2.1.1 Willingness to Act... 9
2.2 The Social Media Framework ... 11
2.3.1 Viral Marketing Campaigns ... 14
2.3 MNEs vs. SMEs: Implications for companies towards Social Media ... 19
2.3.1 Social Media implications for Large Multinationals ... 19
2.3.2 Implications of Social Media for SMEs... 20
2.4 Research Gap... 22 2.5 Conceptual Model ... 23 3 Methodology ... 23 3.1 Procedure ... 24 3.2 Manipulation Test ... 25 3.3 Pre-Test ... 26 3.4 Sample ... 27 3.5 Measures... 27 3.5.1 Dependent Variable ... 28 3.5.2 Control Variables ... 29 3.5.3 Independent Variable ... 30 3.5.4 Moderator ... 32 3.6 Data Analysis ... 32 4 Results ... 34 4.1 Manipulation Check ... 35 4.2 Correlations ... 38 4.3 Hypothesis testing ... 40
4.3.1 Impact of willingness to act on Customer brand Engagement ... 40
4.3.2 Impact of viral marketing campaigns on Customer brand Engagement ... 42
4.3.3 The effect of company size on customer brand engagement ... 50
5 Discussion ... 56
5.1 Results ... 56
5.2 Theory Implications ... 62
5.3 Managerial Implications ... 64
5.4 Limitations and Future Research... 65
5
7 References ... 68
8 Appendix ... 72
8.1 Tables ... 72
6
1. Introduction
In the past decade, the western world has been overwhelmed by the arrival of online social communities, social media platforms and the ability for an individual to stay connected twenty-four a day. The social media hype, has swept through the Western world, influencing the way marketers go about their daily activities. Social media, an online tool which allows people to create, share and exchange information, has influenced the tools of the trade such as the strategies companies instigate to communicate with their consumer (Mangold, 2009) The average consumer is bombarded by slogans, campaigns, brands and companies on a daily basis and social media has progressed to incorporate these brands onto their social media platforms. Examples of social media platforms which companies use to try and interact with their customers include Facebook, Twitter and LinkedIn.
Social media has made it possible for companies of all sizes to reach out to consumers all over world. The increased influence of social media, on the consumer, has changed the way marketers think about their online marketing strategies. The ability to engage a consumer has become a daily task in social media marketing, as the consumer is constantly seeking for better quality products online (Brodie, 2011).
Customer engagement is defined as a cognitive, emotional and behavioral connection between a consumer and a brand or company. Customer brand engagement will be the focal point of this research paper as we define customer brand engagement by its four building blocks; brand experience, commitment, loyalty and satisfaction (Hollebeek, 2011). These building blocks will be used to consider the effectiveness of current online marketing
strategies through the use of a company’s social media campaign. A comparison will be made between Multinational enterprises (MNEs) and small and medium enterprises (SMEs), to
7
establish the relationship between effectiveness of social media strategies and the size of a company.
The motivation for this research paper was a comparative research paper to investigate the effect a social media strategies has on customer brand engagement and to specifically look at the difference of this effect between MNEs and SMEs. Previous research has shown the influence of size on online marketing strategies, specifically SMEs not participating
effectively due to various barriers compared to their MNE competitors (Michaleidou, 2011). However, a comparative study between companies of different sizes and the effect of size on customer brand engagement was missing in this particular field of research. This research paper will assess the effective use of various social media strategies and present a comparison between MNEs and SMEs specifically focusing on the four building blocks of customer brand engagement.
2. Literature Review
Consumer-to-consumer communication has increased significantly in the past few years with the introduction of social media. A single entity is no longer connected to another single entity but possibly hundreds or thousands at a time. The rise of social media has changed the way marketers’ use marketing strategies, as the impact of consumer-to-consumer
communication has increased the marketplace (Mangold, 2009). To fully understand the effectiveness of social media strategies and its relationship with Customer brand engagement, the following section of the paper will introduce those topics influential to this research paper. These topics provide a brief introduction into customer brand engagement, including the building blocks of customer brand engagement. While additionally the various types of social media strategies are identified and how they are set up through a social media
8
strategies for MNEs and SMEs. The literature review will conclude with a research gap and a conceptual model.
2.1 Customer Brand Engagement
Previous research has found that a clear definition has yet to be identified for customer brand engagement. Although a clear definition was never formed, research has found that
researchers have formed their own specific definition of customer brand engagement considering their own particular type of research. Hollebeek (2011) defines customer brand engagement as an engagement process through a two-way interaction between object and subject. The specific definition for her research deals with the level of an individual’s state of mind, characterized by specific levels of cognitive, emotional and behavioral activities with regard to brand interactions. Patterson et al. (as cited in Hollebeek, 2011) expands on this definition by adding that customers behavioral brand related activity may be expressed through a consumers level of energy exerted in interacting with a brand, known as willingness to act. Brodie et al (2011) establishes his definition for customer brand
engagement for his research within a virtual brand community, which is based upon his five themes review (Brodie, 2011) of customer engagement. The definition of customer brand engagement used for this research paper will be based on the definition introduced by Hollebeek (2011) and Bowden (2009) and structured into a found definition by Brodie (2011). Considering this, consumer brand engagement is defined as a multidimensional concept compromising of cognitive, emotional and behavior dimensions, within a virtual brand community, which involves a specific interactive experience between consumer and the company/brand (Brodie et al, 2011).
The definition of customer brand engagement identifies the most important variables considering customer brand engagement. These variable building blocks of customer brand engagement enhance the level of customer interaction with a brand or company. Brodie
9
(2011) explains that customer brand engagement is directly related to a number of brand relationship outcomes; brand experience, commitment, loyalty and satisfaction. These four outcomes will be classified as the building blocks of customer brand engagement. Hollebeek (2011) defines these four building blocks in her customer brand engagement matrix. Brand experience is defined as a subjective, internal consumer response, either through feelings, sensations and cognitive reaction and behavioral response. These responses are induced by brand-related stimuli such as packaging, identity, communications and environment. Commitment is defined as the valuation of an ongoing relationship and the willingness to maintain a relationship with the brand, which may be amplified by a strategic implementation of marketers. Loyalty was defined as a trusted brand, which entails repeated purchases
prompted by strong disposition over a period of time. Satisfaction, finally, was defined as a customer’s overall evaluation of the performance of a brand to date. Furthermore, Gummerus (2012) states that loyalty may be seen as the highest form of customer engagement, there are certain levels of behavioral, which are not characteristics of a high degree of loyalty. One behavioral aspect which will be crucial in this research paper is the consumers’ willingness to act, which will be explained in the following section of this paper.
2.1.1 Willingness to Act
In order for customer brand engagement to be initiated, a consumer must have a certain willingness to act. Willingness to act is based on the instinct of the consumer, to initiate a relationship with a brand or company. In the research conducted by Hollebeek (2011), she established three key customer brand engagement themes based upon her definition of customer brand engagement in which willingness to act is necessary. These three key factors include, immersion (cognitive), emotional (passion) and activation (behavior). The first key factor is immersion, also known as involvement, which deals with the customer’s level of brand-related concentration in particular brand interactions. The second factor is passion,
10
described as the degree of a customer’s positive brand related effect in particular brand interactions such as the notion of being proud whilst being associated with a brand. The final key factor was activation, the informant’s behavioral willingness to act, was defined as a customer’s level of energy, effort and/or time spent on a brand in specific interactions. (Hollebeek, 2011). These three factors can be related to the first building block of customer brand engagement, namely brand experience, as brand experience is defined as the behavioral part of customer brand engagement.
Willingness to act consists of two factors in this research paper. These factors will be willingness to spend and willingness to share. First, willingness to spend is defined as the intention of purchase a consumer has toward a specific brand, which can be understood as a behavioral willingness to act and is essential for companies, as this should lead to sales. Purchase intention, may be defined as the phase in which the consumers interest in the brand deals with the possibility of buying the branded product. Kim et al. (2011), describe a
purchasing decision as a function of emotional (passion), functional (behavior) and social values (cognitive), which he refers to as the customer value framework. Kim et al. (2012), relates that measuring purchase intention assumes that the willingness to spend of the
consumer is directly responsible for further contributions towards a brand, which is consistent with the definition of loyalty in Hollebeek (2011) dimensions of customer brand engagement.
Secondly, research by Hollebeek (2011), Kim (2011) and Kim (2012) introduce the
importance of willingness to share. The activity and intention of a consumer as a key factor of willingness to act upon towards a brand or company. Taking into account the research done concerning brand experience and loyalty, the first hypothesis of this research paper can be established. The first hypothesis of this research paper is drawn from the research
11
willingness to act, in this research paper, consist of willingness to share and willingness to spend. Accordingly the research about customer brand engagement should entail that there is a positive relationship between willingness to act and the two building blocks of customer brand engagement, brand experience and loyalty, based on the behavioral aspects of a consumer’s willingness to act.
Hypothesis 1: Willingness to Act has a direct positive relationship to customer brand engagement
Hypothesis 1a: Willingness to share has a direct positive relationship to brand experience
Hypothesis 1b: Willingness to spend has a direct positive relationship to loyalty
2.2 The Social Media Framework
In the last few years social media has become part of daily life in western society. However, the research on social media quickly is lagging behind and outdated due to the continuous updates performed on the social media platform. Previous research has not given a clear answer on how to effectively deal with social media yet, however some attempts have been made, Huberman (2008), Kaplan (2010) and Mangold (2009). Trusov (2008, and 2009) and Kozinets (2010) help establish specific areas of social media such as how to communicate effectively online or how to deal with word of mouth on social media. Although such research gives insight for marketers on how to deal with social media, it lacks to inform effectively about leveraging social media, specifically on Facebook. Due to the nature of this research paper, first a basic framework of a social media platform will be established before it is possible to show what is important for companies presenting their social media campaigns which will be measured towards customer brand engagement.
12
Kietzmann et al. (2011) set up a social media framework in which seven functional building blocks of a social media platform are explained. These building blocks for social media are essential in defining selected social media campaigns, which will be introduced in the following section of the paper. The seven building blocks of Kietzmann’s social media framework consist of: Sharing, groups, conversation, identity, presence, relationships and reputation groups. Kietzmann mentions that these building blocks do not naturally have to be included in all social media strategies but form a basis to help understand the various levels of social media. Due to the type of social media strategies used, this paper will concentrate on those building blocks which relate to the elements of customer brand engagement, namely sharing, groups, conversation and identity.
The first dimension in Kietzmann’s framework discussed is sharing. As mentioned in the previous section of the paper, the willingness to act and the willingness to share are essential for online marketing strategies as this relates directly to brand experience through the
behavioral aspect of customer brand engagement. Kietzmann mentions that sharing can be anything ranging from videos, pictures, recommendations and complaints about a company or a campaign. Additionally his research mentions that consumers must have a similar interest in the type of product/service that a company is known for, since without this initial interest the probability of sharing is very low. The second dimension is groups, which are defined as the extent to which the users are able to join a particular network. As this online network expands, so will the possibility of its users sharing items within the online
community. As social media platforms become flooded with excess information, these online communities function as a filter to brand-specific posts (Kietzmann, 2011).
The third dimension of Kietzmann’s framework is conversation. Conversation deals with the extent to which users communicate within their community. A company has the distinct
13
ability to initiate communication, which if done correctly could lead to an increase in the brand experience of a consumer, possibly leading to increased brand loyalty. A company’s campaign going viral is initiated by the conversation dimension, as an online consumer will like, discuss or share the campaign with their online community. The final dimension that relates directly to this research paper is the identity of the consumer. Identity deals with the overall image of a user on social media, ranging from name, age, gender, ethnicity or geographical location. The identity dimension could result in problems for companies as consumers are being more concerned about their privacy and what companies are doing with their personal details. Women have become more pro-active and seem to protect their
personal information better than males (Grubbs-Hoy, 2010). Due to this gender difference, it is essential for a company to create a balance between sharing identities and privacy of users since this balance is crucial when trying to achieve a successful social media strategy.
The other three building blocks of Kietzmann’s social media framework include relationships, presence and reputation. Relationships deal with the connection between various users on a social media platform; this can range from a formal to an informal relationship. The building block of presence is based on the idea that the company is
available on social media platforms as well as real life. The final building block is reputation, for which Kietzmann mentions that the problem with reputation lies not by the popularity of a brand or user, but by how many other users actually read the posts/videos that are placed on social media. Taking into consideration Kietzmann’s framework for social media platforms, the basis for online marketing campaigns is identified. This basis is essential when finding the difference between the effective uses of social media marketing strategies focusing
specifically on customer brand engagement. The following section of the paper will introduce the different types of online marketing campaigns which apply to this research paper.
14
2.3.1 Viral Marketing Campaigns
Online marketing campaigns that go viral are instigated by a company or an individual, by uploading or sharing a campaign, image or video with their online community, with the intent for others to share within their own respective online communities. Viral social media
marketing occurs when influential social media users share a campaign through social media at a rapid pace. Kaplan (2010) initiates in his research that a viral marketing campaign needs three basic conditions; the right environment, a good message and the correct people to act as influential messengers. Viral marketing can be a very effective marketing tool and lead to positive results for a brand or company. However, viral marketing can also lead to a negative effect towards a brand or a company if the campaign is negatively oriented towards the brand, through a bad campaign or negative recommendations. Kaplan et al (2011) has attempted to introduce a new kind of framework in which four types of viral online marketing campaigns are identified. These four types consist of negative and positive outcomes of social media marketing. These campaigns were named nightmare campaigns, strokes-of-luck campaigns, home-made issues campaigns and triumph campaigns. Table 1 shows the framework
introduced by Kaplan (2011) and also introduces the aspect of a strategic campaign versus a non-strategic campaign, which will be clarified in the following section of this research paper.
Table 1: Viral Marketing campaigns organized through strategy and orientation of campaign. Positively Oriented Campaign Negatively oriented Campaign Strategic planned campaign
Triumph campaign Nightmare campaign
Non-strategically planned campaign
15
2.3.1.1 Positively oriented viral marketing campaigns
Kaplan (2011) initially introduces the campaigns with either a positive or negative
association, from the perspective of the company. The triumph and stroke-of-luck campaigns are perceived as positive campaigns and the nightmare and home-made issues campaigns are seen as negatively oriented campaigns. Additionally, this research paper labels those
campaigns being either a strategically oriented campaign or a non-strategically planned campaign. The non-strategic campaigns entail that a marketers did not assume the campaign to go viral, either in a positive or negative manner, as a successful manager would never purposely create a negatively orientated campaign for their company. Therefore these campaigns can also be seen as lucky or as unlucky, as opposed to a non-strategic campaign from the point of view of a marketer. Kaplan’s (2010) earlier research shows the importance of a strategic approach for brand on a social media platform. Specifically choosing a social media platform, ensuring activity alignment and a social media integration plan are vital in the survival of a brand on social media. Additionally, a brand must be active, interesting, humble, honest and must include a hint of unprofessionalism to show that a brand, like their consumer, is not always perfect (Kaplan, 2010). Research has shown that a strategically oriented brand, which includes the right elements in their social media strategies, will result in a healthy relationship with the consumer, which brings forth the second hypothesis of this research paper.
Hypothesis 2: Strategically oriented marketing campaigns will have a direct positive effect on customer brand engagement
Having dealt with the strategic aspect of the campaign, the first type of campaign as perceived in Kaplan’s research of viral marketing is the triumph campaign. A triumph campaign is a positively oriented social media campaign in which a company is capable of creating its own viral success through a strategic marketing campaign. A triumph campaign
16
distinguishes itself as if it was a strategic move of the marketers to make that specific campaign go viral. A viral marketing campaign is successful when consumers, whom might have no affiliation with the company, are willing to share the campaign within their Facebook community, through groups and conversations. The second positively oriented type of viral marketing is the stroke-of-luck campaign. The stroke-of-luck campaign is positively oriented and distinguishes itself as being created through positive reactions from consumers. The managers at the company were unaware of the consequences of their campaign, when they presented their initial online marketing strategies. Therefore the stroke-of-luck campaign is known a non-strategically planned campaign. The Stroke-of-Luck campaigns increases brand awareness as consumers, with no possible association to the brand, will share the campaign due to the positive perspective of a campaign and remember the brand in a later stadium due to the positive nature of the campaign. Both the triumph and the stroke-of-luck campaign are characterized by the consumer being positively effected by the campaign and sharing the campaigns with the intention to spread only the positive aspects of the campaign.
As understood from the building blocks of customer brand engagement and commitment, commitment deals with an on-going relationship with a brand, which can be amplified by a strategic marketing campaign. Where brand experience, deals with a subjective internal consumer response. Taking into consideration the various types of positive viral campaigns, the third hypothesis of this research paper can be established.
Hypothesis 3: Positive marketing campaigns will have a direct effect on customer brand engagement.
Hypothesis 3a: A positively oriented strategic campaign will have a positive effect on customer brand engagement, specifically on brand experience.
17
Hypothesis 3b: The positive non-strategic campaign will have a positive effect on customer brand engagement, specifically on commitment.
2.3.1.2 Negatively oriented viral marketing campaigns
Brodie’s research (2011) reviews virtual brand communities and states that conversations and sharing occur at almost no cost and are able to spread rapidly, possibly go viral, through both online and offline channels. He continues to mention that negative reviews on social platforms tend to have a larger negative effect on sales than the positive reviews. Hennig-Thurau et al. (2004) reports eight factors which motivates consumers to make contributions to a virtual community, including venting negative feelings, concerns for other consumers, self-enhancement, advice-seeking, social benefit, economic benefit, platform assistance and helping the company.
Research by Barbara (2012) and Wilson (2012) mentions the increased power of consumers on social media. These consumers are empowered to behave worse on an online platform. This negative behavior is due to no feeling of guilt, the mob of other users to back them up, relative anonymity and no accountability for their actions on such platforms. These
comments can create a snowball effect, through conversations and sharing, in which more and more consumers will be aware of the campaign as the mob expresses their complaints about the company’s campaign. A negatively oriented viral marketing campaign, whether strategically intended to go viral or not, is not initiated by the marketers as a negative campaign, but by the consumers, reacting towards these campaigns.
Kaplan’s types of viral marketing also include two negatively oriented viral marketing
campaigns, a nightmare campaign and a home-made issues campaign. A nightmare campaign is a viral marketing campaign, initiated by consumers posting or sharing negative
campaigns/comments about the company on social media platforms. The nightmare campaign can be related to the triumph campaign if the marketers of the brand had the
18
strategic intent for the campaign to go viral. If the marketers had the intention of the campaign going viral positively, the nightmare campaign is such a campaign in which the negative reactions of the consumers lead to the campaign going viral negatively. Since, no marketer would ever set up a negative campaign towards their own brand, a nightmare campaign occurs through either errors in the campaign or a generally poorly constructed campaign.
The final type of viral marketing is the negatively oriented home-made issues type. The home-made issues campaign distinguishes itself, due to the fact that the company themselves are the instigator of negative viral marketing. Kaplan uses the case study of Sony, in which they made up a fake blog, through which they tried to convince its target audience to buy the new PSP. However, the blog went viral as consumers found out the blog was fake and created by Sony, whom initially denied any association with the blog, which led to additional
negative viral marketing until Sony finally admitted to their mistake a few weeks later. (Kaplan, 2011). A home-made issues campaign has the non-strategic label attached to the campaign as the marketer has attempted to use current events or a debatable occurrence to try and increase brand awareness which negatively projects the brand in the eye of the consumer.
The building block of customer brand engagement has shown that satisfaction deals with the overall interpretation of a brand and loyalty deals with trusting a brand to ensure repeated purchases. A non-strategically planned campaign could interfere with the trust a consumer has towards a brand as the might seem incompetent, whereas a strategically planned campaign will effect the overall satisfaction of the brand. Through the research of Brodie (2011), Barbara (2012) and Wilson (2012) regarding negatively oriented viral marketing campaigns, the fourth hypothesis of this research paper can be proposed.
19
Hypothesis 4: Negative marketing campaigns will have a direct effect on customer brand engagement.
Hypothesis 4a: A negatively oriented strategic campaign will have a negative effect on customer brand engagement, specifically satisfaction
Hypothesis 4b: A non-strategically planned campaign will have a negative effect on customer brand engagement, specifically on loyalty.
2.3 MNEs vs. SMEs: Implications for companies towards Social Media
The final section of the literature review will introduce the size of a company and the effects size has on the social media marketing strategies, the implications on MNEs and SMEs with regard to the various types of social media strategies presented earlier.
2.3.1 Social Media implications for Large Multinationals
Most research conducted with regard to social media deals with large companies and the opportunities that social media platforms offer these companies to extend their brands into the lives of their consumers. In the past, companies have been able to carefully communicate what they wanted to communicate to the outside world about their company through public relations managers and perfectly scripted press releases, however with the arrival of the social media platforms, information is shared by consumers is faster and more publically visible leading to companies losing control about what is available for their consumers to read (Kaplan, 2010). For marketers of large multinational companies, the marketing game has changed dramatically since the beginning of the 21st century. The introduction of social media has brought forth both challenges as well as opportunities for marketers. However, through social media platforms such as Facebook, the marketing power has shifted from the marketer to the consumer since they have gained the ability to share knowledge and
20
experiences about a certain brand or company within their own social community (Berthon, 2012).
Xiaoyun (2014) suggests that social media platforms over time can lead to higher sales performance as well as increase the online engagement with consumers. The online
engagement with consumers, takes time to be set up but Xiaoyun (2014) has shown that there is a positive correlation between positive interaction with the online consumers and the size of the online community. Although these correlations might be true for smaller firms as well, the following section of the literature review will interpret the implications of social media for small and medium enterprises (SMEs) as most SMEs are low resources and IT
capabilities for large online marketing campaigns (Xiaoyun, 2014).
2.3.2 Implications of Social Media for SMEs
Taking into account the knowledge gathered about social media usage for large
multinationals, the following section will take a closer look at those small and medium sized companies and investigate the implications when using a social media platform. Research by Spence (2003) shows that SMEs, although small in size individually, as a whole make up for a large percentage of a countries economy. In an offline scenario, an SME is more likely to have an influence on the local community as opposed to a national market level. Perceived barriers such as lack of money, time and training, negative views and unfamiliarity are consistent for all SMEs (Michaleidou, 2011). However, through the use of a social media platform, a SME is able to extend its community through its online members, creating a buzz about their organization (Spence, 2003).
The arrival of social media platforms offers SMEs a platform in which they can promote their product or brand through an easily accessible medium. Durkin’s (2013) findings about SMEs in Ireland show a negative affiliation with social media strategies. His findings contribute to employees having a lack of motivation or an incorrect agenda through which social media
21
strategies could be successfully implemented to enhance the customer brand engagement. The employees of SMEs in Ireland were aware of the possibilities of Facebook as an Internet marketing tool; however the adaptation of a proper marketing strategy was not properly implemented by the managers of the SMEs (Michaleidou, 2011). The social media strategy path for SMEs is complex, as a small business will need to remain persistent in their online presence to create an active online community, willing to act upon their online campaigns. An SME is able to increase customer brand engagement by the use of social media
marketing, if persistence and engagement from all employees involved in the social media strategy planning process are adequate (Xiaoyun, 2014).
Nevertheless, in theory, the online possibilities for both types of companies are the same, with regard to their social media strategies. Due to the nature of an SME being active offline mostly in their own local community as opposed to globally operating MNEs, the ability to create a larger online community is an advantage for MNEs. Given the nature of the types of companies as well as the four viral online marketing strategies, including strategic input and orientation of the campaign, the introduction of the fifth and final hypothesis of this research paper is possible.
Hypothesis 5: The effect on customer brand engagement on a social media marketing campaign has a direct relationship with the size of a company.
Hypothesis 5a: A positively oriented marketing campaign for an MNE will result in a more positive effect on customer brand engagement, then a positively oriented marketing campaign for an SME.
Hypothesis 5b: A strategically oriented marketing campaign for an MNE will result in a more positive effect on customer brand engagement then, a strategically oriented marketing campaign for an SME.
22
2.4 Research Gap
The social platform has greatly expanded over the last few years and many companies are trying to catch up or keep up with the altering marketing possibilities of the use of social network platforms. How to effectively reach the consumer and through which medium this can be done, is a daily question for all marketers. This research paper seeks to investigate the differences between the effectiveness of social media strategies for SMEs and MNEs,
specifically for Facebook. Using the customer brand engagement dimensions created by Hollebeek (2011) and Brodie (2011) and Kietzmann (2011) social media framework to help define the viral online marketing definitions of Kaplan (2011).
This research paper will be a comparative research paper to help identify the differences for SMEs and MNEs with regard to their social media strategies and how this effects customer brand engagement. Current research focusses primarily on large companies while looking closely at the positive effects of social media while the research on SMEs and their effects of social media is lacking in size. Similarly, research comparing the differences of MNEs and SMEs is also incredibly small. SMEs, considered by Michaleidou (2011), are a driving force of economic growth; deal with additional complexities as lack of money, motivation and training.
Brand experience, satisfaction, commitment and loyalty are the essential themes of customer brand engagement; these key terms will be investigated as various viral social media
marketing campaigns are presented. This paper investigates the gap that lies between
orientation and strategic input of marketers of social media campaigns introduced by Kaplan (2011) and how this effects customer brand engagement of SMEs compared to MNEs. Specifically,
23
How does the effective use of social media strategies effect customer brand engagement and how do these effects compare between Multinational Enterprises (MNEs) and Small and medium sized enterprises (SMEs).
2.5 Conceptual Model
The following figure shows the conceptual model of this research paper.
3 Methodology
Easy access to social media, through the use of smart-phones and tablets, has established an easier way for the consumer to communicate within their own network. The fact that people wish to remain internationally connected
with their friends and relatives, ensures the pro-longed power of social media. The social media platform discussed in this paper is Facebook. As Facebook has developed from a Harvard-only platform to an international social media
platform, companies were given the
Willingness to Act Company Size Customer brand engagement Orientation of viral social media campaigns -Positive/Negative Strategic implementation of viral social media campaigns - (non) Strategic Social Media Framework
24
ability to promote their business through the use of Facebook. Facebook has become an attractive medium as businesses are now permitted to develop their own “apps”
(applications), that could link their Facebook page to their own website, but also lead to online shopping behavior through Facebook (Dekay, 2012). Barbara (2012) mentions that in the world of “new marketing” every company has become a media company, in which every post, twitter feed or blog is an opportunity to reach the consumer in a fast and effective manner. A small glimpse into the context of a company’s ability to implement Facebook successfully as a way to market their products effectively online. This research paper will use Facebook as the only type of social media platform. The following section of the paper will introduce the procedure of the experiment, including the manipulation tests, the pre-tests as well as the sample and the variables used during the experiment.
3.1 Procedure
This experiment consisted of a factorial experiment in which a 2x2 factorial design was implemented. Through this method, each factor is able to be tested on the response variable as well as their individual interactions. Due to the nature of the setup of this experiment, an online survey, named Qualtrics, was used to gather the data. Through a survey each
dimension of the factorial design can be collected as the respondents are separated into two groups. Firstly, respondents were assigned to a group, the first group was assigned to the MNE section of the survey whereas the other group was assigned the SME section. Secondly, each viral social media marketing campaign was assigned to a respondent randomly. A quota was set on the amount of respondents per social media campaign as well as the amount of respondents per company size, to ensure an equal sample size for all factors.
The data will consists primarily of quantitative data. Respondents will randomly be placed in either an MNE or SME scenario. The survey will inform the respondents about the type of
25
firm they are dealing with, either an MNE or an SME. Next, respondents will be asked to react to a specific marketing campaign followed by their opinions about the marketing campaign. As mentioned before, the marketing campaign will also be randomly selected by Qualtrics. Finally, a few basic questions about the respondent’s social media usage and their general demographics will be asked. The survey will be completely anonymous and there will be no incentive provided to those that do fill in the survey. The survey will include a
manipulation check, to investigate whether or not the respondent fully understood the
scenario described to them. Additionally, a realistic test was included to identify how realistic the campaigns and the company were in their opinion.
The respondent received a link to this online survey electronically; by use of a social platform such as Facebook to those people connected to the network of the author. Social Media was the only method of spreading the link, to make sure no respondents participated that were not using social media. The author of the paper requested respondents, specifically those with a large preferably international network, to share the survey link so that Snowballing might be initiated as respondents repost the survey enlarging the target audience and ensuring that convenience sampling was not the only method of sampling.
3.2 Manipulation Test
The purpose of a manipulation check was to check if the selected scenarios of this research survey were realistic and that the respondent had a clear understanding of the type of scenario that was presented. The respondent was asked whether they believed the campaign as well as the company, Active Apes, to be of a realistic nature. Secondly, the manipulation check was implemented into the survey. At the end of each marketing scenario, a binary question informed about the scenario to see whether or not the purpose of the scenario was clear. The question conditions mentioned in this question included contradicting statements such as
26
whether or not they believed the scenario to be planned or unplanned, a success or failure, lucky or unlucky. The respondents had to select a minimum of one answer and a total of eight answers; this was to ensure the consumer truly understood which keywords were of
importance. An equation was created to calculate the outcome of the manipulation test. If the amount of wrong labels was greater than the amount of correct labels, they would fail the manipulation test.
The survey was manipulated according to both campaigns and size of the company. The respondents would randomly be assigned either an MNE or an SME at the start of their survey. Keeping all factors equal, only the amount of employees and the location of their HQ were manipulated. Next, the respondent would randomly be assigned one of the four viral marketing campaigns. Each campaign would have a similar layout, including a Facebook logo and the appropriate like and share buttons. Through this manipulation, a single
respondent was only granted one specific scenario, out of eight possibilities, four campaigns and two companies.
3.3 Pre-Test
Before the survey was launched a pre-test was conducted among 19 randomly selected respondents (both male and female). The pre-test was set up to ensure the survey would suffice to the necessary requirements needed to account for the collected data. The
manipulation check, realistic check, willingness to act, online activities and customer brand engagement variables were all recorded as being on a reliable scale.
The manipulation check was also tested during the pre-test to identify whether or not the company and marketing scenarios were accurate and realistic. The second part of the manipulation check, with regard to the type of marketing scenario showed that the
27
whether or not the scenario was a success of a failure and whether or not the company was responsible for the success/failure.
3.4 Sample
The survey, designed for the average every-day consumer, should have had at least a minimum sample size of 200 participants and had no maximum number of respondents. However, a threshold of 240 participants was desired as this would entail each online viral marketing scenario to be completed by 30 different individuals. In order to gain a large response rate, the survey was short and concise to keep the focus of the respondents. There were no set demographics for the respondent as each individual is a consumer of the market. The minimum therefore represents the smallest desired sample size to make assumptions about an entire online-shopping community. All respondents were supposed to be active on social media as the link to the survey was sent through Facebook posts and messages only, to ensure non-Facebook users would effect the results.
3.5 Measures
In this section of the paper, the various variables used in the experiment are introduced and defined. Initially the four distinct pillars of customer brand engagement; consisting of brand experience, commitment, loyalty and satisfaction. In order to remain consistent among all dependent variables, all were based on the methods introduced by Hollebeek (2011) and Brodie (2010) on the key themes of customer brand engagement. Additional variables, including the control variables, independent variables and the moderator will be clarified in the following section to help understand the various measurements used to develop the proper research.
28
3.5.1 Dependent Variable
Customer brand engagement: (4 items, α: .88) Customer brand engagement was assessed through the definition set forth by Brodie (2010). Brodie defines Customer brand engagement as a specific multi-dimensional interactive experience between consumers and the brand including immersion, passion and activation. Customer brand engagement as a dependent variable will consist of its four building blocks specifically selected for this research, namely brand experience, commitment, loyalty and satisfaction. In order to understand the variable in a better manner, the following section will define the sub-variables responsible for our
dependent variable customer brand engagement.
Brand Experience: (4 items, α: .87) Brand Experience is the first building block of customer brand engagement and was assessed Hollebeek (2011) by means of their Customer brand engagement matrix. Hollebeek defines brand experience as a subjective internal consumer response, triggered by brand related stimuli. An example of brand experience within the survey would be their likelihood to join a loyalty program. The items responsible for this variable were rated on a 5 point Likert scale ranging from 1 (Very Unlikely) to 5 (Very Likely).
Commitment: (2 items, α: .65) Commitment is the second building block of customer brand engagement and was assessed Hollebeek (2011) by means of their Customer brand
engagement matrix. Hollebeek defines commitment asvaluing an ongoing relationship with a specific other party so as to warrant maximum efforts at maintaining it, willing to maintain a relationship. An example of commitment would be their likelihood to visit the company’s websites. The items responsible for this variable were rated on a 5 point Likert scale ranging from 1 (Very Unlikely) to 5 (Very Likely).
29
Loyalty: (4 items, α: .9) Loyalty is the third building block of customer brand engagement used during this research paper. Loyalty was assed and defined by Hollebeek (2011) as repeated purchases prompted by a strong internal disposition over a period of time. An example of Loyalty would be group interaction about the campaign. The items for loyalty were rated on 5-point Likert scale ranging from 1 (Very Unlikely) to 5 (Very likely).
Satisfaction: (4 items, α: .87) Satisfaction is the final building block of customer brand engagement throughout this research paper. Satisfaction was assessed and defined by Hollebeek (2011) as a customer’s overall evaluation of the performance of an offering to date. An example of satisfaction would be their feelings towards the campaign. The items for satisfaction were measured using a 5-point Likert scale.
The four variables, brand experience, commitment, loyalty and satisfaction will be analyzed individually before they are combined to represent the final variable of customer brand engagement, in order to compare all marketing scenarios properly using the factorial dimensions.
3.5.2 Control Variables
Online activity: In order to entail that the respondents of the survey were active users, questions were asked with regard to their online activity. Online activity did not just include time spent online but also time spent specifically on Facebook. Respondents were asked the average time they would spend online a day followed by a question asking which percentage of this time online was spent on social media such as Facebook. Additional information was asked with regard to their Facebook activity, these including the number of friends on Facebook, the active groups in which they participated as well as the amount of pages they had liked for their individual Facebook profile. All these factors were combined to interpret how active the respondent was online and specifically, how active they were on Facebook.
30
Demographics: The survey will end with several basic demographic variables, which could indicate if other factors could influence the outcome of the results. These demographic variables include age, gender and educational level. After the age and gender questions, the respondents were asked to mention their educational level, which had 5 possibilities, ranging from a primary school level to a University –Master level. Additionally, the respondents were able to leave a question or comment at the end of the survey or leave their email address if they wished to be informed about the results of the research conducted.
3.5.3 Independent Variable
Social Media Strategies: Social media strategies were assessed by the four online viral marketing campaigns from Kaplan’s (2011) research on the dimensions of viral marketing. The four campaigns were each labelled as a negative or positively oriented campaign as well as strategic and non-strategic. The triumph and stroke-of-luck campaigns were considered positive and the negatively oriented campaigns were nightmare and home-made issues campaigns. Accordingly, the triumph campaign and the nightmare campaign were labelled strategic and the stroke-of-luck campaign and home-made issues campaign was labelled non-strategic.
The triumph campaign was a mass-action oriented campaign in which consumers were challenged to get up and do something active and involve their friends by tagging and
additionally challenging their friends. This campaign is seen as strategic as managers had the ability to set up the campaign and make it a happening including a lot of re-posts as people would challenge their friends, using the brand name. The stroke-of-luck campaign was an April-fool’s joke campaign, in which a toy turtle would supposedly be delivered by an actual turtle, free of charge. As this was a campaign, only funny for a single day, it is considered as non-strategic, as the main purpose is merely for entertainment and is useless once April fool’s day has passed. These two campaigns were supposed to establish a positive feeling for
31
consumers as they were instigated to get involved with friends and family in doing something active.
The nightmare campaign consisted of a large spelling mistake on an Elmo toy box, which had the label “crap your hands” as opposed to the quote saying “clap your hands.” The
introduction of the Elmo was strategically planned for months by the marketers to be posted a few days before the release of the new toy in the Asian market. However, the marketers did not see the error printed on the box as they posted their campaign on Facebook. The final campaign was the negatively oriented home-made issues campaign, in which the company tried to free-ride on a devastating fatal storm, introducing a discount during a storm, which has cost many their lives. This campaign is also non-strategic as the marketers tried to quickly gain some brand awareness and free ride the bad storm to get people motivated, however backfired as many people lost their lives during the storm.
In order to remain constant, each marketing campaign was projected as an actual Facebook-campaign, including a share, like and reply button with the same amount of likes for each campaign. Please view the appendix for an example of each of the viral online marketing strategies.
Willingness to Act: (4 items, α: .7) The main social media marketing campaigns used in this research paper are based on people acting upon the social media posts with their friends and family, willingness to share and acting upon the marketing campaigns and their willingness to spend on these campaigns would set the basis for willingness to act. Questions were set up to identify the willingness to share of the respondent as this could have a significant effect on the results of people are not willing to share much on social networks. The respondents were given a 5-point Likert scale question, asking about the sharing frequency on Facebook with regard to negative and positive campaigns. The frequency could be an important factor as
32
people might not share on Facebook, however, remain active through non interactive
communication. Secondly, willingness to spend was measured as this entails the intention to make direct or future purchases, the respondents were questioned about their purchase
intentions, using a 5-point Likert scale, with regard to both online and social media marketing campaigns as well as their purchase intention with regard to the Active Apes. The willingness to act variables consists of both purchase intention decisions as well as willingness to share a campaign.
3.5.4 Moderator
Company size: The moderator of this research paper was company size; size was either an MNE or an SME. In order for all data to be valid the moderator was manipulated at the start of the survey. Each respondent would randomly be assigned a survey for either an MNE or an SME. A quota maintained a balance between both types of size to ensure that both had a similar sample size. The introduction slides for the companies were kept constant, except for the amount of employees each company had and their headquarters. An SME had 10
employees all positioned at a HQ in Amsterdam, while the MNE had over 50,000 employees based in six HQ all over the globe. As mentioned, all other facts about Active Apes remained constant for both scenarios.
3.6 Data Analysis
Once all the data has been collected, the raw survey data will be processed using SPSS which has the ability to perform various tests to gain the desired results of this research paper. Initially, descriptive statistics will be run to inform about the frequencies, standard deviations and the means of the variables. A chi square test will be used to report the manipulation check as well as the realistic check. The analysis with regard to willingness to act and customer brand engagement will use a regression analysis to investigate the effect of
33
the experiment, One-way ANOVA tests will be used to classify distinct significant
differences between the means of the variables for orientation of the campaigns as well as the strategic intent of the campaigns. The last hypotheses of this research paper will use a
MANOVA test which is a multivariate analysis of variance. A multivariate analysis, as opposed to a univariate analysis makes use of many dependent variables, with any number of independent variables. Due to the nature of the dependent variable, consisting of four
different building a block, this test was proposed as the best method of extracted the desired results. This is due to the fact that the purpose of a MANOVA test is to test the dependent variables and the effect of independent variables on the dependent variable. The MANOVA test was run multiple times to investigate whether the types of campaigns, either
34
4 Results
The following section of this research paper will include the results of the experiment. Initially a few descriptive statistics will be given, followed by the manipulation test and the realistic test. Following these initial tests, the correlations between the variables will be discussed followed by an in-depth discussion of the hypothesis testing.
Table 2: Descriptive Statistics for all Variables
N Minimum Maximum Mean S.D.
Brand Experience 206 1.00 5.00 2.3576 .99645 Commitment 206 1.00 5.00 3.4563 .93862 Loyalty 206 1.00 5.00 2.5813 1.07794 Satisfaction 206 1.00 5.00 3.0837 1.03827 Size 206 1.00 2.00 1.4951 .50119 Campaign Orientation 206 1.00 2.00 1.4854 .50101 Campaign Strategy 206 1.00 2.00 1.5291 .50037 Willing To Act 206 2.25 12.00 5.2221 1.69975
Social Media Time 206 0.00 3974.55 546.2522 402.95679
Table 1 shows the final sample size of the experiment consists of 206 accepted responses selected for the data analysis. (Note: the original sample size was 258; this will be explained in the manipulation check section of the results) The table shows the mean and standard deviation accordingly. Table 2, show the frequencies which occurred during the experiment. Looking at the frequency for size we are able to assume that the balance between MNE respondents and SME respondents is almost at 50%. The campaigns are equally balanced between respondents as they only range from 22, 8% for Nightmare respondents and 27.2% for stroke of luck respondents.
35
Table 2: Frequency chart(s) for Size and Campaign
Table 3 and 4 of this analysis show a Chi-square test in which all demographics of the respondents are tested with regard to the orientation of the campaign (table 3) and the campaign strategy (table 4). A campaign is tested according to the demographics of the respondents, through age, gender and level of education. As can be concluded from the p-value, there is no distinct significance which entails there is no relationship between the demographics of the respondents and the campaign to which they responded.
Table 3: Chi-Square applied to Campaign Orientation for Demographics
Chi-Square df p
Age 34.10 35 0.51
Gender 0.40 1 0.53
Education 1.43 4 0.84
Table 4: Chi-Square applied to Campaign Strategy for Demographics
Chi-Square df p
Age 31.99 35 0.61
Gender 0.17 1 0.39
Education 2.78 4 0.59
Having looked at the descriptive statistics of this experiment, the following section of this paper will investigate the effect of the manipulation check including the realistic check.
4.1 Manipulation Check
The manipulation check for this research paper was measured through the analysis of the marketing campaign which was presented to the respondents. Each of the four viral marketing campaigns was given specific labels to correspond with the campaign. The respondent was given contradicting labels including success/failure, luck/unlucky,
36
planned/unplanned and a desirable/undesirable. The manipulation check used a binary choice to establish whether the respondent understood the purpose of the campaign. Table 5 shows a cross-tabulation between the variable manipulation test, created to test whether or not the respondent passed or failed the test, and the type of campaign.
Table 5: Manipulation Test vs. Campaign Orientation cross-tabulation
Campaign Type Total Positive Negative
Manipulation Fail
Count 21 31 52
% within Campaign Orientation 16.5% 23.7% 20.2%
Pass
Count 106 100 206
% within Campaign Orientation 83.5% 76.3% 79.8% Total
Count 127 131 258
% within Campaign Orientation 100.0% 100.0% 100.0%
As table 5 shows, the percentages of people that failed the manipulation test are 16.5% for the positive campaigns and 23.7% of the respondents failed the negative campaigns, created a total percentage of 20.2% of the respondents failing to properly address the type of
campaign. Table 6, a Chi-square test, shows that there is no-significance between both variables (p = .152).
Table 6: Chi-square test of Campaign Orientation for Manipulation test.
Looking closer at the percentage of the positive marketing scenarios (M: 0.83 SE: 0.033) versus the negative scenarios (M: 0.76 SE: 0.037), gathered from table 4, we see that the problem lies in with the negative oriented marketing campaigns. The individual marketing campaign in reference to the manipulation test we see that the positively oriented campaign Triumph (M: 0.82 SE: 0.05) and Stroke of Luck (M: 0.84 SE: 0.04) as well as negatively
Value df p
Pearson Chi-Square 2.036a 1 .154
Continuity Correction 1.617 1 .203
Likelihood Ratio 2.048 1 .152
37
oriented campaign Home-made issues (M: 0.82 SE: 0.49) show a better response rate as to the negatively oriented Nightmare campaign (M: 0.71 SE 0.056). Table 7 shows the frequencies of the malfunction test per specific viral marketing campaign.
Table 7: Frequency table for Malfunction test among all campaigns
The nightmare campaign recorded 66 respondents, which entails that 19 respondents did not comprehend the purpose of the nightmare marketing campaign to its fullest. Due to the specific set-up of the manipulation test, no respondents had been dropped until this point in the experiment. However, after careful consideration and in order to remain consistent among all campaigns, all those respondents which failed the manipulation test, not just the nightmare campaign, will be disregarded from the results, reducing the sample size from 258 to 206.
In order to test if the campaigns were realistic and therefore making this research paper realistic, the respondents were asked to report if they could imagine this marketing campaign to occur in real life. Table 8 and table 9 show a cross-tabulation between the orientation of the campaign and variable regarding a real campaign and a realistic check of the company.
Table 8: Real Campaign vs. Campaign Orientation cross-tabulation
Campaign Orientation Total Positive Negative
Real Campaign
Real Campaign
Count 89 77 166
% within Campaign Orientation 84.0% 77.0% 80.6%
Unreal
Count 17 23 40
% within Campaign Orientation 16.0% 23.0% 19.4% Total
Count 106 100 206
38
Table 9: Real Company vs. Campaign Type cross-tabulation
Both tables show, that the respondents had a realistic view of both the company Active Apes (M: 1.2 SE: 0.025) as well as the various campaigns (M: 1.22 SE: 0.026), as the percentages of unreal for positive campaigns show 16% believed the campaign to be non-realistic and 23% of the negative respondents did not find the campaign realistic. Accordingly, 15% of the positive respondents believed the Active Apes was not able to operate as a real-life company and 24.4% of the negative campaigns respondents were not convinced by Active Apes. The percentage difference between the positive and negative is peculiar as the layout for the company was identical, except for the size of the company.
4.2 Correlations
Through the use of descriptive analysis and an APA Correlations Matrix, found in table 10 below, the correlating variables will be described in the following section of this paper. Note that all respondents which failed to pass the manipulation test will be excluded in all
experiments for the remainder of the data analysis.
The first correlations which are most vital for this paper are the correlations among the dependent and independent variables. It is interesting to see that the Campaign orientation negatively correlates with the four dependent variables, Brand Experience, Commitment, Loyalty and Satisfaction. Although these correlations are rather small (between r = -.36 and r = -.69, at p < .01), it shows that positive marketing campaigns have a positive effect on our
39
four dependent variable. As a result, Satisfaction is mostly effected by this negative correlation as this shows the largest correlating factor (r = -0.69 at p < .01). The strategic nature of the campaign shows almost no correlation with the four dependent variables (between r = 0.01 and r = 0.05).
Secondly, Willingness to Act correlates differently with each of our four different dependent variables. Willingness to act correlates positively with both Brand Experience (r = 0.58 at p <0.01) as well as Loyalty (r = 0.63 at p <0.01). These correlations make sense because as a consumer is willing to make the extra effort, it increases loyalty as well as their brand experience.
Table 10: APA Variable Correlations Matrix
Mean S.D. 1 2 3 4 5 6 7 8 9 1 Brand Experience 2.36 1.00 (0.87) 2 Campaign Orientation 1.49 0.50 -.392 ** - 3 Campaign Strategy 1.53 0.50 0.01 0.00 - 4 Commitment 3.46 0.94 .637 ** -.401** 0.05 (.66) 5 Loyalty 2.58 1.08 .815 ** -.358** 0.01 .641** (.87) 6 Satisfaction 3.08 1.04 .615 ** -.693** 0.05 .599** .633** (.91) 7 Size 1.50 0.50 0.07 -0.03 -0.02 0.00 0.11 0.11 -
8 Social Media Time 546.25 402.96 .205 **
0.07 0.12 .149* .257** 0.13 -0.02 (.40)
9 Willing To Act 5.22 1.70 .583 **
-0.14 0.06 .420** .631** .319** 0.11 .419** (.70)
Note: **. Correlation is significant at the 0.01 level (2-tailed). N:206 Note: *. Correlation is significant at the 0.05 level (2-tailed). N:206 Note: (.xx) = Cronbach’s Alpha Coefficient
Finally, we see that all four dependent variables, correlate among each other. All correlations among the dependent variables show the interrelatedness of the customer brand engagement model presented by Hollebeek (2011). It therefor makes sense that these variables are correlating as they all inter-dependent on each other.
40
4.3 Hypothesis testing
The following section of the paper will include the testing of the hypotheses, in which the dependent variables, as well as the moderating effects will be discussed in relationship with the given hypotheses. First the impact of willingness to act on customer brand engagement will be discussed followed by the impact of viral social media campaigns and their impact on customer brand engagement. The hypothesis testing portion of this research paper will conclude with the analysis with regard to effect of company size on customer brand engagement.
4.3.1 Impact of willingness to act on Customer brand Engagement
The main dependent variable of this experiment is Customer brand engagement which consists of four building blocks Brand Experience, Commitment, Loyalty and Satisfaction.
The first hypothesis in this research experiment states that the willingness to act has a direct effect on customer brand engagement (H1). Additionally the sub-hypotheses states that willingness to share has a positive relationship to brand experience (H1a) and willingness to spend is has a direct positive effect on loyalty. A multiple regression analysis was to test if the willingness to act significantly effected customer brand engagement. The results of the multiple regression analysis indicated the predictors explained 65, 6% of the variance (R2=.43, F (4,201) =37.99, p<.01). Furthermore, table 11 shows the regression table for willingness to act and the building blocks of customer brand engagement.
Table 11: Regression analysis of willingness to act and customer brand engagement
Willingness to Act β S.E t p (Constant) 2.80 .358 7.83 .000 Brand Experience .423 .164 2.58 .011 Commitment .072 .136 .528 .598 Loyalty .823 .154 5.34 .000 Satisfaction -.307 .120 -2.55 .011
41
As can be concluded from these results that loyalty, is significantly related to willingness to act (β = .522, p<.001). Additionally, it can be stated that only commitment has no significant relationship with willingness to act (p = .598). Therefore, H1 is supported as all three
significant variables of customer brand engagement show a positive relationship with
customer brand engagement as a result of willingness to act. However as stated in research, it was understood that willingness to act was directly related towards the behavioral aspect of the consumer when it comes to customer brand engagement, therefore the sub-hypothesis will provide a better understanding.
In order to test both sub-hypotheses, additional multiple regression tests were conducted to test the influence of willingness to share on brand experience and willingness to spend on loyalty. For the first sub-hypothesis the components for willingness to share were tested through a multiple regression analysis with brand experience as a dependent variable. The results of the multiple regression analysis showed that the two indicators explained 57, 8% of the variance (R2=.33, F (2,203) =50.91, p<.01).
Table 12: Regression analysis of willingness to share with brand experience.
Brand Experience β S.E t p Share Positive .208 .057 3.656 .000 Share Negative .259 .049 5.237 .000
The table shows that the willingness to share of a consumer has a positive influence on brand experience as willingness to share a positive campaign (β = .208) and willingness to share a negative campaign (β = .259) are both significant. The second sub-hypothesis was set up to show the relationship between willingness to spend and loyalty. The multiple regression analysis of this hypothesis showed that the both indicators of willingness to spend explained 40, 5% of the variance (R2=.164, F (2,203) =19.965, p<.01).
