Increase your brand Customer Engagement : choose the right sequence of fit between your Online Content Posts and the Brand Fan Page

(1)

Increase your brand Customer Engagement:

Choose the right sequence of fit between your Online Content Posts and the Brand Fan Page MSc Thesis Business Administration – Digital Business Assignment: Final version Supervisor: Sara Valentini Student: Yoram Knorringa – 11129816

(2)

2

Statement of Originality

This document is written by Student Yoram Knorringa who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are 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)

3

1. Abstract

In today’s highly dynamic and digital business environment, many brands feel the need to engage in this interactive environment to increase their online customer engagement. With the rapidly increasing use of social media networks by customers, the interest in social media marketing has also grown tremendously among marketers and scholars. Despite this interest, research about how to increase the online customer engagement of a brand is limited to date. The brand extension literature discusses the concept of perceived fit, however, little is known about how this perceived fit influences the level of customer engagement. This thesis aims to examine how the sequence of the perceived fit plays a role in the level of customer engagement. The findings are based on a cross-sectional survey design with 254 respondents. Three statistical analyses, ANOVA, MANOVA and linear regression, are conducted to test three hypothesized relationships. The research shows that predominantly broadcasting Facebook posts that are perceived as high fit by the brand extension literature, gradually intervened by Facebook posts that are perceived as low fit, achieves the highest level of customer engagement. Also, this sequence achieves a significantly higher number of comments than all other sequences of perceived fit and significantly achieves more clicks than the sequence that predominantly broadcasts Facebook posts that are perceived as low fit and gradually intervene those by Facebook posts that are perceived as high fit with the parent brand. Keywords: customer engagement, social media marketing, perceived fit, sequence, brand extension, online content post, brand fan page

(6)

6

2. Introduction

In 2025, less than ten years from now, 80 percent of all advertising worldwide is expected to be fully digital (Benjaminsen, 2017). The push strategy of Zalando can lead to their own demise and the YouTube channel owned by L’Oréal can help them to become even bigger. The Googles, Amazons and Facebooks of this world are known as online heroes, but the older heroes like Coca Cola, Nike or Volvo are also evolving as online brands. There is no way back, the importance of the digital marketing strategy of a brand is more important than ever. Research has shown that there are 3.03 billion active social media users and that 91% of retail brands use 2 or more social media channels (Bandwith, 2017). The interaction that takes place on this social media channels can shape the online reputation of a brand. The social media marketing strategy will help brands to maintain customer relationships and increase the engagement of the customers with the brand (Wang & Kim, 2017). Social media gives brands the opportunity to reach their customers via their own channel with content created by themselves (Edelman & Salsberg, 2010). Via the brand fan pages brands can choose what they post, when they post and what kind of interaction is hidden in the post. Why did T-Mobile launch their Pets Unleashes marketing campaign that has nothing to do with the product they sell? And why should Shell post content about free coffee? The online content posts on a brands’ own fan page can be seen as owned media (Edelman & Salsberg, 2010). The brand chooses to post this content themselves and shares this with their online community. However, this chosen content can simultaneously result in earned media. The more people engage with the post, the more likely other social media users will get in contact with the post at the same time. An example of this can be sharing the post with their friends. An online content post that matches the recipient’s interest is more likely to be shared by someone (Zhang, Moe and Schweidel, 2014). Sharing a post illustrates a certain level of engagement with the post. Previous literature has shown that the specific content of these online posts can play a role in the level of engagement of online

(7)

7 customers (de Vries, Gensler and Leeflang, 2012). Heath, Bell and Sternberg (2001) corroborate this finding by showing that online content that triggers emotions is more likely to be shared. In addition to this, Wang and Kim (2017) suggest that brands merge social media into their marketing strategies to improve customer engagement and firm performance. This research examines the congruity between the online content post and the brand fan page, indicated as: the perceived fit. This phenomenon originates from the context of brand extensions and can be based on the product extension being compatible with the parent brand (Kim and John, 2008). The research of Kim and John (2008) and Völcker and Sattler (2006) even suggests that the perceived fit is the most important determinant of brand extension success. This literature focuses on the degree of fit between the brand and the extension product (Aaker and Keller, 1990) and states that a higher level of fit is more favorable than a moderate fit (Kim and John, 2008). In a similar vein, the online generated content of a brand on their brand fan page can be perceived as a high fit or low fit with the brand. A car manufacturer, like Ford, can launch a new marketing campaign on their Facebook page by introducing their new Ford Focus. This Facebook post can be perceived as high fit since its complementing with other products sold by the parent brand Ford (Broniarczyk and Alba, 1994). On the other side, if a telecom provider, such like T-Mobile, launches a new marketing campaign focused on pets this fit can be perceived as a low fit because pets have never been the main focus of the brand and they have nothing to do with their core business. A Facebook post that is perceived as high fit can increase the likelihood of the customers and form more positive attitudes (Aaker and Keller 1990; Kim and John, 2008) whereas a Facebook post perceived as low fit can have a positive impact on the level of customer engagement since moderately incongruent content is expected to increase the customer’s attention (Meyers-Levy and Tybout, 1989; Mandler, 1982). The literature review will give a more extensive analysis and discusses the relevant findings in this topic area. However, the literature lacks

(8)

8 research about the right order of high fit and low fit posts. Little is known about when to choose a high fit or low fit post and in which sequence you can achieve the highest level of customer engagement. In this thesis, I aim to examine how the sequence of perceived fit plays a role in the level of customer engagement. Starting from the literature by Naik, Mantrala, and Sawyer (1998), I propose five different sequences a brand can use to find the right balance between the high fit posts and the low fit posts of a brand generated online content post and the brand fan page to achieve the highest level of customer engagement as the main goal. Little is known about the focus on the variety of sequence. Considering this gap in the literature, this research contributes by answering the following research question: To what extent does the sequence of fit between brand generated online content posts and the brand fan page have impact on the online customer engagement level of a brand? Moreover, which sequence strategy can be seen as most effective? This research contributes to the theory by extending the literature with the proposed literature gap as mentioned above. The results and conclusions of this research can help managers or digital marketers to make the right social media strategy decisions. This way the research also has a managerial contribution. Furthermore, this research can give a base for further research about this topic. The next chapter contains the literature review that identifies and explores the existing literature gap.

(9)

9

3. Conceptual Framework

The main focus of this research is to analyze if the sequence of perceived fit between a brand generated online content post and the brand fan page has an influence on the level of customer engagement. This direct relationship will not be measured on a single post, but rather on a sequence of posts. Before discussing the relevant literature, the conceptual framework is visualized in figure 1 with the sequence of perceived fit as independent variable, existing of five sequences, and customer engagement as the dependent variable. Figure 1: Conceptual framework Sequence of Perceived Fit (IV) Between the Online Content Post and the Brand Fan Page (at time t) _{Customer Engagement (DV)} (at time t) Likes _Comments _Shares _Clicks The framework illustrates that this research will analyze the relationship between the sequence of fit and customer engagement. Five different sequence strategies are taken into account which all include a different sequence of posts in terms of degree of fit, this difference will be explained later on in the literature review. More, the literature review will start by giving some background information about the current marketing methods, the impact of a brand fan page and the importance of content in combination with the concept of virality. Thereafter, the concept of the dependent variable, customer engagement, will be discussed and the literature review ends with a more in-depth focus on the perceived fit, the sequence of posts and the strategies that are formulated.

(10)

10

4. Literature review

Within this part of the research, all relevant literature is discussed to elaborate on the conceptual framework. The literature review will discuss some overlapping themes as it will discuss more specifically done research to get a more in-depth view of certain issues. At the end of the literature review, the hypotheses are formulated, based on the discussed literature.

4.1 Social media, brand fan pages, online content posts and virality

Since the rapidly increasing use of social media networks by customers, the interest in social media marketing has also grown tremendously among marketers and scholars. Trusov, Bucklin and Pauwels (2009) suggest that the fundamental nature of social media is offering a platform for users to interact with each other. In addition to this, the authors conclude that the use of social media leads to higher response rates and a higher level of customer engagement than the traditional marketing strategies. Before this research goes deeper into the concept of customer engagement, three other aspects regarding the social media platform will be discussed. 4.1.1 Traditional vs social media marketing Brands are forced to use new ways of effective communications since the young consumer are more likely to rely on digital marketing communications and have a less focus on the traditional ways of marketing such as television or print (Generation Z., 2013). Traditional marketing is mostly focusing on a broad target group, where digital marketing makes it easier to target people individually. Although, Edelman and Salsberg (2010) state in their research that traditional marketing still plays a major role, they also acknowledge that there are new, less costly, forms of media. Traditional marketing such like television, newspapers or billboards etc. are most commonly ‘paid’ media. Within this form, the company pays for advertisement place or sending a message via a third party (Edelman and Salsberg, 2010). However, the downside of this form of marketing is that there

(11)

11 is very little interaction possible between the medium used and the customers. This way of only providing information doesn’t give the company much customer behavioral information except for the hope that these people patronize with the brand. This form of one-way communication was mostly conducted by a marketing department operating separately from the other departments whereas Stone and Mason (1997) in earlier research already mentioned that marketing should be something from the whole organization. It should play a more central role. The technological revolution of the last decade has evolved the internet in a dynamic network where people can easily connect with each other at any time of the day (Cheung, Lee, and Rabjohn 2008; Rooderkerk and Pauwels, 2016). Social media platforms give customers the possibility to interact with each other, conceive information about products or services and be in touch with organizations. The organization on the other hand, can also make contact with the customer by posting online content or asking questions on their brand page. This two-way interaction has grabbed the attention of scholars and marketing departments on how to succeed in this dynamic network. Schubert and Selz (1999) already mention before the 21 century that we should leave traditional marketing behind us and that it’s time to focus on web based marketing. Cho, Fu and Wu (2017) have identified the most popular research topics in the marketing journals between 1995 and 2014. Their research state that the focus have moved from a more product and channel coordination focus towards a more service dominance and customer centrality focus. Cho et al. (2017) suggest that the current customer has overcome materialism and consumerism which resolves in a different way of interpreting the marketing practice. Instead of focusing on the hedonic and individualistic approach, marketing practice now focusses on co-creation of experiential and social value. Marketers focus themselves on serving the customers mental interest instead of exploiting their weaknesses (Cho et al., 2017).

(12)

12 As discussed, brands focus more on digital marketing nowadays than the traditional marketing (Generation Z., 2013) and even more particular on social media marketing. Deploying a digital marketing strategy is essential for every brand nowadays. First of all, digital marketing gives brands the opportunity to measure and quantify results with tool such as Google Analytics, Google Adwords and Moz Pro. The generated data within these tools are creating valuable insights in which marketing campaign is valuable (McClurg, 2012). Secondly, ninety-eight percent of the customers start their customer journey online (Marketingfacts, 2016, p.16). If a brand wants to capture its customers it should be active in this online environment. Third, digital marketing offers brands the possibility to reach an enormous amount of people. It connects the brand with many people around the world and in combination with social media it can even multiply this reach (McClurg, 2012). More, social media enables the possibility for a digital marketing campaign to go viral. The virality of an online content post will be discussed in the paragraph 4.1.3. This research appeals to the challenge of many marketers who are struggling with their social media marketing strategy. The dynamic and emergent nature of social media makes it difficult for marketers to effectively use social media as a marketing communication channel (Keegan and Rowley, 2015). The two-way conversation exchange is a central aspect of the channel which makes it different to other digital marketing channels (Shih, 2009). Customers nowadays expect brands to adopt and manage their social media platforms to interact with them (Labreque, 2014). Thereby, many brands feel this need to engage in existing social media conversations and are investing in the channel because they want to increase their customer engagement, drive online sales or protect their brand reputation (Keegan and Rowley, 2015). However, as the authors also mention, the return on these investments is not always easy to quantify (Keegan and Rowley, 2015). They contribute to the dilemma why social media marketing

(13)

13 (SMM) is difficult to use effectively as marketing channel by stating that: ‘most challenges arise from the relative novelty of SMM and therefore the limited experience with SMM of both parties’ (p.26). This uncertainty and lack of knowledge, potentially, makes it harder to establish strong and long-term relationships with the customer via social media. The research of Confos and Davis (2016) also addresses this relationship between the brand and the customer via social media. Though, they translate these relationship challenges into opportunities where brands are considered an interactive partner of the customer. Social media channel Facebook for example offers this interactive relationship where the brand can share and converse with the customer. These characteristics of social media in combination with the two-way communication principle and the increasing presence of customers online (Marketingfacts, 2016) offer brands the possibility to establish potentially long-term relationships (Confos and Davis, 2016). Besides the relationship aspect of social media, as discussed above, social media marketing has more to offer. According to Srinivasan, Rutz and Pauwels (2015) in Tornhill, Xie and Lee (2017) the novelty of social media marketing exists of: better targeting, better tracking and better responding. The first, as mentioned before, focusses on the capability to target the customers personally whereas traditional marketing lacks this capability. Additional, retargeting makes social media even more valuable which can include: referrals, likes or previous earned messages that can increase the online sales (Brettel, Reich, Gavilanes and Flatten, 2015). Second, the tracking of customers makes it possible to better understand the customer decision process and the customer relationship with the brand (Edelman and Singer, 2015; Tornhill et al., 2017). And third, brand managers can measure the customers sentiment based on likes, comments, shares and clicks for example to overlook which products and services are working or not. Based upon this measurement the brand managers: ‘can echo consumers’ positive comments of user experience to a new product, solve service failures by bringing solutions and suggestions in managerial responses (Xie, Zhang and Zhang, 2014; Tornhill et al., 2017). In addition to

(14)

14 this, marketers and brand managers can respond via social media by offering discounts, responding to questions or make announcements. This paragraph has discussed the advantages of digital marketing and in more particular the advantages and challenges of social media marketing. The next paragraph gives an in-depth view of the brand fan page and it’s most important aspects and considerations. 4.1.2 Brand fan page The brand fan page allows brands to interact with their community members, potential customers and gives customers the possibility to interact with each other. In this research, the Facebook brand page will be used as predominant example of a brand fan page. Whereas in 2012 Facebook had more than 1.5 million businesses with a brand page (Jeanjean, 2012), this number has increased to over 70 million business users in 2017 (BirdSongAnalytics, 2017). Top leader Red Bull has almost 49 million fans on their Facebook brand page (SocialBakers, Ferbuary, 2018) which can be explained by the more than 1 billion active users per day (SproutSocial, February, 2018). The information on brand fan pages is shared in the form of posts and they appear in the central part of the brand fan page which is known as the ‘timeline’. All Facebook users that have liked the brand page once, the fans, can see the posts on their timeline and have the possibility to engage with the post through liking, commenting or sharing the post (Luarn, Lin and Chiu, 2015). These brand fan pages can create, strengthen and enhance customer relationships (Tsimonis and Dimitriadis, 2014). Besides, the fan brand page can be used to gain brand awareness, easily interact and listen to your customers to gain engagement with people who like the brand (Tsimonis and Dimitriadis, 2014). The interactive character of the brand page gives brands the possibility to build relationship with people already at a young age. More, evidence suggests that customers experience a brand message from the online web not even as advertisement, but more as information which they obtain through their

(15)

15 experience with the brand (Moore, 2004; Confos and Davis, 2016). Despite these opportunities Luarn et al. (2015) suggest that brands shouldn’t create a Facebook brand page just because others have done so. The negative impact of a brand page will be discussed later on in this section. Brands are in control about what kind of content they generate on their brand fan page. These owned social media channels are controlled by the brand, such like Facebook or Twitter pages (Tornhill et al., 2017). Other examples of owned media can be catalogs, websites, e-mail and company-owned retail stores which can be used for marketing purposes (Edelman and Salsberg, 2010). The authors Edelman and Salsberg (2010) discuss not only owned media in their research, but also elaborate on paid and earned media. Whereas paid media is a more traditional way of marketing, earned media is playing an important role in the social media marketing strategy. The authors state: ‘Earned media is generated when the quality or uniqueness of a company’s products and content compel consumers to promote the company at no cost to itself through external or their own “media.”’ (p. 3). This definition explains the important role of the interactive users on the social media platform. Without any costs for the initial company, promotion of the post can happen via the engagement of the users where a transition exists from owned media towards earned media. The owned media, controlled by the brand, can resolve into earned media when the users engage with the online content (Tornhill et al., 2017). As explained above, the brand page can be used to increase customer engagement with the brand. This form of earned media exists of individuals that like, comment or share the specific post and is purely effective for the initiated brand. More, it does not have any spill-over effects which can happen with normal advertisement post regarding a certain product line (Tornhill et al., 2017). Other tactics can be to encourage purchase behavior with sales promotions or extending the online experience by linking to other brand related websites in the generated posts on the Facebook brand page (Confos and Davis, 2016). Whether you want to increase sales or extend the online experience, the main focus is gaining

(16)

16 the biggest reach with your post. This starts with the intention of customers to engage with the post and can end up into a viral post with enormous reach. Although, the brand fan page gives brands plenty of opportunities, a brand should also be cautious with their brand fan page. As the research of Luarn et al., (2015) concludes: you shouldn’t start with a brand fan page because others have done so. A brand should be aware what the consequences of a brand page can be. The brand page gives a brand the direct possibility to engage with their community, but simultaneously gives the customer the opportunity to directly address the brand in the form of criticism, an attack or to ask controversial questions. Edelman and Salsberg (2010) state that it’s very important to develop a clear community or social networking strategy. Since all online content can be seen by everyone all over the world, one-to-many communication, it’s more than important that a brand chooses its particular content carefully. In addition to this, a brand should agree upon a set of rules and principles for managing and responding to an attack via the brand page (Edelman and Salsberg, 2010). Due to its one-to-many characteristic, a failure to respond appropriately can be high (Ries and Trout, 1981). To obtain most customer engagement a brand should rebalance time and resources regarding their social media marketing strategy. A well-structured brand page requires cultivation, patience and sustained engagement. 4.1.3 The importance of content and content virality In this paragraph both the content of an online post and the concept of virality is discussed. As the previous paragraphs has shown, it’s known that there has been a shift in the marketing strategy of most companies from a traditional strategy towards a more interactive, social media marketing strategy. Thereby, the controlled brand page of a company can play a significant role in the decisions which content to broadcast to their community and potential customers.

(17)

17 Besides these facts, also the importance of the content of the post is discussed in the literature. De Vries et al. (2012) say that: ‘positioning the brand post on top of the brand fan page enhances brand post popularity’. This popularity can be, just like engagement, measured by the number of likes, comments, shares and clicks. Moreover, the authors also conclude that other, content based, variables influence the popularity of a content post such like videos, messages, quizzes, information, and other material. In more recent research the authors Barcelos, Dantas & Sénécal (2018) have pointed out how the tone of voice of the online content post can have an influence on purchase decisions. The authors make a difference between a more human tone of voice and a traditional corporate voice. The study shows that using a human voice can increase the purchase intentions of a customer with the knowledge that the purchase intentions can be stronger when the customer has a hedonic goal in mind. On the other hand, using a human voice can decrease the intentions to purchase of a customer due to the perception of risk humanness brings along (Barcelos et al., (2018). The authors Schulze, Schöler & Skiera, (2015) conclude in their research that there is no easy, one-size-fits-all solution for social media marketing. Instead of spending time and resources to expenditure an overall social media strategy, companies should tailor their social media marketing strategies that fit each product exclusively. Useful products will benefit form a different strategy than for example games (Schulze et al., 2015). In this sense, the authors build upon the statement that the content of a post matters. Choosing the right tone of voice in combination with the right information shared regarding the sold product or marketing campaign can actively contribute to the engagement of customers with the online post. Rooderkerk and Pauwels (2016) contribute to this idea with their research in which they investigate what features and characteristics can define the number of comments that a post receives. The authors not only find out that the who and the when matters, where who is the person/company

(18)

18 posting the content and the when considers the timing of the post, but even more important for this research is the what and the how of an online post. The what, considers the actual content of a post not based on visibility such like emoticons, but rather on the response of the customer it evokes (Rooderkerk and Pauwels, 2016). This is in line with earlier mentioned research of De Vries et al. (2012) and Luarn et al. (2015) who have also shown that the type of content is a crucial element in posting strategies. People are more likely to engage with more controversial content, because it motivates people to give their opinion about the controversial standpoint of the author (Rooderkerk and Pauwels, 2016). In contradiction to this, when a topic is ambiguous, people are uncertain about the meaning of the content and rather choose not to respond. Besides the what, the how also matters considering the urge of engagement of others. The length of a post should be informative and not too long, where a short message evokes more response. In addition to this, a brand marketer could also decide to ask a question which can lead to a higher response rate (Rooderkerk and Pauwels, 2016). Brand marketers are choosing the right content to achieve the highest level of engagement with the ultimate goal of a post going viral. This concept of virality is discussed by Berger and Milkman (2012) who show that positive content, including smileys, have a higher possibility to go viral. Their research is replicated by Heimbach & Hinz (2016) three years later and they have extended the analysis to three additional communication channels and the non-linear relationship between virality and positivity. The author’s findings are consistent of the earlier findings of Berger and Milkman (2012) across all communication channels. A channel like Facebook has the capability to let an online post going viral by its features to like, comment, share and also tag a person. The Ice Bucket Challenge in August 2014, where donating for ALS was the central message, can be seen as an example where people could activate others by posting the content online and ‘tag’

(19)

19 others in the post to make sure they would do the same. This tagging, possible on Facebook, encourages friends, family or others to engage with the post and even share the post with their own network (Confos and Davis, 2016). This type of viral marketing can be used by brands, persons and charities just like the ALS Ice Bucket Challenge. Considering the fact that a post that goes viral achieves the highest level of customer engagement with the post, a brand marketer should consider the research of Berger and Milkman (2012). Where positive content is more likely to go viral, the content of the post should also match the message you want to get across. To achieve virality it’s not only important that customers engage with the post by liking and commenting, but also sharing the post which is seen as the highest level of engagement (Kietzmann, Hermkens, McCarthy and Silvestre, 2011). This will be discussed more in-depth during the next chapter regarding the concept of customer engagement.

4.2. Engagement

The literature discussed in the previous chapter must be considered by brands when they decide to enroll their social media marketing strategy. The characteristics of social media marketing, the role of a brand page and the importance of choosing the right content are all necessary to take into account when a brand marketer wants to achieve the highest level of customer engagement. The next chapter will discuss the concept of customer engagement, its definition and the several levels the phenomenon can adopt. Since the concept customer engagement is used many times in this section and the next it will be indicated with CE to improve the reading. 4.2.1 Definition of engagement and its importance Despite the extensive use of the word ‘engagement’ throughout the literature, there is no concession about the ‘real’ definition of this concept. Mostly its referred to as ‘consumer/customer engagement’ or ‘brand engagement’. The term has been used in many different disciplines including sociology,

(20)

20 psychology, political science and also in marketing. Although the concept is used in several disciplines Brodie, Hollebeek, Juric and Ilic (2013) highlight that: ‘the role of interactive customer experience and co-created value are the underlying conceptual foundations of customer engagement (CE)’ (p. 257). The upcoming paragraphs will elaborate more on this. Many organizations have used the level of engagement as a predictor, heightening the level of engagement can contribute to an increase of the productivity and profitability of a company (Greenwood, 2007; Hollebeek, 2011). Not only is CE framed different across the disciplines, also many authors differ from definition. Both Bowden (2009a) and Bejerholm and Eklund (2007) consider customer engagement as a psychological process, whereas Fredricks, Blumenfeld and Paris (2004) define CE as a more context dependent variable. In 2012, Gummerus, Liljander, Weman and Pihlström also considered the emergence of new media and its relationship towards CE. In their opinion CE consists of ‘all consumer-to-firm interactions and consumer-to-consumer communications about the brand.’ (p.858). In here, the authors take into account the two-way communication characteristic of CE. The definition of CE that will be used in this research is the definition of Hollebeek (2011): “The level of an individual customer’s motivational, brand-related and context-dependent state of mind characterized by specific levels of cognitive, emotional, and behavioral activity in brand interactions (p.790)” The ‘brand interactions’ can be seen as the interaction between a customer and an online post of a brand on their Facebook brand page. The customers’ ‘state of mind’ can be formed due to earlier brand related communication which evokes a certain motivation (or not) to engage with the post. The definition also takes into account the context of the post. The previous section has shown that posting online content on your brand page can multiply the popularity of the post where also timing can play a role. The specific levels in brand interaction that Hollebeek (2011) mentions is about the perception of

(21)

21 the brand interaction. How the customer judges the post on a cognitive and emotional way also depends on the content as discussed in the previous chapter. A more controversial post can evoke a customer to respond because of the emotional feeling and the urge to contribute with their opinion (Rooderkerk and Pauwels, 2016). The behavioral part can exist of the action to like, comment, share or click on the post based upon the state of mind. Reading a post is one thing, but engaging with online content posts via social media channels is a next step. Sashi (2012) explains the rise of the interest in CE and points out the importance of CE for the success of an organization. The use of social media plays an important role in achieving a higher level of CE. This interactive, digital media not only allows the brand to share and exchange information one-to-many with their customers, but it allows customers to do the same with one another as well (Sashi, 2012). Considering the above, the engagement level of a post is not only due to the chosen content of the brand, but is also based on the peer-to-peer communication between the customers and their willingness to like, comment or even share the post with their network. Kietzmann, Hermkens, McCarthy and Silvestre (2011) underpin the importance of understanding the role of social media in CE. They illustrate several building blocks which a brand should take into account to understand the engagement needs of their audience. A brand that wants to engage with their customers based on relationships should use Facebook to achieve this best (Kietzmann et al., 2011). Just like Sashi (2012), Brodie et al. (2013) also point out the importance of engaged customers. In their research, the authors state that the engaged customers also play a key role in viral marketing activity by ‘providing referrals and/or recommendations for specific products, services, and/or brands to others.’ (p. 252). As previous literature illustrates, viral marketing is a sincere focus point for brand

(22)

22 marketers. Given this importance of CE and its definition, the next paragraphs will discuss the levels of engagement and how we will measure CE in this research during the methodology. 4.2.2 Levels of engagement Customer engagement can be measured based on several levels. The possible levels of engagement with a Facebook brand page post can be: liking, commenting, sharing or clicking on the post. When speaking about CE levels, these levels will be considered in the research. Liking a post can be before or after commenting, sharing or clicking on a post where the number of likes can indicate the reputation or popularity of that post (Kietzmann et al., 2011). A post with many likes can indicate that the content is of interest and simultaneously stimulates more people to have a look at the online content (Luarn et al., 2015). This can be seen as the Facebook algorithm where more likes resolve in a bigger reach, with more people interact with the post, more people getting interested and therefor like the post. This results again in more reach and so on. Previous research has shown the relevance of a comment considered as an engagement factor (Brodie et al., 2013 & Hollebeek, 2011). Their results show that someone who comments more or who’s comments are longer than the average, indicates a higher level of engagement. More, a high number of comments on a post suggests that customers have taken their time to share their opinion. This can be used as an indicator to measure the degree of success or impact of the post (Luarn et al., 2015). The third level, sharing, has been mentioned before in the virality paragraph 1.3. Taking into account that viral content achieves the highest level of CE, a major part of this is people sharing the content with their own network. More, sharing a post can be seen as the highest level of engagement (Kietzmann et al., 2011). A person that shares the brand generated content on their personal time-line indicates that he identifies himself with the content or feels the urge to let others know about this content.

(23)

23 Fourthly, besides the more visually engagement factors of liking, commenting and sharing, customers also have the possibility to click on a Facebook post. This mostly reveals extra information about the subject of the post and is used in the literature as an indicator for engagement (Peterson, 2006). Clicking on a post indicates that the information shared by the brand is interesting enough and socially motivates the customer to endure his interaction with the brand on that moment (Andrew and Parsons, 2002). More, clicking is the best indicator to explain conversion. After a customer clicks on the post, this firstly means that he or she is attracted, and secondly, a customer can either bounce or continue in the conversion funnel (Barbieri, Silvestri and Lalmas, 2016). Since conversion predominantly is the main goal of a brand’s advertisement campaigns, the number of clicks can be seen as an important customer engagement indicator. Table 3 summarizes the variables that are used in this research to measure the level of customer engagement. Table 1: Levels of customer engagement Likes A post with many likes indicates that the content is of interest (Kietzmann et al., 2011) Comments A high number of comments on a post suggests that customers have taken their time to share their opinion (Luarn et al., 2015). Shares Sharing a post can be seen as the highest level of engagement (Kietzmann et al., 2011). Clicks Good content motivates customers to click on the post and endures their interaction with the brand (Andrew & Parsons, 2002).

(24)

24

4.3 Sequence strategy of posting

The previous literature chapter has discussed the concept of customer engagement, its multiple levels and how to measure it, but how can a brand marketer increase this customer engagement? In this literature chapter I discuss the history of marketing advertisement and the importance of media planning. Knowing this, five sequences will be discussed to give an answer to the question above. Earlier research has shown that the type of content can play a crucial role in posting strategies (De Vries et al., 2012; Luarn et al., 2015), but this research goes beyond the already known literature. In the next chapter, we discuss the perceived fit between the brand generated online content post and the brand fan page. Besides, I conceptualize five sequence strategies that a brand marketer can use to decide which variety of high fit and low fit posts achieve the highest level of customer engagement. 4.3.1 Perceived fit The right social media strategy makes it possible for a brand to broadcast digital campaigns to reach their customers. This strategy is predominantly focused on earned reach which occurs when customers themselves are taking over the broadcasting and transform in the channel itself by sharing the online content with their network (Edelman and Salsberg, 2010). The online content that is broadcasted by the brand doesn’t necessarily have to focus on the brand or its main product or service. This is what we call an online post with a low fit where the online content is not congruent with the brand. In 2015, T-Mobile launched their ‘Pets Unleashed’ campaign which gave people the fictive possibility to stay in touch with ‘their other family member’, their pets, and let them join T-Mobile. Where pets are not the main focus of the brand, the campaign earned 194 print and online news pieces and 52 unique broadcast segments across CNN, Fortune, Wall Street Journal and others (CP, 2015). The novelty and fun factor of this campaign made it go viral and the reach was tremendously. Although this was not congruent with their brand, in increased their brand awareness.

(25)

25 A second example, the strip figure Maleficent used her brand page to promote discounts for Disney Land which not only increased the reach of her online content, but also resolved in more Disney Land visitors who could see Maleficent in real life. Third, Shell promotes ‘free’ coffee on their Facebook brand page which you can win by sharing the content with your friends. The tagging and sharing increases the reach and the coffee can only be consumed at the Shell gas stations. Where coffee is not the main focus of Shell, it does increase the customer engagement of their online posts and lead customers to their gas stations which increase profits. Previous research has shown that the degree of perceived fit has an impact on the success of the brand extension (Kim and John, 2008). Also, the difference in content influences the number of likes, comments and shares (de Vries et al., 2012; Berger and Milkman, 2012). Therefore, I propose that also the variety, in terms of fit, of sequence in the postings strategy can have an impact on the level of customer engagement on the brand fan page. Theoretically, the concept of perceived fit is drowned from the brand extension literature, where the parent brand can decide to do a brand extension. Aaker and Keller (1990) have shown that this extended product category will increase the likelihood and forms more positive attitudes the moment this extension overlaps with the parent brand. This can be seen as a form of high fit. With the scope of this research, that would suggest that an online content post with a high fit between the post and the brand fan page, considering that positive content has a better chance to go viral (Berger and Milkman, 2012), will lead to a higher level of customer engagement. This is supported by the research of Buil, Chernatony and Hem (2009) who conclude that: ‘extensions close to the original brand that have a high degree of similarity with the existing product categories of the parent brand and maintain a coherent image will benefit from a more favorable attitude’ (p. 1315/1316). Both findings are consistent

(26)

26 with the research of Kim and John (2008) and Boush et al. (1987) who propose that customers respond more favorably if they are able to perceive a fit between the extension and the parent brand. On the other hand, using the brand fan page to broadcast online content incongruent with the brand, perceived as low fit, could be more effective and lead to more engagement since moderately incongruent content is expected to increase customer’s attention (Meyers-Levy and Tybout, 1989). The unexpected online content can lead to a response of the customer which is in line with the research of Rooderkerk and Pauwels (2016) regarding controversial content. In addition to this, Mandler (1982) argues that a more favorable effect will be generated when the content is moderated incongruent with the brand more than when its congruent or extreme in-congruent with the brand. Despite these standpoints, brands should take into account that a wrong extension could create extremely damaging associations with the brand that will not be reversible after the moment it has come online (Ries and Trout, 1981; Aaker and Keller, 1990). If the extension turns out to have a negatively impact, this will not only reduce interest of customers in your brand, but is simultaneously a waste of resources such as time and money. The concept of perceived fit discussed in this paragraph will take a central point in formulating the different strategies in paragraph 4.3.6. 4.3.2 The sequence of fit Considering the concept of perceived fit, a brand marketer should enroll a strategy that increases the customer engagement level of certain online content posts, short term, and increasing the customer engagement level of the brand fan page, long term. Although scholars have discussed the difference between low fit and high fit, the previous paragraphs reveal that there is no distinct line in which approach is best. Should a brand marketer only broadcast online content in line with the main brand or

(27)

27 should they broadcast as many pets related commercials on their brand page as possible to achieve the highest level of engagement? This research tries to answer such questions as mentioned above by focusing on the sequence of posts in regard to the perceived fit between a brand generated online content post and the brand fan page. In paragraph 4.3.6 several strategies will be formulated that will eventually contribute to both the gap in the current literature and management practicalities by answering the research question of this work. Focusing on the sequence of posts rather than a single post reveals the possibility to give a strategical advice at the end of this research. 4.3.3 Advertisement strategy To conceptualize different strategies that investigate which sequence of posts can be most effective regarding the increase of CE, some relevant literature in regard to media planning and advertising marketing will be discussed. For a brand marketer to choose the right advertisement strategy isn’t as easy as shooting a fish in a barrel. Especially, not in the current situation where the use of the web can amplify the reach of certain campaigns which can lead to an increased level of customer engagement or damaging associations (Ries and Trout, 1981; Aaker and Keller, 1990). The connection between the web advertising strategy and the corporate strategy should be aligned, especially for larger brands with a huge spread of their operations and influence (Boudreau and Watson, 2006). Besides, brands could also struggle with the well-known ‘global-local problem’ (Mooij and Hofstede, 2010). This struggle indicates the question of a brand to choose to either standardize their advertising for efficiency reasons and broadcast this to a broad public or adapt local habits to reach particular target groups. Also, understanding the culture of your target group is increasingly important (Mooij and Hofstede, 2010).

(28)

28 Furthermore, the standardization debate addressed by many scholars has been going on for more than four decades. The ones in favor of standardizing a brands’ advertising state that the rise of global market segments requests for a general strategy across markets (Melewar and Vemmervik, 2004). The major benefits that underpin the argument are economies of scale and a consistent brand image. However, scholars against the standardizing advertising strategy say that differences between cultures is so rooted that standardized marketing leads to a decrease of competitive advantage and lower sales (Melewar and Vemmervik, 2004). These kinds of fundamental issues are also addressed in earlier literature regarding the importance of the right media planning schedule (Naik, Mantrala, and Sawyer, 1998). 4.3.4 Media planning schedule Besides long discussed issues such as the global-local and standardization discussions, the moment of advertising the right content is also broadly debated. The right media planning schedule considers the moment of advertisement, weather you should continuously advertise about a product or service or if particular periods of advertising will be more effective (Naik et al., 1998). Also, a period of no advertising can enhance customers’ attention to ads (Grass and Wallace, 1969). This suggest that the period of no advertising will let the customer forget the particulars of a given ad and may feel that the ad is new when it is reintroduced (Calder and Sternthal, 1980). Naik et al. (1998) point out the continuous strategy and pulsing strategy. The first strategy concerns continuous advertising during the whole year whereas the latter is more often used at certain time slots to raise effectiveness of a campaign (Naik et al., 1998). In similar research the pulsing strategy regarding advertisement and budget allocating is also discussed (Feinberg, 1999; Park and Hahn, 1991). Both articles consider the pulsing strategy as most effective. A third strategy can be identified as ‘flighting’ in which the media effort is varied, but there are also times where there is no advertisement

(29)

29 at all (Klever, 2009). Klever (2009) defines continuous advertising as a constant schedule just like Naik et al., (1998) and the author defines pulsing as: ‘a variation of media efforts with no interruption, typically with a “heavy up” before prime selling times and less spending in off-season or anti-cyclical to the sales curve’ (p. 12). Although time is to be considered of importance in these strategies mentioned above, this research will only take into account the sequence as variable and time is seen as a constant variable. The strategies are mimicked and the logic is used to explain the five sequences in this research. The advertisement literature considers several strategies where there is continuous advertisement which, related to this research, can be continuously advertise with a low or high perceived fit between the broadcasted content and the brand page. More, the pulsing strategy illustrates that there can be a mix of intensity which, related to this research, can be a mix between low or high perceived online content posts. Contributing to this, the concepts of wearin, wearout and forgetting will be discussed. As the first paragraph in this section already mentions, a moment of no advertising can enhance the customers’ attention (Grass and Wallace, 1969). This break in advertising, resolves in the fact that the customer will ‘forget’ the particulars of an ad. Subsequently, the customer perceives the reintroduced ad as fresh which increases the number of cognitive responses (Calder and Sternthal, 1980). Also, this rejuvenating effect can simultaneously enhance the motivation to process the ad message (Pieters et al., 2002). The concept of forgetting plays a role in both the wearin and wearout phenomena that are produced by advertising repetition (Pechmann and Stewart, 1988). The authors define those two phenomena as follows: ‘An ad is said to have worn in at a particular level of repetition if, when consumers are exposed to it, it has a significant effect on them .. An ad that has worn in is said to have worn out at a particular level or repetition if, when consumers are exposed to it, it no longer has any

(30)

30 significant effect on them or even has a significant negative effect’ (1988, p. 286). The process of wearin happens immediately after being exposed to a marketing message. Both the research of Penchmann and Stewart (1988) and Calder and Sternthal (1980) illustrate that it takes three exposures for wearin to occur and for positive thoughts to outnumber the negative thoughts that can arise after being exposed. After these three exposures, the message begins to wearout in that negative thoughts increasingly outnumber the positive thoughts (Belch, 1982). The repeated exposure in this case may cause a campaign to lose its effectiveness (Calder and Sternhthal, 1980). The logic of these phenomena can be used in order to formulate five different sequences of perceived fit. The literature suggests that customers that know the advertisement are easily able to recall the information of the add and directly start with processing the message, which indicates the wearin phase. However, at one point, this recognition of the ads can negatively impact the attitude and advertisement response (Lehnert, Till and Carlson, 2013). Considering the five sequences of perceived fit in this research, the repetition of high fit posts, where the customer can start to recognize the ads, can lead to both wearin and wearout processes. More, the moment of no advertising, forgetting (Bass et al., 2007), can be mimicked by a repetition of high fit posts intervened by a low fit post. In this case, the low fit post can avoid the process of wearout by letting the customer focus on something else and subsequently enhance the customers’ attention by exposing him to a high fit post again. Based on the literature, this high fit post has the potential to achieve a high number of positive cognitive responses. In this research, the advertisement and media planning literature is mimicked and is taken as an explanation how the strategies in the next paragraph are conceptualized. Where the literature takes ‘time’ into account, this variable is not considered in this research, but is seen as a constant variable. This way the concepts of continuous and pulsing strategies and the concept of wearin, wearout and

(31)

31 forgetting advertising are not copied, but the logic of the concepts is used to define several strategies for the difference in the sequence of posts. The literature in this chapter shows that certain issues have already been there for a couple of decades and it’s not easy to enroll a digital media strategy without taking into account all possible consequences. Where the focus now is on social media marketing, even there it’s difficult to give an in-depth analysis about the best advertising strategy (Knoll, 2016). Seen this complexity, this research won’t be able to give the overall answer and solve all issues, but rather contribute to the literature with a more specific focus on the right sequence and perceived fit of online content posts. 4.3.5 Sequence of Perceived Fit: Strategies and hypothesis The literature in paragraph 4.3.4 has pointed out different strategies used in the advertising marketing which can be mimicked for this research. However, as explained, several variables such as time are not taken into account in this research, but the variety of sequence of the posts is the central focus point. Based upon the discussed literature five strategies, displayed in table 2, are conceptualized and will be tested individually during the methodology chapter.

(32)

32 Table2: Set of Possible Sequence Strategies Day of the week Continuous High Continuous Low

Pulsing High Pulsing Low Pulsing Mix

Monday High fit Low fit High fit Low fit High fit

Tuesday High fit Low fit Low fit High fit Low fit

Wednesday High fit Low fit High fit Low fit High fit

Thursday High fit Low fit High fit Low fit Low fit

Friday High fit Low fit Low fit High fit High fit

Saturday High fit Low fit High fit Low fit Low fit

As table 2 visualizes, there are five strategies which all have their unique sequence of posts regarding the perceived fit. Where high fit stands for a brand generated online content post that is perceived in line, or congruent, with the brand and low fit stands for a post with content that is perceived as incongruent with the brand. These strategies are developed for the current environment and disclose the earlier defined advertisement strategies. As Truong, McColl and Kitchen (2010) state: ‘Today there are concerns that advertising strategies based on models and experiences developed with traditional media may need to be reviewed in the digital market’ (p. 709). Since this research focusses on the current digital market, the strategies are also conceptualized as such. Where the literature has shown the increase use of social media, it’s more important than ever for brands to enroll the right social media marketing strategy (McMillan and Childers, 2017).

(33)

33 Investigating the strategies in table 2 will contribute to the daily practices of a brand marketer to make a relevant choice about which sequence of fit results in the highest level of customer engagement. The research of Aaker and Keller (1990), John and Kim (2008) and Buil et al., (2009) conclude that social media posts congruent with the brand will result in a higher level of customer engagement. However, the research of Meyers-Levy and Tybout (1989), Rooderkerk and Pauwels (2016) and Mandler (1982) suggests the opposite. They argue that incongruent content will lead to more attention, more positive attitudes and so resolves in a higher level of customer engagement. Although, several incongruent marketing campaigns have shown a high success rate to increase the reach of the online content, such like the Pets Unleashed campaign of T-Mobile, a sequence of continuous incongruent posts with the brand can be damaging and erode the brand’s core associations (Ries and Trout, 1981; Aaker and Keller, 1990). Additionally, the brand extension literature suggest that incongruent posts can lower customers’ attitude towards the brand and that customers respond more favorably if they are able to perceive a high fit between the extension and the parent brand (Kim and John, 2008; Boush et al., 1987). This leads to the formulation of the following hypothesis: H1: Continuous High strategy achieves a higher level of cumulated customer engagement than the Continuous Low strategy in a time period t. Besides these two continuous strategies, three pulsing strategies are formulated as well (Naik et al., 1998) where moments of high fit will be intervened with two posts of low fit (Pulsing High) and where consequent low fit posts will be intervened by two moments of high fit (Pulsing Low). The third pulsing strategy contains the sequence in which an even divided mix is conducted between high and low fit posts. Rooderkerk and Pauwels (2016) explain that topic ambiguity can resolve in uncertainty for the post recipients. This can happen when a brand chooses for the sequence of fit where only high fit posts

(34)

34 are broadcasted and suddenly a low fit post: Pulsing High. This post has the possibility to confuse the customer by the ambiguous broadcasting which evokes fewer comments. However, the authors also state that the online content with a high fit can be seen as practical use (Rooderkerk and Pauwel, 2016). The moment one or more low fit content posts are followed up by one high fit post these can be seen as practically useful, because the post tells something about the brand where the customers are interested in. This can resolve in a higher level of customer engagement (Rooderkerk and Pauwel, 2016). Also, moderately incongruent online content is expected to increase customers’ attention (Meyers-Levy and Tybout, 1989) and a more favorable effect will be generated when the content is moderated incongruent with the brand (Mandler, 1982). The literature on traditional advertising media planning, considering wearin, wearout and forgetting, can be aligned with those findings. This literature suggest that moments of no advertising can enhance the customers’ attention (Grass and Wallace, 1969) and the motivation to process the ad’s message (Pieters et al., 2002). The switch between high fit and low fit posts in the pulsing strategies can be indicated as a ‘break’ in the advertising by the customers. Subsequently, this break let the customer perceives the reintroduced high or low fit post as fresh which increases the number of cognitive responses (Calder and Sternthal, 1980). These cognitive responses can provoke emotions that lead to a higher level of customer engagement (Rooderkerk and Pauwel, 2016; Heath, Bell and Sternberg, 2001). Besides, the literature suggests that the process of wearout happens after being exposed to the same kind of ad more than three times (Penchmann and Stewart, 1988). The posts will than no longer have a significant effect on the customer and will lead to an increase of negative thoughts of the customer (Belch, 1982). The feeling of novelty of an ad enhances the attention of a customer which stimulates the wearin process (Lehnert et al., 2013). These phenomena can play a role in achieving a higher level of customer engagement.

(35)

35 Based on the brand extension literature in combination with the traditional media planning literature, the following two hypotheses can be formulated: H2: Pulsing strategies achieve a higher level of cumulated customer engagement than the Continuous High strategy in a time period t. H3: Pulsing strategies achieve a higher level of cumulated customer engagement than the Continuous Low strategy in a time period t. These two hypotheses take the continuous strategies as control groups. The level of customer engagement of all three pulsing strategies will be measured against both the Continuous High as Continuous Low sequence. Considering the different pulsing strategies, I also expect to find differences in the level of customer engagement between the pulsing strategies. However, the existing literature doesn’t possess enough elements to formulate a hypothesis for this. In addition to this, I’ve also decided not to include the comparison between the pulsing strategies as one variable and the continuous strategies as one variable because this will not clearly enough expose the differences between the sequences and reduce the overall information. The variety in the sequences is the main focus for this research. Table 3 summarizes the hypothesis. The next chapter will discuss how these hypotheses will be tested, which design is chosen for this research and what the strategically procedure has been. Afterwards, the results of those tests are proposed and discussed.

(36)

36 Table 3: Overview of hypothesis H1 Continuous High strategy achieves a higher level of cumulated customer engagement than the Continuous High strategy in a time period t. H2 Pulsing strategies achieve a higher level of cumulated customer engagement than the Continuous High strategy in a time period t. H3 Pulsing strategies achieve a higher level of cumulated customer engagement than the Continuous Low strategy in a time period t.

(37)

37

5. Methodology

In this chapter, the empirical part of the study will be provided. The methodology will take you chronologically through all steps that are taken before analyzing the results and hypothesis. At first, the research design, sample and procedure are discussed. Secondly, an overview of the measures, variables and stimuli will be provided. The third part highlights the statistical procedure of this research.

5.1 Research Design

This research used an online survey-based experiment to test the hypothesized relationships. This cross-sectional survey design analyzes data from a population at a certain moment in time which provides a way to examine causal relationships between the dependent and independent variables (Lewis & Saunders, 2012). This experimental design is characterized by both the random selection of participants and the random deviation of participants into the study groups (Campbell, Stanley & Gage, 1963). In this research, two control groups and three experimental groups are taken to test the hypothesis. The survey was conducted via Qualtrics, an online platform that Master students can use for free. This carrying out online, via the web, is a convenient way of gathering data which gained popularity (Couper, Traugott, and Lamias, 2001). It has several advantages such like low delivery costs, low entry time of data and multiple design options (Fan and Yan, 2010). The survey consisted of fixed response questions in which the answers where already formulated and in which the questions needed to be answered before it was possible to continue the survey. By providing a lively introduction text, visuals and images the survey tried to motivate the participants to finalize the survey.

5.2 Sample

The sample that is chosen in this research is a convenience sample. This sample is a type of a non-probability sampling method in which the sample is taken from a group in the population that is easy to

(38)

38 contact and reach (Ferber, 1977). Most participants are directly from my own network and are reached via e-mail, WhatsApp, Facebook and LinkedIn. Also, I used two websites: surveycircle.com and poll-pool.com where I uploaded my survey. These websites motivate you to fill in surveys of other students to make sure you get other participants assigned as well. This reciprocity helps many students to get enough participants. Since my network is predominantly Dutch, most participants of the survey are born and raised in the Netherlands. However, since also my internationally fellow master students could fill in the survey and I uploaded my survey at the two websites that are also used by other nationalities, a part of my participant group would be more internationally oriented. I could argue that because of the international character of the participant group the results of this research could be generalized to a wider audience. Though, the number of participants is still not big enough to generalize the results in any way (Green and Glasgow, 2006), this will be discussed in the limitations chapter as well. A minimum of 40 respondents per study group was set. From the 281 participants, I had to exclude 26 participants. 8 participants were excluded because they didn’t pass the trap question. This trap question was installed to see if participants where reading the questions or just going as fast as possible (Liu and Wronski, 2018). 6 more participants had to be excluded because they didn’t fill in their gender. And another 12 participants didn’t completely finish the survey. Although, Qualtrics should have filtered those participants already, there were still some left that needed to be excluded. A total of N=254 participants are used to test the hypothesized relationships.

5.3 Stimuli and measures

All participants of the survey have been exposed to a sequence of six Facebook posts of the brand Netflix. Those Facebook posts were simulated by the researcher and not taken from the Facebook page. In total, fifteen Facebook posts were developed using a scenario-based logic which were divided in seven High fit and eight Low fit posts, all identical to the original lay-out of Netflix’ Facebook posts.

(39)

39 Twelve of those fifteen posts were used to expose to the participants which are chosen based on a Simple T-test, this test is discussed in the next paragraph. See appendix 1 for the full overview of the simulated Facebook posts. All constructs were measured using a seven point Likert scale. To measure the likelihood of the participants to engage with an exposed Netflix Facebook post the same scale is used as in the research of Berger and Milkman (2012). This scale is a probability scale in which 1=extremely likely and 7=extremely unlikely. All respondents were asked what the probability is that you like, comment, share or click on the exposed Facebook post. The dependent variable customer engagement is measured with this seven point Likert scale. Every participant has seen six different Facebook post. These six Facebook posts could be part of five different sequences. This research focuses on the variety of sequence and more particular, which sequences achieve the highest level of customer engagement. Those six Facebook posts could be divided in the sequences shown in table 2 in the Literature Review. All twelve Facebook posts are labeled based on the content they expose. One of the Facebook posts is about the Comedy assortment of Netflix (high fit) where another Facebook post is about the Energy sector (low fit). Below, in table 4, all five sequences are now shown with the exact Facebook post. The full overview of the Facebook posts can be seen in appendix 1. Table 4: Variety of sequences including the developed posts Continuous High Comedy – Account – Casa – Agenda – Narcos – Free Continuous Low Franklin – Energy – Socks – Mark – Pets – LGBT Pulsing High Narcos – Mark – Account – Casa – Pets – Agenda Pulsing Low Energy – Casa – Pets – Franklin – Comedy – LGBT Pulsing Mix Casa – LGBT – Narcos – Socks – Free – Pets

(40)

40 In addition to this, the participants where asked after seeing the six Facebook posts what the probability is that they will: 1) visit the Netflix Facebook page by the end of next week? 2) click on Netflix Facebook posts to get more information by the end of next week? 3) like, comment or share a Netflix Facebook post by the end of next week? This is the same probability scale as Berger and Milkman (2012) use in their research. Answering positively to these answers would indicate that the showed sequence motivates the participant to engage with Netflix in the near future. This is also valuable to know for a brand and marketing manager since this indicator can be used to measure the level of customer engagement. These three questions, indicators, are computed to one variable: MotivationEngagement. Also, at the end of the questionnaire the participants where asked about the level of involvement with Netflix as a brand: 1) Netflix is a product that interests me, 2) watching Netflix products is enjoyable, 3) I enjoy discussing Netflix products with my friends, 4) Watching Netflix is satisfying and enjoyable. To measure this involvement the seven Likert scale (1=strongly disagree, 7=strongly agree) was used. The variable is called InvolvementTotal.

5.4 Quantitative pre-test

The pre-test was conducted via a Qualtrics questionnaire, just like the main test. As mentioned, fifteen Facebook posts were developed, divided between high fit and low fit posts. However, before using the posts it was necessary to test if those self-made Facebook posts also measured the right fit. 55 participants (N=55) have completed the pre-test and a minimum of 25 participants per post was set. All posts had minimum of N=25 and a maximum of N=29, this is also visualized in table 5. Table 5 also reveals more relevant information about the X_Total variables. These Total variables are computed by taking the likes, comments, shares and clicks of a particular Facebook post and compute this into a different variable which is X_Total. The last, Mean_Total, indicates the mean of

Increase your brand Customer Engagement : choose the right sequence of fit between your Online Content Posts and the Brand Fan Page