藉內省意見領袖優化社群媒體B2C2C 網路口碑行銷影響力 - 政大學術集成

全文

(1)國立政治大學資訊管理學系碩士學位論文. 藉內省意見領袖優化社群媒體 B2C2C 網路口碑行銷影響力. 政治大 Optimizing B2C2C 立 eWOM Likelihood in Social Media ‧. ‧ 國. 學. through the Introvert Leaders. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v. 指導教授：苑守慈博士研究生：黃志騰撰中華民國一○四年七月.

(2) Optimizing B2C2C eWOM Likelihood in Social Media through the Introvert Leaders 藉內省意見領袖優化社群媒體 B2C2C 網路口碑行銷影響力. by. Chih-Teng Huang. A Dissertation Submitted in Total Fulfillment of. 政治大 the Requirement for the Degree of 立. ‧ 國. 學. Master of Science. in. ‧. n. al. er. io. sit. y. Nat. Management information Systems. i Un. v. Supervisor: Soe-Tysr Yuan, Professor, MIS, NCCU. Ch. engchi. Department of Management Information Systems. NATIONAL CHENGCHI UNIVERSITY. July 2015. © Chih-Teng Huang 2015. 2.

(3) 謝辭在研究所的學術生活中，我從資訊管理範疇更深一層的接觸到了服務科學領域。這兩年的成長，我非常感謝我的論文指導教授－苑守慈老師，對我的疑問循循善誘，對論文的細節慢慢琢磨。除了理論架構必須嚴謹，也敦促我要有實作的能力，將系統開發出來後，更要求系統務必提供使用者最好的使用者操作體驗。在口試期間宋同正老師、林宛瑩老師給了我很多的指導。另外周思妤老師、呂伊婷老師在論文上也給了我很多的啟發。感謝這一路上所有給予我指導建議的學術前輩們，讓我受益良多。. 立. 政治大. 論文的完成，很感謝一起奮鬥的夥伴－楊維正和巫承安，一起腦力激盪解決. ‧ 國. 學. 各種問題，你們的態度是我的榜樣，感謝這兩年有你們相伴。SSRC 實驗室的學長. ‧. 姐、學弟妹們各個熱心助人，感謝席筠、威辰、韻平、庭毅、承豫、祖韵和曉晨，. sit. y. Nat. 以及所有政大資管所的同學、學長姐、學弟妹們在論文這條路上的協助。還有大. io. 了許多寶貴的資料讓這份研究更具價值。. n. al. Ch. engchi. er. 力支持我們的 Vocals 和茗熙坊，願意使用我們設計的新服務流程，讓我們蒐集. i Un. v. 最後我要非常感謝家人這兩年來的支持和體諒，在我追尋學術夢想的道路上給我滿滿的溫暖和祝福，讓我的學生生涯沒有後顧之憂。在我遭遇瓶頸的時刻給我安慰，在我有所求的時刻默默付出，在我疲憊的時刻給我鼓勵，讓我勇敢的追求成為一個更好的我，感謝你們！. 2015/07/30 黃志騰筆.

(4) 摘要在社群網路發展蓬勃的時代，各企業都要求行銷人員尋求在社群網路曝光最大化的方法。企業也渴望透過培養忠實顧客得到穩定的顧客關係;換句話說，企業無意識地在顧客參與度的方面進行著墨。除此之外，企業習慣於藉著整合(聯合)性行銷來增加企業曝光度以及減少行銷成本；但僅止於與利益關係夥伴的行銷而喪失了更進一步培養顧客關係的機會。在我們的研究中，我們發展了原創性的概念性架構去幫助企業完成三項目標：優化口碑行銷之可能性、發展顧客參與度、以及加強與合作夥伴整合性行銷的效益。本研究設計了顧客參與平台，提供. 治政企業以協助企業管理粉絲專頁、發展顧客參與度以及挑選適合的合作夥伴以進行大立整合性行銷。我們研究發現，內省意見領袖之顧客參與度與文章型態以及內容有 ‧ 國. 學. 顯著關係，他們較容易受到知識性文章內容所吸引；而一般 Facebook 使用者也. ‧. 容易因朋友分享文章而有更高度閱讀的意願。我們也藉由這些發現針對兩間企業. y. sit. io. n. al. er. 顧客參與深度。. Nat. 進行聯合行銷的個案研討，成功提高其 B2C2C 在 Facebook 上的口碑行銷廣度和. Ch. engchi. i Un. v. 關鍵字：社群媒體上的口碑行銷，顧客參與，內省意見領袖，品牌價值共創與整合行銷.

(5) Abstract In view of the explosive growth of social network, businesses want their marketers to find ways to be more ad exposure. Furthermore, businesses desire to get loyal customer for sustainable customer relationship; in other words, businesses are interested in developing customer engagement, even if they do not realize that they are so. Besides, businesses used to increase their brand exposure and decrease marketing cost through integrated marketing, but merely cooperating with stakeholders lost opportunity to further cultivate customer relations. In this study, we develop an original. 治政 conceptual framework to assist business to complete 大 three objectives: locally 立 optimizing eWOM likelihood, developing customer engagement, and enhancing ‧ 國. 學. effectiveness of integrated marketing with cooperation stakeholder. We design an. ‧. engagement site to help businesses choose their cooperation stakeholders, manage their. sit. y. Nat. fan page and further engage their customers on their fan page through our services. io. al. er. provided on engagement site. Our finding includes the insights of the significant engagement degree between the informative content article and introvert leaders, the. n. iv n C introvert leader to arouse their friends’ interest by the article, the method to enhance hen gchi U B2C2C eWOM through the introvert leaders and the practices of cooperating campaign between two businesses on Facebook.. Key words: Word of Mouth on Social media, Customer Engagement, Introvert Leader, Co-branding integrated marketing.

(6) INDEX CHAPTER 1 INTRODUCTION ............................................... 11 1.1 Background and Motivation .......................................................................... 11 1.2 Research Question ......................................................................................... 12 1.3 Research Method ........................................................................................... 14 1.4 Purpose and Contribution .............................................................................. 16 1.5 Content Organization ..................................................................................... 17. CHAPTER 2 LITERATURE REVIEW .................................... 19 2.1 eWOM............................................................................................................ 20 2.2 Customer Engagement Behavior ................................................................... 21 2.3 Introvert leader ............................................................................................... 25. 政治大 2.4 Stakeholder co-branding marketing ............................................................... 27 立. ‧ 國. 學. CHAPTER 3 MOTIVITATION APPLICATION ...................... 28 3.1 The Conceptual Framework of iEngagement ................................................ 28. ‧. 3.1.1 Situation – Organization and Eco-stakeholders .................................. 28 3.1.2 Organism – E-empowerment .............................................................. 29 3.1.3 Behavior – E-empowerment CEB ...................................................... 30 3.1.4 Consequence - Value conversion ........................................................ 31 3.2 The System Architecture of iEngagement ..................................................... 33 3.3 System Scenario ............................................................................................. 35. er. io. sit. y. Nat. al. n. iv n C h e n g.............................................. CHAPTER 4 METHODOLOGY 37 chi U 4.1 Conceptual Framework .................................................................................. 37 4.2 System Architecture ....................................................................................... 41 4.2.1 Post Measuring Module ...................................................................... 44 4.2.2 Introvert Leader Measuring Module ................................................... 44. CHAPTER 5 APPLICATION SCENARIO .............................. 48 5.1 An overview of system context ...................................................................... 48 5.2 The system journey of the application ........................................................... 49 5.3 Detailed Views of the iEngagement System Experience ............................... 51. CHAPTER 6 EVALUATION .................................................... 59 6.1 Propositions.................................................................................................... 59.

(7) 6.2 Assumptions ................................................................................................... 61 6.3 Experiments Design Details ........................................................................... 61 6.3.1 Data Collection Procedure Design ...................................................... 61 6.3.2 Design principles and details of questionnaire and case study ........... 63 6.4 Evaluation Results ......................................................................................... 75 6.5 Discussion and Additional Data ..................................................................... 84. CHAPTER 7 CONCLUSION ................................................... 90 7.1 Contributions.................................................................................................. 90 7.2 Managerial Implications ................................................................................ 92 7.3 Limitation and Future Works ......................................................................... 92. APPENDIX A ............................................................................ 94. 治. 政 APPENDIX B ............................................................................ 97 大. 立. ‧ 國. 學. APPENDIX C .......................................................................... 106 APPENDIX D .......................................................................... 107. ‧. REFERENCE ........................................................................... 108. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v. 7.

(8) TABLE INDEX Table 2.1 Three factors of customer engagement (Van Doorn et al. (2012)) ............... 22 Table 2.2 Worlds used to describe extroverts and introverts ....................................... 25 Table 6.1 Design Evaluation Methods (Source: Hevner et al., 2004) .......................... 59 Table 6.2. Details of the simulated businesses ............................................................. 63 Table 6.3 Service awareness ........................................................................................ 64 Table 6.4 Case study business information .................................................................. 75 Table 6.5 Descriptive statics of different content to introvert leader ........................... 77 Table 6.6 Descriptive statics of different content and article type to introvert leader . 77 Table 6.7 Data from Vocals Facebook fan page ........................................................... 81 Table 6.8 Interview of Ming Shi Fang ......................................................................... 82 Table 6.9 Descriptive statics of stakeholder cooperation’s user engagement .............. 83 Table 6.10 Descriptive statics of user engagement from the article source ................. 84. 政治大 Table 6.11 The summary of the entrepreneurs’ suggestion and opinion ...................... 85 立 ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v.

(9) FIGURE INDEX Figure 1.1 Hevner’s Design Science Research Framework (Hevner, 2004) ............... 17 Figure 2.1 Engagement Rate with Facebook Posts Worldwide, by Type (2011) ......... 24 Figure 3.1 iEngagement Conceptual Framework ........................................................ 28 Figure 3.2 iEngagement System Architecture ............................................................. 33 Figure 4.1 Conceptual model ....................................................................................... 38 Figure 4.2 System Architecture.................................................................................... 43 Figure 4.3 Actor’s role distributed by ρ and γ (Zubair Shafiq et al. 2012) .................. 46 Figure 4.4 Distribution of average shortest path length (Zubair Shafiq et al. 2012) ... 47 Figure 5.1 The system journey of the business user .................................................... 49 Figure 5.2 Business owner information registering page ............................................ 51 Figure 5.3 Service concept introduction page.............................................................. 52 Figure 5.4 Service detail introduction page ................................................................. 53. 政治大 Figure 5.5 Brand alliance page .................................................................................... 54 立 Figure 5.6 Campaign laboratory page .......................................................................... 55. ‧ 國. 學. Figure 5.7 Example of Facebook fan page post ........................................................... 56 Figure 5.8 Facebook fan page information in campaign thermometer page ............... 57. ‧. Figure 5.9 Facebook fan page information in information observation page .............. 57 Figure 6.1 The data collection procedure .................................................................... 62 Figure 6.2 Being not influenced by neighbors - positive part...................................... 66 Figure 6.3 Being not influenced by neighbors - negative part ..................................... 67 Figure 6.4 Inward interaction positive part .................................................................. 67 Figure 6.5 Inward interaction negative part ................................................................. 68 Figure 6.6.a Different content of articles ..................................................................... 69 Figure 6.6.b Different content of articles ..................................................................... 70 Figure 6.7.a Different types of informative articles ..................................................... 71. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v. Figure 6.7.b Different types of informative articles .................................................... 72 Figure 6.8.a Different imagery of informative articles ................................................ 73 Figure 6.8.b Different imagery of informative articles ................................................ 74 Figure 6.9 Introvert leaders’ analysis through our questionnaire ................................ 76 Figure 6.10 The behavior on Vocals Facebook fan page during campaign ................. 78 Figure 6.11 The behavior during June and July ........................................................... 78 Figure 6.12 The behavior of the post on 7/9 ................................................................ 79 Figure 6.13 The behavior of the campaign post .......................................................... 79 Figure 6.14 The trend of interaction rate in Vocals Facebook fan page ...................... 82 Figure 6.15 Differentiation between exiting approach and our approach of eWOM .. 88.

(10) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v. 10.

(11) CHAPTER 1 INTRODUCTION 1.1 Background and Motivation Busy pace of modern life, businesses often spend a lot of money on marketing activities to attract the attention of consumers. In addition to attracting the attention of consumers, businesses are increasingly focusing on maintaining customer relationships. Businesses are very eager for long-term and sustainable loyal customers and they rely on how businesses retain, sustain and nurture its customer base (Jenny et al., 2010). In the case of business resource being limited, which channel shall be chosen and how to. 政治大. optimize the benefit of marketing activities become big questions?. 立. In 2013, Taiwan’s online population spent 3 hours and 24 minutes a day, more than. ‧ 國. 學. 1 hour compared to the time spent on watching television (FIND, 2013). Further, the number of active users per month in Facebook being up to 14 million. ‧. (www.Facebook.com, 2013), the use of such a large number of people explains why. y. Nat. io. sit. businesses scramble to establish their own brand community on Facebook fan page.. n. al. er. Facebook fan page does offer a number of features to maintain customer relationships;. Ch. i Un. v. such businesses can post some information of product with text, photo or video, and. engchi. customers can use text response the post. However, do businesses have a two-way communication platform meaning to maintain a stable customer relationship? Current businesses are good at commodity-oriented marketing strategy, and their Facebook fan pages are too emphasizing on B2C’s information transference to improving customer engagement. Customer engagement, it is defined as “behaviors going beyond transactions and driven by motivation, and specifically refer to customer’s behavioral manifestations that have a brand or business focus (Van Doorn et al., 2010, p. 254). Customer more engaged in a business brand may be more of loyalty, thus the business can reduce the cost of maintaining customer relationship (Yeo et al., 11.

(12) 2005). Irena et al. (2013) addressed that in this new marketing era the terms engagement and participation became the central non-transactional constructs (i.e., going beyond purchase and driven by motivation) accordingly can be used to describe the nature of social platform participants' specific interactions and/or interactive experiences.. 1.2 Research Question Customer engagement behaviors can contribute to the business’s performance in two ways: (1) CEBs in interactions with business and their employees result in cost advantages for business while ensuring rewarding customer experiences (Hoyer et al.. 治政 2010) and (2) CEBs in interactions with other customers 大 affect other customers’ 立 attitudes and behaviors toward business (Gupta and Harris 2010). In the other words, ‧ 國. 學. customer can engage business through B2C’s interaction and C2C’s interaction.. ‧. However, the businesses’ traditional social marketing practices have no concept of. sit. y. Nat. customer engagement, and this leads to high turnover of customer (Divol, Edelman,. io. al. er. and Sarrazin, 2012). This is more than business using these following three criteria to. n. measure a fan page management being good or bad: counts of like, response, and share on each post. These criteria. iv n C are h the outermost behavior e n g c h i U on. social platforms like. Facebook and show only partial B2C word of mouth (WOM, defined as person-toperson communicating between a receiver and a sender, and the person may be friends, family members or strangers behavior (Jose´ et al., 2012).) Through these criteria we still do not know the degree of customer engagement and the degree of interaction between customer and customer. We must let fans have a willingness to revisit the fan page; after research, we believe that the best way is through introvert leader (Zubair Shafiq et al., 2012). Introvert leaders, as mentioned by Zubair Shafiq et al. (2012), refer to those who would well digest prior to responding in a social media when they get information or feedback from their neighbors. 12.

(13) Between businesses and businesses, there exist competitive relationships, these relationships don't let customers flow smoothly between businesses on social platforms. No one will let competitor expose on their own fan page. Nonetheless, when information cannot be streamed on social platform, businesses will lose the opportunity to make their own exposure. Thus, it makes customers struggle to search information about the product they need on the Internet, resulting in low degree of product / service complexity and low process simplification for and business’s products (Martin et al., 2014). In addition, WOM is also an important task for business. If the business only. 治政 concerned about B2C WOM, it would be not completely 大 trusted by customers (Jesse 立 and Orit, 2008); if the business hopes that C2C WOM is able to achieve all the publicity ‧ 國. 學. effect, but the results have no control over the business and there’s no guarantee that. ‧. C2C WOM will meet the minds of business. The only way to solve this problem is. sit. y. Nat. through B2C2C WOM to operate the most efficient social marketing. We define. io. er. B2C2C WOM as business engages customers through introvert leaders. Jose et al. (2012) proposed that it has the same importance between “in-group”. al. n. iv n C (WOM with close friends or family)hand “out-group” U e n g c h i (WOM with persons who are out. of one’s social circle) WOM marketing. Hennig-Thurau et al. (2010) described the impact of new media on customer relationship as playing pinball (marketing managers continue to guide “marketing ball” path with “flippers,” but the ball will not always go where it is intended to and the slightest miscue can be amplified into a catastrophic crisis.), implying the difficulty of controlling WOM on new media (eWOM), and the importance of C2C WOM. Unfortunately, current Facebook fan page managers focus only on B2C WOM, and therefore “Like-oriented” fan page operating strategy highlights the lack of measuring of customer engagement and C2C WOM’s effect. Nowadays, businesses used to collect likes for fan page after hosting an event. However, 13.

(14) Facebook deleted the mechanism for business to collect likes after the events to encourage customers' clicking “like” on their fan page, and it may have great impact on the existing models of managing fan page (Facebook, 2014). Thus, the BC2C eWOM of the actual characteristics still needs to be studied and verified. Katrien et al. (2013) had done a research about managing engagement behaviors in a network of customers and stakeholders. They mentioned that the stakeholders include business and their employees, customers and their family. However, the common joint marketing stakeholders are the other businesses, and mistrust that exists between them would be one of the reasons to obstruct mutual cooperation. But these obstructs still. 治政 cannot be a reason to not to cooperate with each other大 among businesses. Besharat et 立 al. (2013) pointed out that co-brand is a business trend: a successful co-brand can reduce ‧ 國. 學. marketing costs for both businesses and can produce win-win result. According to. ‧. above, we believe that business cooperates with different kinds of businesses will have. sit. y. Nat. different degrees of impact on customer engagement, so we will provide some service. io. other.. al. er. on an engagement site that can let them have a good cooperated experience with each. n. iv n C In this research, we investigate h two research questions. e n g c h i U The first is how to execute. and optimize B2C2C eWOM through the influence of introvert leaders. The second is how we establish a platform for businesses to cooperate to each other in the smooth way through imagery matching mechanism. This research proposes a resolution method and system in response to two questions and the method has two main components – business cooperation and B2C2C eWOM – that will be briefed later.. 1.3 Research Method In our system, there are some components: business, business’s Facebook fan page, engagement site. Engagement site is a website that creates and sends some information 14.

(15) or communication to customers and prospective customers (Rodney et al., 2013). Businesses can put link of their brand website or fan page on engagement site that can redirect customer to visit their brand website or fan page. And engagement site also stores that communication on itself so that it can provide the data that we can analyze and improve in terms of the following proposed services from the business point of view, an engagement site provides business with (1) finding out introvert leaders on their Facebook fan page, let business realize their main engagement target, (2) measuring the eWOM of Facebook fan page and each post, let business know our mechanism is effective, (3) picking out the great eWOM performance article, let it be. 治政 the reference for business to design the article in the future. 大 We will explain in the next 立 paragraph of the value of these three services. ‧ 國. 學. First, choosing joint marketing scope allows business to choose their cooperation. ‧. partner categorized by geographic factor or product/service type. These are two of the. sit. y. Nat. most common practices for businesses selecting cooperation partners. How do. io. al. er. businesses find the suitable cooperation partners? Through engagement site, Facebook community performance will be used as a reference. According to Ali et al. (2014),. n. iv n C brands tend to find the higher population to perform co-branding, so we need to h e nbrand gchi U analyze the characteristic of their fan page for matching each other. Second, checking eWOM measurements is defined by counts of like, response, and participating in the discussion of proportion (i.e., the extent of the discussion of the post exceeds the previous level or not). We believe comprehensive measurement can avoid the disadvantage of counting only the counts of like. Third, in order to add to B2C eWOM by managing C2C eWOM, we provide business’s Facebook fan page manager with appropriate materials for posting. After businesses finish using our services on engagement site, we will refer their eWOM performance to improve our service, providing some more appropriate posting 15.

(16) material. In this study, the eWOM performance is defined as follows. The eWOM performance is the attractiveness due to extent of an article. There are several data that we can collect on fan page articles, and these data are also that Facebook uses to measure an article’s popularity to decide whether to let this article appear on fans’ wall (Facebook, 2010). These data are “like”, “share”, and “response” on the article. So we will collect these to measure our eWOM performance.. 1.4 Purpose and Contribution The aim of this research is to optimize B2C2C’s eWOM likelihood on social media.. 治政 Our engagement site provides Facebook fan page managers 大 with new measurement of 立 social marketing (B2C2C eWOM) and the posting materials they need. Through ‧ 國. 學. B2C2C system we build, posts on fan page will meet Facebook’s ranking criteria so. ‧. that we can make the post have higher chance to expose on their fans’ wall. Besides, in. sit. y. Nat. the past the way to attract fans is more difficult to learn how to reach the target customer. io. al. er. base. Now, through contacting with introvert leaders, who have influence and effect their neighbors by inward interaction, we can attract the target customers (i.e., those. n. iv n C fans who are attracted by introvert in private message box with h leaders). e n g cMessage hi U friends is one of the most common inward interaction on Facebook, it is hard to trace by crawler, but this information propagation way greatly decreases the commercial feeling to message receivers. Therefore, through the introvert leader, the information dissemination are often better accepted by others. Further, through joint marketing service, business will be easier to find the appropriate cooperation partners and establish positive ecosystem. Their cooperation stakeholders are suggested through our system that is able to match each business by the characteristic of business, the type of goods and area of location. We believe our 16.

(17) system can improve business’s B2C2C eWOM performance and business will acquire long-term and sustainable loyal customers.. 1.5 Content Organization The content of the research is divided into four main sections and followed by the design science research framework (Hevner et al., 2004) (See Figure 1.1)., which aims to define the environment of research problem, to apply the foundations and methodologies in knowledge base, to build the IT artifacts to solve the relevant problem and relate them in relevance and rigor.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v. Figure 1.1 Hevner’s Design Science Research Framework (Hevner, 2004). . Chapter 1 – Introduction This chapter elaborates the current environment to provide an overall understanding of research background, motives, and purposes. By reading this 17.

(18) chapter, readers would have a vivid understanding about the current state of quo in order to define the problem space and the business needs. . Chapter 2 – Literature Review In this chapter, the literature review depicts the foundations and methodologies in knowledge base. The relationship between introvert leader and customer engagement and stakeholders are elaborated so as to clearly explain the existing knowledge to organizational problems. Through discussing the possible relationships in these issues, these could response to address important unsolved problems in Chapter 1 and to relate to the IT artifacts design described in Chapter4.. the iEngagement (for. 學. ‧ 國. . 治政 Chapter 3 – Motivation Application 大立 Establish the motivational conceptual framework of. enhancing customer engagement) service system. Describe how the system. Chapter 4 – Methodology. sit. y. Nat. . ‧. operate to improve customer engagement.. io. al. er. Establish the conceptual model framework of research which is followed by. n. developing the system. . Chapter 5 – Conclusion. Ch. engchi. i Un. v. This chapter elaborates conclusions and future work of this research.. 18.

(19) CHAPTER 2 LITERATURE REVIEW. In this chapter, we are going to discuss the foundations and methodologies as the background knowledge of our research. The review of WOM, customer engagement, introvert leader and stakeholder co-branding marketing are elaborated in order to find the theoretical basis as the required rigor to sustain the research. (1) eWOM: eWOM is one of the most common marketing methods for business to spreading their reputation, and using Facebook is one of the most popular. We think. 政治大 effectiveness about transferring the information to customer. 立. if we can grasp the properties of the eWOM, then we can have the better eWOM. ‧ 國. 學. (2) Customer engagement behavior: How to get loyalty customer has been a very critical goal to business if they attach importance to customer relations. Some. ‧. businesses used coupon to attract customers, but they came for what and then they. sit. y. Nat. will leave for what. We think only when we get customers’ heart, we can get loyalty. n. al. er. io. customer and co-create value with them.. i Un. v. (3) Introvert leader: We knew that opinion leaders can enhance eWOM power, but. Ch. engchi. different kinds of opinion leader lead to different results. Understanding them may help us identify the factors to optimize eWOM. (4) Stakeholder co-branding marketing: Some businesses hope spreading their reputation through operating co-branding marketing, but there are some failures. If we can sort out the correct way to perform on social media, it will benefit businesses a lot.. 19.

(20) 2.1 eWOM When the popularity of the Internet and its speed is increasing, more and more online applications appear. Social networking is one type of online application that has grown rapidly in prevalence in recent years (Tiffany et al., 2009). Besides more and more users on social media, and it can also reflect the real word life style more or less. Muniz and O’guinn (2001) defined brand community as “a specialized, nongeographically bound community, based on a structured set of social relationships among admirers of a brand”. Further, Nicole et al. (2007) also believed that Facebook represents an understudied offline to online trend in that it originally primarily served. 治政 a geographically-bound community. Above literature 大meant Facebook community 立 similar to the real world, and more and more people spend more time on online life. It ‧ 國. 學. is not difficult to understand why businesses spend so many efforts on online marketing.. ‧. Many businesses choose eWOM as their social marketing strategy, because research. sit. y. Nat. indicated that eWOM may have higher credibility, empathy and relevance to customers. io. al. However, operating eWOM has a great risk.. er. than marketer-created sources of information on the Web (Bickart and Schindler, 2001).. n. iv n C Hennig-Thurau et al. (2010) took marketing” metaphor “pinball game.” h e“eWOM ngchi U. This analogy illustrates the difficulties of operating eWOM marketing that we cannot make 100% accurate prediction of the direction of public opinions. This study identified new media phenomena, and customer’s interaction on those new media. It illustrated the complexity of eWOM but how to improve eWOM’s operating is scanty. In our research, we are going to solve this difficulty through operating introvert leaders’ eWOM marketing. Our method will make the “pinball” go on the way that we want it to go. Besides, Jose´ et al. (2012) pointed out that there some difference between ingroup WOM and out-group WOM. The group, individual’s network of friends, often 20.

(21) has the similar life circle and share the information to each other. In-group WOM makes a substantial overlap of information already exists among the group with same members; and out-group WOM means an individual contacts the people who do not know the information. It gave us a complete exposition in the internet usage motivation and eWOM in which users’ behavior is driven by mood enhancement for social interaction and experiential learning.. Facebook also uses the similar way to filter out the article. appropriated to every user. Each user has a relatively high probability to get the article forwarded from in-group WOM. However, for user business fan page is out-group neighbor and it makes eWOM diffused from fan page to user more difficult.. 治政 Richard et al. (2011) gave us an integrated marketing 大 example, “We’re All Fans”. 立 The advertising agency had used social media, traditional channel, and some different ‧ 國. 學. device to execute “We’re All Fans.” And authors shared their observations and got a. ‧. concept of ecosystem for co-creation online marketing. In the study, scholars provided. sit. y. Nat. some guidelines to us for operating a co-creating online marketing and we will try to. io. al. er. implement these suggestions: (1) visualize the ecosystem, (2) identify and track key performance indicators, (3) being with your story, (4) social media does not require. n. iv n C elaborate budgets, (5) being unique. includes stakeholders as one of the h Our e nresearch gchi U important factors in optimizing eWOM performance.. We will go to explore how the. two businesses collaborate on the community platform.. 2.2 Customer Engagement Behavior Many scholars have such same views: if business wants to get loyalty customers, they may focus on managing customer relationship, and they may be better aware of the concept of customer engagement. Customer engagement is sometimes used to denote the highest form of loyalty (Bowden, 2009; Roberts and Alpert, 2010).The common mean of customer relation is that everything is customer-oriented, not 21.

(22) business-oriented or product-oriented. However, the concept is still not clear enough so far to be able to create a loyal customer through contacting customers on social media. Van Doorn et al. (2010) considered that customer engagement behaviors needs to go beyond transactions, and it may be resulting from motivational drivers. Lusch and Vargo (2006) thought that customer co-creation that involves the customer participation in the creation of the core offering itself. It can occur through shared inventiveness, codesign, or shared production of related goods or service. What is the basis of customer engagement is also important. Van Doorn et al. (2012) put the affecting factors of customer engagement into three dimensions to explain: (1). 治政 business-based factors, (2) customer-based factors, and 大(3) context-based factors (see 立 Table 2.1). These are three main factors to affect customer engagement. In Facebook, ‧ 國. 學. customer-based and context-based factors play important role. Our research will. ‧. implement architecture to let business more comfortable to enhance their customer. io. al. er. based factor (article posted by introvert leaders).. sit. y. Nat. through customer-based factor (interactions between fan and fan page) and context-. n. iv n C Table 2.1 Three factors of customer (Van Doorn et al. (2012)) h e n gengagement chi U Affecting factors of customer engagement Characteristics of the brand—actual and perceived by the customer—can strongly influence CEBs. First, brands with high reputation or high levels of brand Business-based factors equity are likely to engender higher levels of positive CEB. However, in cases of failure, the negative fallout in terms of CEB may be higher as well.. 22.

(23) Customers may have specific consumption goals such as maximizing consumption benefits or maximizing relational benefits. The goals themselves Customer-based factors. can influence how the brand is used and consequently how customers engage with the brand. Sometimes the goals may be unrelated to the brand or product experience. Context-based factors affecting CEBs may largely the political, economic/environmental, 治政大 aspects (P.E.S.T.) of the social and technological arise from. Context-based factors. 立. society within which business and. customers exist.. ‧ 國. 學. Competitors and their actions also create a. strong. ‧. contextual force affecting customer engagement.. sit. y. Nat. io. al. er. Hennig-Thurau et al. (2012) gave us a concept of creating customer engagement on online social communities in three steps: (1) understanding consumers, (2) consumer. n. iv n C interactions, and (3) measuring customers/ outcomes. This research gave us h e n gcustomer chi U an example of consumer support forums, and we are going to apply to Facebook communication: (1) finding introvert leader, (2) interaction between fan and fan page, (3) measuring eWOM performance on fan page. For creating customer engagement behavior, we tend to stimulate the engagement model for self-reinforcing. In addition to the above-mentioned conditions, there are still some factors that can also have great influence on community engagement: social benefits, entertainment benefits and satisfaction (Johanna et al., 2012). Entertainment benefits in turn mediated the influence of both community and transactional behaviors on satisfaction and loyalty, 23.

(24) whereas economic benefits such as bonuses or lotteries had no influence on either satisfaction or loyalty. Scholars also pointed out that being entertained in the community was more important for the brand relationship outcome than the social activities of, for example, getting to know other community members, staying in touch with or helping community members. Such member-to-member activities have been shown to be very important in other online communities (De Valck et al., 2009). Businesses can get feedback from those reactions of fans. Godes et al. (2009) even mentioned that businesses benefit when users like their content because it encourages customer-customer and business-customer interaction, gauges the popularity of their. 治政 posts, and allows users to provide their personal endorsements. 大 Besides, content type 立 also has influence on customer engagement (Figure 2.1). We can enhance customer ‧ 國. 學. engagement through spreading specific content type as well.. ‧. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v. Figure 2.1 Engagement Rate with Facebook Posts Worldwide, by Type (2011) Rodney et al. (2013) declare an idea of engagement site. Engagement site is for marketing a business on Internet and for optimizing and improving the business’s visibility on search results pages of conventional Internet search engines that find out the most relevant result. We think this idea can have its analogy on social marketing as well. And we hope we can provide business with their prospective customers through social media channels. Businesses often market themselves on internet through these 24.

(25) channels. We think we can collect such channels’ marketing result data and selfreinforce our engagement model. 2.3 Introvert leader Zubair Shafiq et al. (2012) grouped users of online social network into four categories: Introvert Leaders, Extrovert Leaders, Followers, and Neutrals. Leaders are highly influential on those of others and followers are users whose opinions are highly influenced by those of leaders. As for neutrals, their outward interactions are not related to their neighbors or interaction in previous time. Ronald S. Burt (1999) points out those opinion leaders are more precisely opinion brokers who carry information across the. 治政 social boundaries between groups. However, for the leader 大 qualities, there exist two 立 different kinds of properties, extrovert and introvert (Hirsh, S.K. and Kummerow, J.M., ‧ 國. 學. 1987.) In this study, we need to examine which kind of leaders is more in line with the. ‧. objective of our research.. sit. y. Nat. Literatures referred that extroverts are energized by interacting with other people,. io. er. while introverts are renewed by being by themselves. Extroverts prefer the outside. al. world of people and things, while introverts enjoy the inner world of concepts and ideas.. n. iv n C Scholars also compiled a table about used to describe extroverts and h e nthatg worlds chi U introverts (see Table 2.2) (Hirsh, S.K. and Kummerow, J.M., 1987). Table 2.2 Worlds used to describe extroverts and introverts Worlds used to describe extroverts and introverts I. Extrovert. II. Introvert. external. internal. outside thrust. inside pull. blurt out a comment. keep in comments. breadth. depth 25.

(26) involved with people, things. interested in ideas/thoughts. interaction. concentration. action. reflection. do-think-do. think-do-think. Based on above literatures, we tend to choose introvert leaders as our research objective. Introverts have less outward interactions but their properties of depth, concentration, and think-do-think (they will have careful consideration before their. 政治大 neighbors. Private message is one of the inward interaction on Facebook, and when 立 behavior occurred, and will be self-reflection after) will have deeper influence on their. ‧ 國. 學. introvert leader forwards the fan page article trough private message and this will make the article less commercial feeling to the users. Besides, from chapter 2.1 we have. ‧. mentioned that business fan page is an out-group neighbor to user, thus we consider. sit. y. Nat. that we can break the boundary through the introvert leader. As the ability to spread the. io. al. er. message through opinion leaders, and we hope they can bring the brand story into each. iv n i U completed eal.n (2012) g c h have. n. user's living area. Therefore, our research may find out introverts and optimize their eWOM likelihood. Zubair. C et Shafiq h. a high degree of. completion research of identifying leaders and followers in online social networks. We will apply the method mentioned to this study and do further research. First, scholars denote friendship graph as a matrix and it stores users’ friendship ties in it. Second, scholars used FJ influence model (Friedkin and Johnsen, 1990) to analyze users’ interaction behaviors Friedkin and Johnsen, 1999). FJ influenced model can show the user’s interaction influenced by their neighbors. Third, they evaluated FJ influence model and tried to use this model to analyze leaders’ properties for distinguish extroverts and introverts. So far, we get the method for finding out introverts and we 26.

(27) are going to optimize their eWOM likelihoods.. 2.4 Stakeholder co-branding marketing A co-brand is defined as the placement of two brand names on a new or perceptually improved product (Kotler and Armstrong, 2013). The use of co-branding has grown by 40 percent annually (Besharat & Langan 2013). Kohli et.al. (2003) noted that co-branding is a special case of brand extension in which two brands are extended to a new product or service. Therefore, both co-branding and brand extensions raise the same basic issues, namely, how brand equity transfers to the new product and how the. 治政 new product subsequently has an impact on brand equity. 大Scholars further noted that a 立 co-branding situation in which each brand makes a significant contribution to the co‧ 國. 學. brand’s core benefits representing “core” complementarity. However, it is not always. ‧. easy to know whether a co-brand involves core complementarity. Thus, we think we. sit. y. Nat. can evaluate the co-brand performance through introvert leaders’ reaction. If introverts. io. n. al. er. react to the articles of co-branding, it will be a positive feedback.. Ch. engchi. i Un. v. 27.

(28) CHAPTER 3 MOTIVITATION APPLICATION. The purpose of this chapter is to describe an overview of our whole research project – “iEngagement”. We will first introduce the conceptual framework of iEngagement research project. Next section describes the system architecture of iEngagement, and the last is the scenario of iEngagement platform.. 3.1 The Conceptual Framework of iEngagement. 政治大. In this study, we identify the CEB stimulation process as four components,. 立. including (1) situation, (2) organism, (3) behavior and (4) consequence (Chou & Yuan,. ‧ 國. 學. 2014) and will further discuss them in the following four subsections.. ‧. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v. Figure 3.1 iEngagement Conceptual Framework 3.1.1 Situation – Organization and Eco-stakeholders Nowadays in digital service ecosystems, there are not only fast-paced, but also turbulent (El Sawy & Francis, 2013). The so-called turbulence is a causal texture of the environment that stems from complex interconnectedness between players (Selsky et al. 2007). The fates of all players are intertwined with that of the ecosystem, competitors 28.

(29) must often work together in “co-opetition”, for example to establish technical standards or common platforms. Hence, there has been a growing recognition among scholars and practitioners that service-dominant logic of value creation in the marketing field (Vargo and Lusch 2004, 2008). This new theory holds that value is co-created by customers and a network of firms and other actors. This means the individual must compete as a value-adder and that the number of competitors may be quite different in a value network to those in a value chain. This combination of cooperative and competitive processes has been termed “co-opetition” (Bengtsson and Kock, 1999).. 治政大 3.1.2 Organism – E-empowerment 立 Recent years have substantial increase of mobile, wireless technology, and new ‧ 國. 學. media channels such as Facebook, Myspace, Twitter, YouTube, and Google, which. ‧. enable customers to take a more important role as market players and reach (and be. sit. y. Nat. reached by) almost everyone, anywhere and anytime. However, after the rise of new. io. al. er. media, the research regarding consumers with extensive options for actively providing information on service have been received attention (Hennig-Thurau et al., 2010).. n. iv n C Between both sides interaction, there some variables not previously appeared. h e has ngchi U. Different from traditional role of Internet channel, new media allow consumers to reach (and be reached by) other consumers and companies almost anywhere at any time through their mobile devices. That is, new media also empowered consumers to promote and distribute their own offers (Hennig-Thurau et al., 2010). In this study, we refer e-empowerment as to the empowerment perceived by customers on the new media toward the business.. 29.

(30) 3.1.3 Behavior – E-empowerment CEB Drawing from resource exchange theory and affect theory of social exchange, Verleye et al. (2013) argued that managerial processes that generate positive customer affect toward the business result in CEBs that benefit the business and its cooperation stakeholders. Resource exchange theory (Foa 1971) holds that people exchange resources that are similar in terms of concreteness and particularism. In line with social exchange theory (Blau 2004), Verleye et al. (2013) labeled exchanges beyond mere economic obligations as “social exchanges”. The affect theory of social exchange (Lawler 2006) holds that social exchanges are driven by customer affect. Based on this. 治政 theory, Verleye et al. (2013) hypothesized that higher levels 大 of customer affect toward 立 the business (i.e., positive feelings toward the business) increase customers’ likelihood ‧ 國. 學. to show CEBs that benefit the business and its cooperation stakeholders. Social. sit. y. Nat. increasing customer engagement and eWOM.. ‧. exchanges theory supports our research to benefit both business and customers from. io. al. er. The importance of e-empowerment perceived by customers toward the business reflects the embeddedness of CEBs in a broader network of customer, business and. n. iv n C other stakeholders. Customers havehseveral opportunities e n g c h i U to show their engagement to. the business and its eco-stakeholders, since they increasingly participate in the creation, production, and delivery of service (Zeithaml, Bitner, and Gremler 2009). In an increasingly networked society, customers cannot only show CEBs in interactions with businesses and their employees. Customers can also show CEBs in C2C interactions by helping other customers and spreading positive eWOM behaviors. Therefore, we identify CEBs into two types of engaged objects: C2B&C2S and C2C CEBs. Customers usually exhibit three types of C2B&C2S CEBs in interactions with business and their stakeholders: cooperation, feedback, and compliance (Verleye et al., 2013). In contexts where CEBs are embedded in a broader network of customers, business and 30.

(31) stakeholders, we hypothesize that customers’ affect toward the organization and consequently their likelihood to show CEBs depends on overall e-empowerment perceived by customers. Drawing from role theory (Kahn et al. 1964), Verleye et al. (2013) assumed that customers’ willingness to show CEBs that benefit the business and its stakeholders depends not only on customers’ affect toward the business but also on their role readiness. In line with Verleye et al. (2013), we define customer role readiness as the degree to which customers feel prepared for encounters with the organization in terms of feeling confident and having the appropriate knowledge and skills. Previous research. 治政 merely demonstrated that customer role readiness affects 大 their compliance with the 立 organizational rules and procedures in various industries (Auh et al. 2007; Meuter et al. ‧ 國. 學. 2005).. ‧. Verleye et al. (2013) proposed that not only customer compliance but also other. sit. y. Nat. CEBs depend on customer role readiness. Customers who do not have the appropriate. io. al. er. knowledge or skills for encounters with the organization might be less willing to feedback to service system, or give suggestions for service improvement. Thus, we. n. iv n C propose that e-empowerment perceived customer increases customer role readiness, h e by ngchi U resulting in higher levels of CEBs.. 3.1.4 Consequence - Value conversion From the El Sawy and Francis’ (2013) observation, with the shifts in digital platforms and network values outlined above, the boundaries of the organization much more porous and it will be more difficult to define where the organization ends and the other parts of the ecosystem begin. Open innovation will likely be a dominant mode of operations as new products and services need to come to market more quickly for diverse customers. In the meanwhile, the notion of presumption will take hold as 31.

(32) consumers of services and products engage in their production through processes that we are already seeing in phenomena such as user-generated content. The value of a customer’s contribution to initiating new product and service innovation ideas needs to be included as a component of CE value. With such customer participation, manufacturers have the potential to enhance product innovation and to speed up the development process, both of which are key objectives of managers to lower costs and improve market acceptance of new offerings. As a result, the extent to which customers are willing to engage in conversations (with other customers as well as the business) can significantly influence a business’s value, especially as it affects. 治政 what customers are prepared to tell others, and what insights 大 they are willing to provide 立 businesses regarding product development and improvement. ‧ 國. 學. Thus, in a digital service ecosystem, value is co-created, co-converted, and co-. ‧. captured together with the different players in the ecosystem: customers, competitors,. sit. y. Nat. complementors, and community. Thus, organization in keystone positions in the. io. er. ecosystem may choose to leave many activities of value creation to others in the ecosystem, while choosing to focus on creating value that is critical to the ecosystem’s. al. n. iv n C prosperity. Therefore, we argue thathCEBs toward theU e n g c h i organization increase the extent of value co-creation by customers, continuously resulting higher levels service offerings of organization and eco-stakeholders. There is a great potential for the use of digital spaces for value creation and extraction by businesses in the service ecosystem, due to the intrinsic nature of digitality: First it creates a new embodiment for value creation (social media, mobility etc..), as an extension of the physical media; Second, it allows the acceleration of links among different spaces (El Sawy & Francis, 2013).. 32.

(33) 3.2 The System Architecture of iEngagement. User Module. Support Data. CEB Cocreation. Stakeholder. Social media. Brand Partnership Module. Identification Module Social Media Module Recommender System Module Search Engine Module. 立. Recommender System. CEB Collection Module. Search Engine. 政治大 Feedback. ‧ 國. 學. Figure 3.2 iEngagement System Architecture. ‧. Our system intends to provide services to help the business to do the customer. sit. y. Nat. empowerment of customer’s CEB. We use stakeholder identification module to identify. io. al. er. the current situation coopetition, three new media module to e-empower the customers, through CEB data from CEB collection module to co-create by brand partnership. n. iv n C module. The following describes inhmore detail of those e n g c h i U six modules:. 1. Stakeholder Identification Module: The main objective of this module is to give recommendation of stakeholder that they can cooperate with to the business using data from the support database. Business nowadays may confused and have little idea of what stakeholder can they approach and partner with, since most of the business on the new media they might have never heard of them. 2. Social Media Module: The main objective of this module is to help the business to optimize the possibility of consumers to do the content like, share and feedback. By spreading through social relationship, it can give both 33.

(34) increasing the customer affect according to social exchange theory and the degree of their acceptance as customer role readiness. At the end, give rise to the C2C and B2C CEB. 3. Search Engine Module: Search engine is one of the most popular way for customer to search for the information. During the process of searching, the user has a higher willingness to learn the information, because they launch the search themselves. Thus, it is important to let the target customers get business’s information when target customers use the term that is related to the business. The main objective of this module is to help the business to increase. 治政 the acquisition of search engine. Through correct 大acquisition, the business can 立 then build up the customer’s awareness of the brand and the basic knowledge ‧ 國. 學. for making the degree of customers feel prepared for encounters with the. ‧. organization (customer role readiness) higher.. sit. y. Nat. 4. Recommender System Module: The main objective of this module is to help. io. er. the business to maximize the possibility of consumers to find the right maven (expert in the domain) to curate the content to further develop the customer’s. al. n. iv n C role readiness. Through the of the customer’s role readiness, h edevelopment ngchi U business can have an arousal on CEB. 5. CEB Collection Module: CEB collection module is to intend to collect the CEB to provide not only the feedback of the present state, but also the feedback to the three new media module so that the three new media module can give an adapted service to the specific businesses. 6. Brand Partnership Module: Brand partnership module’s objective is to help the business to communicate with the multi-stakeholder. In this module, we use the most common way in business, email communication channel, to help the business manage the brand partnership with their cooperators. 34.

(35) 3.3 System Scenario CTaipei website (http://www.taipeing.net/) was established in 2012, it is the website for accelerated Taipei become a creative city. Businesses can create an account on CTaipei, put store information, and be positioning by customers with imagery. But in addition to receiving the customers’ position, businesses are also very concerned about how to engagement customer with new media. Based on such requirements, we use CTaipei website as an engagement site, iEngagement. In the previous situation, R8 bookstore has difficulty to design the keywords for their campaign pages. R8 need to manage their brand on management page of iPeen. 治政 website, and manage Facebook fan page on Facebook 大fan page management page. 立 However, they can’t identify their potential cooperation stakeholders; they have no clue ‧ 國. 學. to find out their mavens and introvert leaders. And these existing customer engagement. ‧. management approaches are complex and decentralized, so iEngagement system. sit. y. Nat. provide R8 with a several services as following:. io. al. er. 1. Optimization of keywords and links of joint campaign pages: R8 and their. n. partners, Elite bookstore, can submit a description of a joint campaign pages to SEO service:. Ch. engchi. i Un. v. We want to hold an exhibition of book selling. In NCCU, and for young people. It will be held in an open space, and the atmosphere will be unfettered. We expect young readers participated in this activity will have the relaxation time. And SEO service will provide R8 with some keywords – Wenshan, young, artist, freedom – for them for improving their joint campaign pages’ content. Besides, SEO service will also suggest R8 with some partners’ website links, such as NCCU. We hope that the use of these two features allow the page rank upgrade. 35.

(36) 2. Influence maven attitude and behavior: R8 can provide us their marketing strategy with some option on iEngagement, and the system will give the suggestion of potential stakeholders – Elite and Kingstone bookstore – to business. After ensuring business and their stakeholders, iEngagement will collect the articles (related to R8 and Elite) on internet and find out business’s maven on recommender. System will process articles to some summarization and influence mavens attitude through these summarization. We expect customer engagement will increase through mavens’ sharing. 3. Optimizing introvert leaders eWOM likelihood: R8 can provide us with the. 治政 link of their Facebook fan page, and system will 大analysis their fans’ behavior 立 data and find out the introvert leaders – user A, B, C – on their fan page. System ‧ 國. 學. will give A, B, and C the articles about R8 and the joint marketing is mentioned. sit. y. Nat. eWOM likelihood and engage fans.. ‧. about. Using some articles that introvert leaders are interested in to optimaize. io. al. er. Each of the service module can share their database with other to regulate their module if need. And R8 can manage their new media marketing and customer. n. iv n C engagement on iEngagement single In R8 case, business has a single site for h esite.n g chi U. managing eWOM on new media, matching cooperation stakeholders, operating joint marketing, and enhancing customer engagement. This would then be our iEngagement project.. 36.

(37) CHAPTER 4 METHODOLOGY. In this chapter, we will describe the conceptual model, system architecture and each module in details. In Chapter 4.1, there are two main concepts, business cooperation part and B2C2C eWOM part. In Chapter 4.2 there are two main modules, stakeholder choosing module and B2C2C eWOM improving module.. 4.1 Conceptual Framework The section aims to provide the conceptual framework underlying our mechanism. There. 治政 are four important concepts in our conceptual framework, 大brand community, collaboration 立 stakeholder, introvert leader engagement, fan engagement (see Figure 4.1). ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v. 37.

(38) Brand Community Content Source Entertainment Inner event Extern event Information Information of brand, product. Incentive. (1) Introvert Leader. Imagery, Video. (1). Collaboration stakeholder Content Source. 立. Entertainment. 政治Like 大 (2) Response. ‧ 國. (2). Share. Extern event. Like. (3) (3). 學. Inner event. Fan Engagement. Engagement. Information of brand,. n. al. sit. Business Cooperation. er. io. Imagery, Video. y. Nat. Incentive. Share. ‧. Information. Response. iv n CBusiness Cooperation U he Figure 4.1nConceptual g c h i model. B2C2C eWOM B2C2C eWOM. Brand Community Content and Collaboration Stakeholder Content: As mentioned in chapter 1, nowadays, people spend a lot of time on browsing Facebook. However, Facebook will filter out 1500 posts for each person per day through their own mechanism – the posts which are most likely to be the reader’s favorite will be chosen (Facebook, 2014). The results of this mechanism depends on each reader’s reading habits, therefore, we need to find out which kinds of post are reader’s favorite. According to the relevant literature, the posts are divided into three types – entertainment, information and remuneration (Cvijikj et al., 2013). The reason why the scholar divided the 38.

(39) posts in these three types is because entertainment and information were considered to have a stronger effect for online engagement, and remuneration is the third frequently motivation for engagement (Muntinga et al. 2011). Such classification can include most types of posts on Facebook. Brand community content refers to the distribution of the types of posts of the reader’s favorite. Business managers can thus be informed and endeavor to make their B2C2C eWOM posts meet the mechanism of Facebook, appeared on reader’s wall.. Introvert Leader Engagement and Fan Engagement: Besides brand community content of different posts’ types, it is important to have introvert. 政治大 reason of utilizing the introvert leader’s behavior as monitoring measurements is introvert 立. leader engagement with the behavior measures of counts of likes, responses, and shares. The. ‧ 國. 學. leaders have a greater influence than others on social media for fans. Fans following an introvert leader who has many interactions with fans but did not expose on Facebook are for. ‧. them having the same interests or personality traits (Zubair Shafiq et al, 2012).. Nat. sit. y. A complete introvert leader will be followed by many fans, and its fans will have some. n. al. er. io. explicit interaction on its post, such as like, share or response (Zubair Shafiq et al, 2021). An. i Un. v. IR shares the post from fan page not only can get a lot of interactions, but also can be more. Ch. engchi. easily exposed on Facebook, because the nature of a group of similar people makes it easier for Facebook to determine where will be more appropriate for the post to appear. The innovative part of the conceptual model is that we only use count of like, share and response, and these will make us easier to measure fan’s engagement. These criteria are not only easy to obtain and observe, but also overcome the difficulty to obtain other information under the Facebook privacy policy restrictions. In this framework, businesses only need to be introvert leader’s recognized, and businesses can also optimize C2C eWOM likelihood through the influence of introvert leader indirectly. Accordingly, this research proposes the following hypotheses to investigate: 39.

(40) Hypothesis 1: In a brand community, different types of articles will have different degree of engagement for introvert leaders with fan page. Cvijikj et al. (2013) pointed out that different types of articles will have different degree of engagement for fans. In our conceptual model, we are interested in whether introvert leaders’ engagement degree will also affected by different types of posts. By enhancing engagement of introvert leaders, introvert leaders will have more likelihood to having some behavior on Facebook; it might be like, share or response. If such a hypothesis holds, our first phase of the B2C2C eWOM approach will then be feasible.. 政治大 Hypothesis 2: The articles from cooperation stakeholders with different types of articles will 立. ‧ 國. 學. have different degree of engagement for introvert leaders with fan page. There have no famous research on the impact of different types of articles from. ‧. cooperation stakeholders on introvert leaders. However, it is also as an important factor in our. Nat. sit. y. conceptual model. Chiranjeev (2003) pointed out that co-branding is a good way to a business. n. al. er. io. to reach in or reach out the market. For this reason, we must understand that if there will result. i Un. v. in different level of engagement for introvert leaders with fan page, then we can make a good. Ch. engchi. cooperation between enterprises. Businesses emphasis on potential value and it depends on the engagement of introvert leaders. When the more engaging of introvert leaders, the more likelihood B2C2C will be successful.. Hypothesis 3: The different kinds of behavior from introvert leaders will drive different degree of engagement for fans on the fan page. We consider that fans may have some different degree of engagement when introvert leaders do not show the same behavior. Roderick et al. (2011) pointed out that there are several degree of engagement, emotional, cognitive and behavioral. If fans respond to introvert leaders’ 40.

(41) behavior, it means fans are in relatively high engagement. And because of observable behavior, business can determine the success of eWOM marketing. On Facebook, there are several behavior types users may make: like, share, response on the post or message with friends in private message box. When introvert leader interacts with the post on fan page, Facebook mechanism might think this post will also attract introvert leader’s friends and push this post to their wall. Through this process, introvert leader will make some influence on their friends, but if introvert leader forwards the article with their own view that may attract their friends to read the article more carefully. This concept is different from the current practice: attracting users’ “like” on the article. We think we should spread the. 政治大. article to everyone who may be interested, and it is the goal of maximizing eWOM.. 立. 4.2 System Architecture. ‧ 國. 學. The system architecture (Figure 4.2) is designed to be able to realize the conceptual model.. ‧. And this section will describe our system architecture and connect it to the conceptual. Nat. sit. n. al. er. io. explain in detail.. y. framework. We will separate system architecture in to a number of modules one by one to. i Un. v. Description from the macroscopic of view, our system is applied to two frameworks,. Ch. engchi. Django for the backend and Drupel for the frontend. We use python, good performance for developing web crawling, to implement Django framework, and embed our service into CTaipei, a Drupel based engagement website. Our system architecture consists of two databases, two modules and two outputs. Those are fan page database, engagement site database, introvert leader measuring module, posting measuring module, introvert leader suggestion result and post suggestion result. Through Facebook crawler, we collect data into the fan page database. Introvert leader measuring module will find out potential introvert leader through its mechanism and the data from fan page. In addition to post measuring module, we also will measure each post’s eWOM 41.

(42) performance as a reference for fan page managers. After the above module operation, we will assess who are the introvert leaders and which post obtains the higher eWOM performance.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v. 42.

(43) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. Figure 4.2 System Architecture. 43.

(44) 4.2.1 Post Measuring Module In post measuring module, we are going to identify good post on the fan page. There will be three article of the most counts of likes, shares and response. These articles may be the reference to fan page manager in the future when they are lack of inspiration. With the collected data from fan page, the module will evaluate a post with its eWOM performance. We can evaluate the performance with the line chart; it is a visual chart for business to compare eWOM performance with previous time slot, higher or lower (Equation 1). In Formula 1, P is eWOM performance measuring result, and eWOM is the performance of. 政治大. WOM. t implies a time period. It may be a week, a month or a time period of a campaign,. 立. depends on business users’ needs.. ‧ 國. 學. P = 𝑒𝑊𝑂𝑀𝑡 − 𝑒𝑊𝑂𝑀(𝑡−1) Equation 1. ‧. The meaning of like and share may imply that fans agree with the post. Hoyer et al. (2010). Nat. sit. y. pointed out that customers interact with business is one of their way to reflect their engagement.. n. al. er. io. And Katrien et al. (2013) also pointed out that when customers have the consistent goal with. i Un. v. business; it will be easier for business to operate value co-creation with customers. In addition,. Ch. engchi. the interaction in a post is one of criterion for Facebook to measure a post that is worthy to spread or not (Facebook, 2010). And the next step, we are going to find out introvert leader in the fan page.. 4.2.2 Introvert Leader Measuring Module In introvert leader measuring module, the module is to find out introvert leaders and measure their performance through their behavior on fan page. Briefly, the module will collect interaction behavior data on each article and it will find out introvert leader. 44.

(45) Specifically, every fan will leave some interaction behavior data on each article, such as like, share or response. We will use some formula to find out introvert leaders who’s sharing or responses may attract a lot of feedback from other fans; this part will be explained in the following sections. And we will use the information in Facebook insight to measure introvert leaders’ performance. We can compare introvert leader’s eWOM performance with its previous record stored in Facebook insight (https://www.facebook.com/[fan page name]/insights/). Facebook insight provide fan page manager each article’s sum of like, share, response and reach. For instance, the “reach” means how many people saw the article on fan page. This information can help decide whether or not to continue to regard them as the introvert leaders.. 政治大 crawling, fan page database and fan page measuring module. We first collect interaction 立. Introvert leader measuring module involves three components: Facebook fan page. ‧ 國. 學. behavior data from the fan page database, then the next stage for further analysis of the process. About introvert leader, Zubair Shafiq et al. (2012) had a complete study. On social media,. ‧. there are two kinds of opinion leader: extrovert leader, introvert leader. Extrovert leaders have. Nat. sit. y. more friends than another but introvert leaders have more interaction with their friends and in. n. al. er. io. the more rapid transfer of information. Referencing to the literature, we’ll use some formula to. i Un. v. extract introvert leaders and it’s interactive with their followers from the fan page.. Ch. engchi. Step1. Calculating Interaction Behavior The literature (Zubair Shafiq et al. 2012) pointed out that each user has two weight terms ρi and γi, and it also provided a formula to solve for ρi and γi for each user υi. The symbol t is time period, and (t-1) is previous time period. The symbol ρ implies whether a user is easy to be impacted by external factors; and the symbol γ implies the relationship between activities levels in consecutive time periods. The symbol m is message matrix, in the following formula (Equation 2 (Zubair Shafiq et al. (2012)) represents the overall interaction behaviors from which the introvert leaders can be identified. That is, this formula can be used to calculate the 45.

(46) value of ρ and γ, and these two variables are two dimensions for business to filter out the people who are not introvert leaders. In this formula, m and t are known. So when value of e close to zero, we can calculate out the value of ρ and γ, and start the next step. 𝑁. 𝑁. 𝑁. ∑ 𝑚𝑖,𝑖′ (𝑡) = 𝜌𝑖 ∑ 𝑚𝑖 ′ ,𝑖 (𝑡 − 1) + 𝛾𝑖 ∑ 𝑚𝑖,𝑖 ′ (𝑡 − 1) + 𝑒𝑖 𝑖 ′ =1. 𝑖 ′ =1. 𝑖 ′ =1. ∀t ∈ N, 1 ≤ t ≤ 𝑡𝑚𝑎𝑥. Equation 2 Step2. Finding Introvert Leaders We use two dimensions ρ and γ to analyze fan page users data. The method we chose is. 政治大. k-mean (I. S. Dhillon, et al, 2004), and we can use R program language to make the distribution. 立. chart. Referencing to literature (Zubair Shafiq et al. 2012), our system is able to find out who. ‧ 國. 學. are the introvert leaders in the fan page. And they will be an important role to operate the B2C2C eWOM.. ‧. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v. Figure 4.3 Actor’s role distributed by ρ and γ (Zubair Shafiq et al. 2012) This show the result of users filtered by step1, and most of introvert leaders are distributed in the lower left corner, with low ρ and negative γ. ρ means network coefficient (i.e., how much degree of that an actor is impacted by their neighbor); and γ means ego coefficient (i.e., how much degree of the probability that an actor may be impacted by their previous behavior) 46.