EVALUATING SUCCESS FACTORS FOR A VIABLE
RABO STARTERSCOMMUNITY:
THE INFLUENCE OF TRUST, ONLINE FEATURES AND COMMUNITY
MANAGEMENT ON THE VIABILITY OF THE RABO
STARTERSCOMMUNITY
Master thesis, MSc BA, specialization Business Development, University of Groningen, Faculty of Economics and Business
February 2011 YVONNE M. DIJKSTERHUIS Student number: 1507117 Heymanslaan 13, 9714 GE Groningen, The Netherlands y.m.dijksterhuis@student.rug.nl University supervisor: Dr. J.F.J. Vos Dr. C. Reezigt Organization:
EVALUATING SUCCESS FACTORS FOR A VIABLE
RABO STARTERSCOMMUNITY:
THE INFLUENCE OF TRUST, ONLINE FEATURES AND COMMUNITY
MANAGEMENT ON THE VIABILITY OF THE RABO
STARTERSCOMMUNITY
Acknowledgements:
I would like to thank my university supervisor J.F.J. Vos for her time and effort. Words of thanks are owed to the colleagues of Rabobank SMG for making this internship a wonderful and educational experience. Especially, the colleagues at the department of Bedrijvenadvies. Finally, special thanks to my friends and family, who supported me throughout this thesis by offering a listening ear when I needed one and for reviewing my work. You all have been a great support.
ABSTRACT
This research focuses on the success factors of the online Rabo starterscommunity and to which extent these have a positive influence in developing a viable community. Rabobank Stad en Midden Groningen (SMG) considers starting entrepreneurs to be an important target group. With that in mind they initiated the Rabo Starterscommunity, an online platform for starters to share experience and knowledge on the tribes and tribulations of setting up a new company. However, the community never became successful, despite the fact that many ideas were developed by the project team.
By conducting a survey and interviews, the success factors were evaluated from the point of view of the members and the community managers. The interviews also highlighted that the collaboration between Rabobank SMG and Rabobank Nederland (RN) should and could have been better. This was confirmed by the observations that were made during the internship.
Furthermore, findings of this research reveal that success factors regarding trust and online features do not correlate with perceived usefulness, through which viability is measured in this case. Moreover, interviews and observation outcomes indicate that Rabobank SMG is highly dependent on RN and with a lack of cooperation between them, the effectiveness of community management falls behind.
Despite these results, the interview and observation outcomes did expose that the success factor effectiveness of community management is a precondition in order to become viable. A dedicated leader is considered to be the driving force behind a successful Rabo Starterscommunity. An enthusiastic community manager is able to motivate and stimulate the project team to meet the other success factors, i.e. information quality, system quality and the level of offline interaction. Accessibility of personal information seems to be a bad indicator for success, since companies actually preferred to profile their business online. Additionally, from observation it became clear that the moment the reigns were slackened, caused by a switch of community managers at Rabobank SMG, the community felt down as well. This finding strengthens the statement that a successful virtual community starts with a dedicated leader.
Further research should be conducted to explore if trust and online features are truly poor factors to indicate success of the Rabo Starterscommunity. In case of further research limitations of this study should be addressed.
TABLE OF CONTENT
1. INTRODUCTION ... 7
2. THEORETICAL FRAMEWORK: SUCCESS OF VIRTUAL COMMUNITIES .... 9
2.1 Virtual Communities ... 9
2.2 From Viability to Perceived Usefulness as Success... 10
2.3 Success Factors of Virtual Communities ... 11
2.4 Trust ... 12
2.4.1 Accessibility of Personal Information.... 14
2.4.2 Level of Offline Interaction... 14
2.5 Online Features ... 15
2.5.1 Information Quality... 16
2.5.2 System Quality... 16
2.6. Effectiveness of Community Management ... 16
2.7. Conceptual Model ... 17
3. RESEARCH METHODS ... 19
3.1 Research Context: Rabo Starterscommunity... 19
3.2 Procedure, Participants and Data Gathering ... 20
3.3 Survey, Interviews and Observation ... 21
3.4 Operationalization of the Variables... 22
3.5 Quality Criteria for Research ... 24
4. RESULTS... 26
4.1 Descriptive Statistics ... 26
4.2 Correlation Analyses ... 27
4.3 Interview Outcomes ... 28
4.4 Information Obtained from Observation... 31
4.5 Overview of the Relevant Findings... 32
5. CONCLUSION AND DISCUSSION... 34
5.2 Implications for Practice ... 38
5.3 Implications for Science... 38
5.4 Limitations and Suggestions for Further Research ... 40
REFERENCE LIST ... 41
APPENDIX A ‘Rabo Starterscommunity’ ... 47
APPENDIX B ‘ List of Items by Construct’ ... 48
APPENDIX C ‘Interview Guide’... 50
“Computer-mediated communication is not just a tool; It is at once technology, medium, and engine of social relations. It not only structures social relations, it is the space within which the relations occur and the tool that individuals use to enter that space”
Steven G. Jones CyberSociety 2.0, 1998: p. 11-12
1. INTRODUCTION
Nowadays, the most popular activity on the Internet is social interaction (Iriberri and Leroy, 2009). Virtual communities such as Twitter, LinkedIn, and Facebook are common property. The emergence of online communities has altered the way people interact with each other (Wu, Chen and Chung, 2009). Consumers are gradually turning to computer-mediated communication which enable them to give or acquire information and knowledge by simply interacting with others across time and space (Igbaria, Shayo and Olfman, 1998). Moreover, virtual communities have proven to have been helpful on an emotional level, for example in sharing pain from illness or loss (Preece, 1998).
Much like the dotcom rush, firms have eagerly joined the virtual communities hype to promote their brands and products without knowledge about how these communities work and how to make them successful (Saulding, 2009). Nevertheless, simply providing a virtual platform does not secure a viable online community (Preece, 2001). Viable online communities rely on people using their sites continually, encouraging active participation among members and, most importantly, amplifying the loyalty of community members (Kim, Lee and Hiemstra, 2004).
Preece (2001) argues that many initially active online communities have neglected to hold their members, because they hoped that customers would be tempted to visit the site through social interaction. This naïve view regarding ‘build-it-and-they-will-come concept’ have often produced “cyber ghost towns”(Preece, 2001).
A company who can agree first hand with these statements is Rabobank Stad en Midden Groningen (SMG), on which this research is based. Their virtual community, the Rabo Starterscommunity is launched in November 2009. The purpose of this website is to encourage fellow starting entrepreneurs (also referred to as starters) to network with each other. By doing so, information regarding the tribes and tribulations of setting up a new company can be obtained easily and without further obligations. However, the community did not have the effects as it was hoped for. After one year of being in the air, the Rabo Starterscommunity attracted few people, who furthermore neglected to start interacting with one another. It is now up to Rabobank SMG to spur conversation, interaction, and reaction between the community members to make the community a success.
community. In the literature a number of factors are mentioned. However, it is impossible to apply them all on the Rabo Starterscommunity. The choice for success factors depends on the kind of community and its characteristics.
My aim in this paper is to investigate the success factors of virtual communities in order to give recommendations for the –up until this point- disappointing virtual community of Rabobank, the ‘Rabo Starterscommunity’, and secure its sustainability.
In order to do so, the following research question will be answered in this study:
What is the influence of the relevant success factors for the viability of the Starterscommunity of Rabobank SMG and to what extent are those factors met?
The sub questions that follow from the research question are:
1. Which success factors are considered to be relevant for the Rabo Starterscommunity in order to gain perceived usefulness?
2. To what extent are the relevant success factors met by the Rabo Starterscommunity? 3. What conclusions can be made concerning the relevant success factor in order to gain
perceived usefulness?
This thesis is organized as follows. Success and success factors of the Rabo Starterscommunity are introduced and explained in the theoretical framework. The relationship between these concepts are translated into a conceptual model. A survey was conducted to test the relationships. Accordingly, interviews were held and observations were done to complete the data gathering. These findings are explicated in the results section. A discussion and conclusion on the research question is discussed in the concluding section.
2. THEORETICAL FRAMEWORK: SUCCESS OF VIRTUAL COMMUNITIES In order to research the success of the Rabo Starterscommunity, I will first define a virtual community and describe when an online community is considered viable. Accordingly, a literature overview of success factors is given. This is followed by the success factors that will be applied on the Rabo Starterscommunity. This section will end by displaying the conceptual model.
2.1 Virtual Communities
The variety of definitions of the term virtual community seems never-ending. The current literature does not provide an unambiguous definition of the concept of virtual communities. This is shown by the fact that online communities have been studied by several disciplines in which each of them brings in its own definition. However, this phenomenon is traditionally defined from a sociological perspective. In line with this perspective, the most quoted author in the online community literature is Howard Rheingold (Iriberri et al., 2009). He defines virtual communities as “social aggregations that emerge from the Net when enough people carry on public discussions long enough, with sufficient human feeling, to form webs of personal relationships in cyberspace” (Rheingold 1993, p. 5). Many scholars have adopted this definition, which became the core of much successive research, and took bits and pieces of it to create a connotation of the virtual community from their own perspective.
Ridings, Gefen and Arinze (2002) for example, describe a virtual community as a group of people who interact regularly on a common interest, problem or task in an organized manner over the Internet. In this way, communities can overcome geographical and time barriers that exist in traditional offline communities (Andersen, 2005). Similarly, Blanchard and Markus (2004) argue that day to day interaction takes place primarily through computer-mediated communication.
The Rabo Starterscommunity can be defined as an online platform that provides facilities for interaction which enables fellow starting entrepreneurs, experts and Rabobank personnel to share experience and knowledge and start discussions on the tribes and tribulations of setting up a new company. Moreover, it provides a network function for starters, a place where they can profile their business.
There are various types of online communities. Hagel and Armstrong III (1996) distinguish four types of communities: communities of transaction, interest, fantasy and
relationship. In each of these communities people interact on a regular basis, however the difference lies in the (diversity of) topics that are covered and the awareness of one another’s identity. For example a community of transaction is purely focused on buying and selling whilst in a community of relationship interaction takes place on a much more personal level. Based on the definition mentioned earlier, the users of the Rabo Starterscommunity are forming a community of interest. In such a community participants are brought together to share information, their expertise, and passion on specific topics. They do not discuss anything outside the topic, nor do they interact about personal affairs (Hagel and Armstrong III, 1996).
2.2 From Viability to Perceived Usefulness as Success
A successful virtual community relies on people who are attracted to visit the site and subscribe themselves as a member. In order to become a viable community the members should be encouraged to take part, and stay involved in discussions and knowledge sharing.
According to Teo, Chan, Wei and Zhang (2003) a virtual community is viable when (new) members adopt and use the information system unquestionably. Members’ adoption is measured by the continual increase of membership, the unquestionable use of the community is determined by the growth of user-generated content. In other words a virtual community is viable when a substantial number of members adopt and make use of the site to gather and share information. However, it is impracticable to use this measurement for this research, since this study aims at evaluating the success factors for a viable Rabo Starterscommunity and analyzes how they are managed. It will not discuss the results before proper executing of the recommendations of this study and the results after implementation. Accordingly, this research does not explicate if such an implementation might result in an increase in members and in a growth of user-generated content.
To solve this issue, a measurable variable should be used in place of viability. The measurable variable must be closely related to viability. This research will measure perceived
usefulness in order to determine the viability of the Rabo Starterscommunity. Many authors have discussed the importance of perceived usefulness for a virtual community and their correlation with viability. For example, Teo et al. (2003) consider that in order to ensure viability, it should be explored whether the community is useful to its users. In addition Wegner, Mcdermott and Snyder (2002) state that members who do not longer perceive usefulness start leaving the virtual community.
Accordingly, Butler (2001) notes that when a virtual community is capable of creating perceived usefulness it increases the ability to establish a viable community. Online communities that are able to provide valuable benefits are better capable of attracting and keeping members, which leads to a more viable community. Many benefits of virtual communities have been identified, including social support, access to information, encouraging discussion and knowledge sharing (Butler, 2001).
Perceived usefulness is generally defined as “the extent to which a person believes that using the system will enhance his or her job performance” (Venkatesh and Davis, 2000, p. 187). In the context of an online community, perceived usefulness means the belief a user has that participating in virtual communities enables him or her to get information from and share knowledge with community members and as a corollary enhance the performance in information exchange (Lin, 2007).
2.3 Success Factors of Virtual Communities
In general, research on success factors aims to offer guidance on how to provide and use resources in an ideal way. Very often these recommendations seem to be insufficient, as the amount of influencing constructs are high (Leimeister, Sidiras and Krcmar, 2006). According to Tromsdorf (1990) one should formulate guidelines which can be controlled by the community manager (also referred to as community leader or moderator) in order to create a strategy that is predicted to be successful.
Many authors have dedicated themselves to determine the success factors of virtual communities. Various authors have mentioned the importance of trust. Ridings et al.(2002) mention that trust influences members’ intentions to both provide information and receive information through the online community while Wu et al. (2009) stress that trust leads to relationship commitment and member stickiness. Casaló, Flavián and Guinalíu (2008) have emphasized the role of consumer commitment and the fundaments of trust that are necessary in order to reach this. They suggest that trust can be defined by three types of beliefs: competence, honesty and benevolence while Ridings et al.(2002) and Wu et al. (2009) distinguish ability, benevolence, integrity, and predictability. Integrity and predictability refer to the righteous behaviour as well as interaction policies and guidelines.
Many authors have dealt with the use of a policy or “netiquette”. Irriberri et al. (2009) argue that moderators need to set a clear standard to be able to nip possible conflicts in the bud. Preece (2001) has concluded that policies guide members’ interaction and contribute to
accepted social norms. In alignment with this issue, several scholars (Wu et al., 2009; Kim, 2000; Eastlick, Lotz and Warrington, 2006) have put emphasis on privacy protection.
Koh, Kim, Butler and Bock (2007) have argued that leader involvement and offline communication influence the activity on the community. In addition, Rothaermel and Sugiyama (2000) stress that off site communication reinforces the relationship-building process of online communities. Moreover they note that an active site organizer positively influences the perceived value a community member has about the website. Ho and Huang (2009) conclude that both leader involvement and off site interaction leads to satisfaction. Lin (2008) has identified the online features information and system quality as the key issues for member satisfaction. Furthermore, Lin suggests that virtual communities meet their needs when members believe they are trustworthy, accessible, and user-friendly. Moreover it should deliver accurate, thorough, constantly updated and personalized information. Therefore, community managers should put time and hard work into securing that their members are satisfied with information and system quality.
As can be revealed from the literature overview, success of a virtual community is influenced by a great array of factors. However, when one pays close attention it can be concluded that most of the success factors of a virtual community boil down to three factors: Trust, Online features & Community management. Moreover, these aspects are more or less intertwined. In addition, trust influences members’ intention to provide and obtain information (Ridings et al., 2002). However, the community should provide a platform that suffices in order to enable members to execute the information exchange.
2.4 Trust
Many authors state that trust is important for enabling members to be active (Casaló et al., 2008). Therefore, it is fundamental to realize that a virtual community is more than just computer-mediated communication. It requires more than providing a website, it requests members who find the community trustworthy (Lin, 2008). Furthermore, a community needs reinforcing measures to link members and spur its interactivity. In addition, Casaló et
al.(2008) state that a person needs to have faith in an online community and in its members before they actually subscribe themselves for it.
Ridings et al. (2002) note that some elements of the virtual community complicate the evolvement of trust, such as the absence of a privacy policy and the lack of
social presence.
Trust between community and members can be build by providing members a sense of control over the disclosure of their own personal information, because people do not easily give away their personal information. The community needs to gain trust by handling member data sensitively. A valuable step to back this up is offering a privacy policy that is lived up to by the community (Kim, 2000). Wu et al. (2009) mention the protection of privacy as a crucial term for online interactivity. This is especially of interest when members like to stay anonymous (Hummel and Lechner, 2002).
In the case of Rabobank SMG, everybody with access to the Internet can see who have joined the community as a member and which actions they have taken so far. As mentioned above, members seem to cherish their privacy. By researching the accessibility of personal information it should show if the members of the Rabo Starterscommunity feel the same about their privacy and if this influence their activity on the community.
Furthermore, a person needs to have faith in the members of the community (Casaló et
al.,2008). Inherent to an online platform is the lack of social presence. In order to gain trust in this area, the organization of offline meetings can be an addition on the virtual community. Rothaermel et al. (2001) argue that off site communication reinforces the relationship-building process of online communities.
Especially in the case of the Rabo Starterscommunity, offline interaction might complement the online activities. A part of doing business, is networking in order to generate a clientele and potential business partners. Particularly in the stage where entrepreneurs just launched their own company and need to profile their business. The organization of an offline gathering could be considered as an extra opportunity to network and to deepen the relationship with the members of the community. For that reason, the organization of offline events might make the community of the Rabobank SMG more attractive to join. Furthermore, meeting each other face-to-face makes it easier to understand, trust and identify other members (Koh et al., 2004).
The above motivates why the success factors might be important for the Rabo Starterscommunity to address. The next section continues by elaborating in more detail on these two factors of building trust.
2.4.1 Accessibility of Personal Information.
In order to gain trust, one of the issues a virtual community should address is the establishment of a policy that contains clear standards that vouch for sensitively handling of member data. However, Kim (2000) argues that, as is applicable in this case, if a community is closely related to “real life”, full profiles help to gain trust between members since a context is provided. In this way, members can form a picture of the person they are dealing with and they might going to meet in person.
Wu et al.(2009) state that the protection of privacy is a crucial term for online interaction. Research on e-commerce has shown that when no privacy and security of personal information can be guaranteed it will prevent online transactions from growing (Eastlick et al., 2006). “Privacy is the right to be left alone, and also refers to the ability of an individual to control, manage, and selectively reveal personal information.” (Wu et al.2009, p.3). In order to gain trust of the (potential) members of the community a privacy policy should be well formulated in which it is explained how the community acquires, gathers, and uses participants’ information (Wu et al., 2009). Moreover, Kim (2000) notes that with a view to building trust with members of the virtual community, promises that are included in the privacy policy should be kept.
2.4.2 Level of Offline Interaction
Virtual communities are characterized by a lack of social presence, which inherently creates communication weakness (Koh et al., 2007). Accordingly, this makes it harder for virtual community members to understand, trust, and identify other members. By organizing offline events, online ties can be strengthened and broadened through face-to-face meetings (Koh et
al., 2004). This is supported by the social presence theory, that explicates that the presence of members, which can be reinforced through offline communication, may influence the ties of community members to their online communities (Short, Williams and Christie, 1976). Rothaermel et al. (2001) state that offline interaction might be a crucial factor in explaining community members’ embeddedness, thus strengthening the relationship-building process of an online community. In addition, Koh and Kim (2003) found that offline communication enlarges the solidarity and cohesiveness of an online community, and reinforces the links between members.
Moreover, Walther (1994) argues that when members do not expect to meet members in person, there is more enticement for participants to benefit (asking for information and
sharing knowledge) from the group and not give back to the group. This implies that some people lurk and do not offer their help, information and support. Conversely, when online interaction takes place on a local level, as is the case at the Rabo Starterscommunity, the ‘free-rider’ problem is less applicable since the chance of meeting one of the participants in person is extremely high (Kavanaugh, Carroll, Rosson, Zin and Reese, 2005). This suggests that when the Rabo Starterscommunity would start organizing offline meetings, it not only strengthens the online ties but also put up a barricade to lurk.
2.5 Online Features
A variety of studies have discussed online features, however the dimensions suggested by the DeLone & McLean Success Model received the most attention. Their model is a framework to measure the performance of an information system. They found that the online features, information and system quality, are an important construct for a successful information system (DeLone & McLean, 2004).
Online features refer to the quality measure of web-based information systems (Lin, 2007). DeLone and McLean (2004) explored that information quality and system quality are essential online features for making an information system successful. Lin (2007) argues that information and system quality are factors that influence the sustainability of an online community. Lin (2008) further suggests that virtual communities meet the needs of the members when they perceive the virtual community as accessible and user-friendly. Also the deliverance of accurate, thorough, constantly updated and personalized information is highly appreciated.
In addition, Koh et al. (2007) found a direct relationship between usefulness and viewing activity. They suggest that members incline to gather and share information more often when they perceive that a community and its content are of use to them. Equally, Lin and Lu (2000) note that information quality is positive related to perceived usefulness. This is in accordance with Lin’s (2007) finding that a higher information quality enhances the usefulness of the online community.
In the same study Lin (2007) found a strong relationship between system quality and perceived usefulness. She explains that this is not surprisingly since the goal of an online community is to facilitate a platform where members can discuss and start topics in such a way that they experience an enhancement in their performance.
The above confirms to research the information and system quality of the Rabo Starterscommunity since perceived usefulness is defined as success in this research. Both online features are detailed below.
2.5.1 Information Quality
Information quality can be defined as the quality of the information contributed by providers of virtual services. According to Hsu and Lu (2004) is the continuous provision of content that visitors consider as useful an important element for the viability of the community. Such content contributes to maintain existing members and to draw new ones to the community (Rothaermel et al., 2001). Wu et al. (2009) stress that a steady traffic flow to the website is vital for an online community to be commercially effective.
Information offered on the community should be customized, complete, of interest, and easy to understand (DeLone et al., 2004). Lin (2007) notes that information quality is inherent for the success of an online community. Perkowitz and Etzioni (1999) associate information quality with usefulness. They argued that the information is only useful if the user considers the information on the website to be accurate, informative and up-to-date.
2.5.2 System Quality
System quality in an online information system determines the possibilities of a website. For example, the ease of navigating the website (usability) and convenience of access (availability) are highly held values (DeLone et al., 2003). In addition, Yoo, Suh and Lee (2002) argue that system quality is particularly fundamental in online communities. Many community members are discouraged to make use of the website when they experience difficulties in accessing and navigating the website. Barriers of entry may have limiting effects on membership, leading to less interactions, less interesting elements, and may over time cause the end of the online community (Teo et al., 2001). Navigation refers to “the ease of use with which the user can move around and find what they want in the community” (Preece 2001, p. 350).
2.6. Effectiveness of Community Management
The success factors that were mentioned in the previous sections seem to be essential for the Rabo Starterscommunity in order to gain perceived usefulness. However, it requires more than just implementing those success factors. It requires a community manager who is
committed to make a success out of the community and who is involved by effectively implementing these factors of success.
For the sustainability of any online community, Preece (2002) points at an effective community leader as one of the critical success factors. Regardless of whether the leader is officially selected and named by the moderator or is a self-proclaimed proprietor of the virtual community, leader involvement is crucial for building relationships and stimulating participation (Figallo, 1998). Leaders are responsible for setting up the standards, organizing events on a regular basis and offer up-to-date content. Moreover, they should create trust within the community in order to stimulate participation among members (Koh et al., 2007). Concerning trust in this research, the community manager of the Rabobank should offer a privacy policy and organize offline events.
Leader involvement is also necessary for encouraging members’ active participation in posting and viewing community content (Koh et al., 2007). Furthermore, the community manager should aim at quick responsiveness and offering a platform that is easy to access and navigate (Rosenkranz and Feddersen 2010).
2.7. Conceptual Model
This study researches the influence of the relevant success factors on gaining perceived usefulness towards the Rabo Starterscommunity. As is discussed before, perceived usefulness will be used to determine the viability. This explains the dotted lines in the conceptual model, displayed below.
Trust and online features are considered to be the most important factors in order to gain perceived usefulness. However, no success can be guaranteed if the Rabobank lacks a community manager who is fully devoted to the Rabo Starterscommunity. Hence, arrows are placed from effectiveness of community management to both of the factors.
Y.M. Dijksterhuis University of Groningen | | | | | | | | --- | --- > | | | | | | | |
Model 1: Conceptual Model |
| 17 Trust Level of Offline Interaction Accessibility of Personal Information Effectiveness of Community Management Online Features System Quality Information Quality Perceived Usefulness
Viability of the Rabo Starterscommunity
3. RESEARCH METHODS
This section presents the research methods that were used for this paper. Moreover, it discusses the operationalization of the variables. In order to value the quality of this research, reliability and validity were evaluated. However, this section starts with providing an insight into the Rabobank and their virtual community.
3.1 Research Context: Rabo Starterscommunity
Rabobank Stad en Midden Groningen (SMG) is a financial service provider and is characterized by its cooperative identity. This entails that Rabobank exists out of 1421 independent local Dutch Rabobanks, that altogether form a cooperation. These banks all provide their own management and their own board of commissioners. Rabobank Nederland (RN), the central cooperation, attends to the interests of the local banks.
The department that serves the middle and small-scale industry (In Dutch: Bedrijvenadvies) considers starters as an important target group. They have made it their aim this year to reach a market penetration among starters of 21,5 %.2
In the past, Rabobank has always been a popular bank for starters. However, recent studies of the Rabobank SMG have shown that starters are not that eager to go to the Rabobank for their financial business as before. Due to this outcome, the bank has decided to look for new means and facilities to attract starting entrepreneurs. One of these facilities is the Rabo Starterscommunity with which the Rabobank hopes to generate customer loyalty towards the bank.
In November 2009, Rabobank launched their new strategy with the Rabo Starterscommunity, a community for business starters, in order to position themselves more strongly in the market of starters. Rabobank aims to contact prospective entrepreneurs in an early stage of their plans, in order to supply full guidance. By providing an online community where they can meet fellow starters, experts, and Rabobank personnel, the bank would supply guidance to starters concerning orientation and entrepreneurship. Moreover, they could meet potential customers and profile their company on the community, which brings a commercial aspect to the community.
The Rabo Starterscommunity shows different layers. First of all, the community consists of a national homepage, where all the members can interact with each other on a
1
In the year of 2010.
2
national level. Furthermore, the Rabo Starterscommunity offers a platform for members to interact on a local level. Accordingly, next to the national part, the community provides multiple local communities which represent all the local banks that subscribed for the community (See appendix A). This is in accordance with their cooperative identity as previously explained. Each local Rabobank can accept or decline taking part in the Starterscommunity. The website is offered by Rabobank Nederland (RN) and can be used at a fee by local banks.
The responsibility of the development of the website of the Rabo Starterscommunity lies entirely with RN. The local bank has virtually no influence on the layout and can only decide how they want to give meaning to their website. This means that in this case, Rabobank SMG is responsible for providing relevant information, promoting the community and spur interaction. However, being a local bank, Rabobank SMG has no influence on certain aspects. Issues like ease of use, the protection of personal data and optimizing the website are not within the power of Rabobank SMG. RN is accountable for that part of the community. This means that two community managers have a responsibility towards the Starterscommunity e.g. the local manager of Rabobank SMG concerning filling up the website and the manager of RN regarding the lay-out and system of the website. Consequently, there is a dependency of Rabobank SMG in relation to RN. That makes it in addition to the commercial aspect, an extraordinary community.
3.2 Procedure, Participants and Data Gathering
This research tries to explore the influence of trust, online features and community management on the viability (through perceived usefulness) of the Rabo Starterscommunity. In order to answer the research question, this study started off with examining the theory concerning success factors. To explore the attitude of the community members towards the factors trust and online features, a survey was conducted. Every member of the Rabo Starterscommunity was approached by email for filling out the questionnaire. Because this research involved a small population, members who subscribed during the period the survey was taken, received a questionnaire as well. Of the 71 people who were approached, only ten returned the survey. This resulted in a response rate of 14,08%.
In addition to the survey, interviews were held with the community manager of RN and the community manager of Rabobank SMG3. This was done to evaluate the effectiveness of community management. Both community managers were asked about their activities concerning the community and their view about a successful community and when they would consider the Rabo Starterscommunity as a success. Additionally, information concerning their involvement was gathered by (participative) observation since I became a member of the project team.
3.3 Survey, Interviews and Observation
In order to ascertain how the members value the success factors of trust and online features, a descriptive survey was conducted. The reasoning behind this kind of survey stems from the fact that this research seeks to learn more about a specific population. In this case the members of the local Rabo Starterscommunity of Rabobank SMG. A descriptive survey aims to determine respondents' views or experiences on a given subject in a fixed structured manner (Gill and Johnson, 2002).
The questionnaire contained closed questions. The respondents were asked to indicate to what extent they agreed upon the statements. Before the survey was sent to the members of the community a pilot survey was done. Five people, who were closely related to the Rabo Starterscommunity, were asked to review the questionnaire on issues such as the ease of filling out the survey, time needed to complete it and its comprehensiveness. After implementing the feedback, the questionnaire was sent to the members.
Every member of the community of Rabobank SMG received the survey per email, in which it was asked how they valued the community. The advantages of using the Internet to conduct this survey, are time savings and cost savings. For example, all the members of the community could be approached, which has a strong benefit over personal interviews in terms of time management. Moreover, email has been found to be a useful mean of conducting this type of research, since the survey population make regular use of the Internet (Cooper and Schindler, 2003).
The survey was distributed during a period of four weeks to all the members of the community, including the members who joined the community for the first time during those four weeks. The completion of the questionnaire took about ten minutes.
3
A switch of community managers took place during this research. The interview and observation outcomes are based on the former community manager of Rabobank SMG.
In addition to the questionnaire, both community managers were interviewed to find more about their involvement and management skills concerning the Rabo Starterscommunuity (See Appendix C). Since the two managers play an essential role in the development and the maintenance of the community, the choice was made to interview them in person.
For both of the interviews a semi structured approach was used. A list of specific questions was presented to the community managers but unlike a survey, this type of interviewing leaves sufficient room for additional information (Van Aken, Berends and Van der Bij, 2007). Moreover, personal interviews enable the interviewer to make use of follow up questions and collect information by observation (Cooper et al., 2003). Both of the techniques were used in order to enhance the richness of the gathered information.
The community manager of RN was interviewed because he can be considered to be responsible for the system quality and the accessibility of personal information. Therefore, it is important to collect specific information concerning this issue and to have the opportunity to ask follow up questions. The same applies to the community manager of Rabobank SMG, who is responsible for the information quality and the organization of offline meetings. It is valuable to know which thoughts lie behind his acts as well.
Although the community managers are responsible for different aspects of the Rabo Starterscommunity, the same interview was used for both. In this way it was also possible to ask their opinion on the issues they were not directly responsible for, which enabled me to get valuable insights.
Additionally, as a member of the project team information was gathered by (participative) observation. Through project meetings, the community manager of Rabobank SMG was closely watched. During the offline meeting, which was attended by both community managers, their actions were carefully observed. Moreover, a good representation of the co working between Rabobank SMG and RN was obtained, since I was the contact person for Rabobank SMG. Hence, extensive contact through email took place regarding the optimalization of the community.
3.4 Operationalization of the Variables
In order to operationalize the variables, most of the items from the relevant success factors were derived from earlier studies on virtual communities. For the measurement of the items, a five point Likert scale was used, ranging from strongly disagree to strongly agree. The
Cronbach’s Alpha was measured, to indicate if there is a high enough internal consistency. As rule of thumb, the Cronbach’s Alpha should be at least 0,60 before the item scale can be identified as reliable.
Perceived Usefulness. In order to measure the usefulness that members perceive of the virtual
community, a three-item scale was used. All items were derived from an earlier study of Lin (2007). Statements were presented to the members concerning sharing and gaining information and knowledge in the area of entrepreneurship. Combining those items resulted in a Cronbach’s Alpha of 0,64, which indicates a high enough internal consistency for the variable perceived usefulness.
Online Features. To determine the quality of the online features, the level of information and
system quality were measured. In order to get a complete view of the information quality, multiple studies were analysed for the measurement of this characteristic. From these earlier studies of Lin (2007), Lee, Strong, Kahn and Wang (2002) and Liu and Arnett (2000), six items were derived. Members of the community were asked if the information provided by the virtual community was accurate, up-to-date and relevant. Furthermore, it was examined if the information covered the needs, is easy to interpret and is easily retrievable. The Cronbach’s Alpha is 0,79, which indicates a high enough internal consistency. Consequently, the six items can be combined into a single measurement of information quality.
With reference to system quality a four-item scale was used. These four items were drawn from articles of DeLone and McLean (2003) and Yoo et al.(2002) which define characteristics of system quality. The members were asked about what they think about the possibilities the virtual community has to offer, the ease of navigating the website, convenience of access and the user friendliness. With a Cronbach’s Alpha of 0,88 the variable has a high enough internal consistency to combine the four items.
Trust. For the measurement of accessibility of personal information there were no items
available from earlier studies. Many papers deal with the accessibility of personal information and identify this factor as paramount for any virtual community that aims at success. However, the papers that were included in the literature review, did not attend an existing scale. Therefore, a four-item scale was composed in order to be able to measure the accessibility of personal information. The questions referred to the ability of the member to
control their own personal information and to which extent the data should be visible by others. The original Cronbach’s Alpha of the four items was 0,54. This indicates that the four items do not all measure the same thing in this case the accessibility of personal information. By deleting one item (See Appendix B) the Cronbach’s Alpha improved to 0,70, which indicates a high enough internal consistency to combine the remaining three items.
Concerning the level of offline interaction the construction of the scale is based on a study of Koh et al. (2004). The items of the study of Koh et al. (2004) had to be adjusted, since the community of the Rabobank is still a virtual community in development, where the aspect of offline meetings is not fully integrated yet. Accordingly, questions like how often do you meet offline were irrelevant. However, it is interesting to know how people assess the possibility of offline interaction. Therefore, members were asked if they believe that offline interaction could be a supplement to the online interaction.
Initially, the Cronbach’s Alpha appeared to be very low, that is, 0,36. After deleting one item (See Appendix B), the remaining items showed a Cronbach’s Alpha of 0,64, representing a high enough internal consistency to combine the items into a single measurement of level of offline interaction.
3.5 Quality Criteria for Research
In judging the quality of this research attention is paid to reliability and viability.
Reliability. Reliability can be defined as the consistency of the results. The results are
independent of certain features of that study and can therefore be replicated in other studies (Swanborn, 1996). Repeating the measurement would enhance the reliability (Van Aken et
al., 2007), however this goes beyond this research.
The interpretation of the researcher also influences the reliability of the study. Especially, in this case where I was one of the project members and gained partially information by participative observation. For this reason the conducted interviews are more dependent on the point of view of the interviewer than the email survey is (Van Aken et al., 2007). In order to make the dependency of me as the person who has conducted the study as little as possible, a semi structured approach is applied. Van Aken et al. (2007) argue that the more procedures are decided on beforehand, the less the subjective influence of the researcher affects the results.
The choice for a certain research approach can also influence results, because different research strategies do not always obtain the same results. To prevent this kind of biases, multiple information sources have been combined (Van Aken et al., 2007). Next to the survey, interviews were held and information was obtained by (participative) observation. The project gatherings and the offline meeting enabled me to closely observe both the community leaders and to have discussions concerning the optimalization of the community. These discussions were furthermore extended by much email contact.
Validity. The validity of a research is partly determined by the reliability and can be divided
in construct, internal and external validity (Van Aken et al., 2007). “Construct validity refers to the quality of the operationalization of a concept” (Van Aken et al., 2007, p.163). Most of the items of the variables in this research were derived from earlier studies, which enhances the construct validity. On the other hand, there were also items that were composed or adjusted to this research. As a result, this may reduce the value of the construct validity. However, the Cronbach’s Alpha test was applied to verify if the items were forming a correct scale. In some cases, items were deleted.
The internal validity measures if the suggested relationships are adequate and complete (Van Aken et al., 2007). The variables and their relationships, as displayed in the conceptual model, are supported through literature. This enhances the internal validity. However, as a researcher one tends to limit oneselves by focusing on one perspective, the one that suits best. This may reduce the internal validity (Van Aken et al., 2007).
External validity can be defined as the generalizability of the research results (Van Aken et al., 2007). This case study focuses solely on the Rabo Starterscommunity of SMG. Therefore, the results cannot be generalized. Moreover, the results are based on a small population, which is not representative for all the starter communities that exist. However, this problem occurs at more Rabo Startercommunities. To them, the results can be of interest. Nevertheless, the external validity of this research can be considered low, because this study does not aim to develop a generalizable theory (Van Aken et al., 2007).
4. RESULTS
This section discusses the survey, interview and observation outcomes. The survey was used to explore how the members of the Rabo Starterscommunity value the success factors. The interviews and observations provided more insights regarding the involvement of the community managers and how the two cooperated. This section ends by giving an overview of the findings.
4.1 Descriptive Statistics
Table 1 shows the success factors that were used to evaluate the success of the Rabo Starterscommunity. Also perceived usefulness, as success is defined in this research, is included in table 1. The table furthermore gives an overview of the frequencies and of the scores on minimum, maximum, mean and standard deviation. A conscious decision is made to include all those measurements, since simply looking at the means does not always provide the right information.
The figure mean shows that, on a scale of 1 to 5, the members responded relatively positive on the integration of offline and online interaction (3,73) and on the fact that their membership enhances their ability to obtain information from members and to share knowledge with them (3,70). In the case of offline interaction 90% of the responses are clustered around 3,50-4,50. Concerning perceived usefulness 60% of the responses ranged between 4,00-5,00.
Information quality scored somewhat lower with a mean of 3,50. With a minimum score of 2,50 nobody valued the information provided on the community considerably low. The rather low standard deviation and the distribution show that the responses are lying fairly close together. Moreover, the difference between the maximum and minimum score are, together with level of offline interaction the lowest.
The responses concerning accessibility of personal information are divided. Forty percent of the members lean towards an attitude, ranging between 2,00-3,00, where they prefer to control their own information and prefer no full exposure of their personal information. Another 40% looks rather neutral at the accessibility of personal information. The final 20% are not that concerned about their privacy. Accordingly, resulting in an average score of 3,0.
The rather low mean of system quality might implicate that members do not perceive the website of the community quite well. However, it is the standard deviation that gives
some more information. Table 1 displays a standard deviation of 1,15, which indicates that the spread between the given answers varies considerably. This is confirmed by the frequencies, which shows that the responses range between 1,25-4,25. Twenty percent of the members believe that the community is not easy to access and navigate whilst 40 % have a positive attitude towards those factors. This case proves that simply looking at the mean does not always provide the full details.
Table 1: Descriptive Statistics
Score Perceived Usefulness Accessibility of Personal Information Level of Offline Interaction Information Quality System Quality 1.0-1.5 1.5-2.0 2.0-2.5 2.5-3.0 3.0-3.5 3.5-4.0 4.0-4.5 4.5-5.0 - - - 20 % 20 % - 50 % 10 % - - 20 % 20 % 40 % - 20 % - - - - 10 % - 70 % 20% - - - - 10 % 30 % 30 % 30 % - 20 % - 10 % 20 % 10 % - 40 % - Minimum 2,67 2,00 2,88 2,50 1,25 Maximum 4,67 4,00 4,38 4,00 4,25 Mean 3,70 3,00 3,73 3,50 2,90 SD 0,67 0,65 0,39 0,48 1,15 4.2 Correlation Analyses
Data for this research were collected on the basis of a Likert scale. Although, Likert scale data has an ordinal measurement level, they may be treated as interval data (Cooper et al., 2003). This implies that parametric tests can be used. However, the condition that the variables are normally distributed should be met. Therefore, a scatterplot was executed (See Appendix D). However, as the scatterplots show, there was no relationship found. This implies that the collected data have to be analysed on an ordinal measurement level.
A Spearman Rho test was performed to measure the correlation between the variables. Since a supposed relationship between the success factors and perceived usefulness was assumed, a one-tailed test was applied. The results are displayed in table 2.
Table 2: Correlation between the Variables 1. Perceived Usefulness 2. Accessibility of Personal Information 3. Level of Offline Interaction 4. Information Quality 5. System Quality 1 - 0,34 -0,37 -0,93 0,05 2 - - -0,08 -0,03 0.81* 3 - - - 0,33 0,11 4 - - - - 0,25 5 - - - - -
* Correlation is significant at the 0,01 level (one-tailed).
According to the results of table 2 there appears to be no relationship between perceived usefulness and the success factors. The relationship between accessibility of personal information and perceived usefulness could not be proved, because no significant correlation was displayed (Rs = 0,34; p < 0,01, one-tailed). Similarly, level of offline interaction does not correlate either with perceived usefulness (Rs = -0,37; p < 0,01, one tailed). Likewise, no evidence was found for a relationship between information quality and perceived usefulness (Rs = -0,93; p < 0,01, one-tailed). Finally, the table does not show correlation between system quality and perceived usefulness (Rs = 0,05; p < 0,01, one tailed).
Consequently, it is not confirmed that a higher score on the success factors will lead to a higher perceived usefulness among the members.
However, accessibility of personal information and system quality do correlate significantly (Rs= 0,81; p < 0,01, one-tailed). This is striking to see, since this relationship was not suggested.
4.3 Interview Outcomes
The community manager of RN believes that a successful virtual community entails a platform where people meet to share the same passion and obtain information concerning this topic. He emphasizes that passion is the key to a successful community. This applies to both members and managers. Moreover, he states that one can create a platform, however, an actual community evolves from a passion. The community manager of Rabobank SMG adds that the information provided by the community should be independent in order to be successful. He believes people consult a forum, because the information comes from members’ own experience and is therefore experienced as self-governing. A forum should not
leave room for commercial advertising purposes. Consequently, he argues that a forum is “the beating heart” of any virtual community.
Furthermore, both the community managers state that a virtual community should offer her members a place where they can find what they need and can drop their questions which can be answered correctly and fast by the community.
Rabobank SMG, as discussed earlier, is dependent on RN. The community manager of SMG argues that this construction has its pitfalls. He believes that Rabobank SMG should be able to make alterations in the website. However, since this is centralized, local wishes are most of the time not accepted. Furthermore, he considers the added value RN has to offer low which is according to him partly due to the structure of the website. He explains this by stating that RN hardly offers their help by optimizing the local website of Rabobank SMG. He understands that not every adjustment can be implemented, but he concludes that this situation makes the website very statical. The website is offered by RN and that is what Rabobank SMG has to deal with. The community manager of RN sets against this that only a few local banks have joined the Rabo Starterscommunity and only a part of them can be considered as active users. He argues that when you strongly invest in the website, it is only for those four to five banks that are actively involved in the virtual community. This consideration concerning the costs determines whether or not adjustments will be carried through.
Regarding the accessibility of personal information, the community manager of RN believes that it is up to the user what kind of personal information they want to disclose. However, they should be aware of the fact that when they participate in a discussion on an open forum, it is public for everyone. The moderator of Rabobank SMG argues that members should have control over their own personal information. Furthermore, the privacy ought to be guaranteed by the virtual community. This means that personal data should not be given to third parties or used for the benefit of the Rabobank. Besides, anonymous registration should be possible. Even though the need for anonymity might be less the case in relation to the Rabo Starterscommunity, due to the commercial aspect.
The community manager of RN considers virtual communities that combine offline and online interaction, as most successful. The moderator of Rabobank SMG knows from others and his own experience that offline meetings catch on. However, he admits that it did not help the community of the Rabobank so far. He argues that the power of repetition might be applicable here.
On the subject of information quality the community manager of Rabobank SMG believes that within the possibilities that the website has to offer, the presented information is sufficient. However, he does prefer to have some more general information regarding starting a company on the local homepage, such as displayed on the homepage of the national community (See appendix A). This issue also came up during the offline meeting which was organized by Rabobank SMG. Nevertheless, no comply was given to this request. The community manager of RN admits that the members are little involved in the development of the website. He argues that this is the case because he is dependent on the possibilities the system has to offer and the budget he has to work with.
Furthermore, the community manager of Rabobank SMG sees the forum as “the beating heart” of the Rabo Starterscommunity, where people can share and obtain information and start discussions with fellow starters, experts and Rabobank staff. These characteristics of a forum enable members to enhance their performance in information exchange and thereby in gaining perceived usefulness. Therefore, more participation from the business advisors is required. Due to the fact that social media is not imbedded in the organization yet and the business advisors have to do this in addition to their normal work, participation is low. The community manager of RN agrees on this issue. Accordingly, he believes that the community should offer its members the information one could expect from the Rabobank concerning starting a business. This requires that the business advisors will use their expertise on the community by, for example, providing useful tips. Although he considers this as important, he realizes this is not happening. He believes it is up to the local banks to provide useful information to their members. The community manager of SMG agrees with him. However, he does believe that the local banks are partly dependable of the website, which is offered by RN.
Although, the community manager of Rabobank SMG acknowledges that the current website lays a good foundation for a successful community, he emphasizes that it is corrigible. Especially with regards to the ease of navigating within the community and the level of user friendliness. The community manager of Rabobank SMG thinks that with some minor adjustments, the members will perceive the website as more appealing. Furthermore, he considers the findability of the community, for example via google, quite good. Nevertheless, within the community he finds the intermediate step to go from the national website to the region page of Rabobank SMG not logical. According to him, this decreases the level of accessibility of the Rabo Starterscommunity SMG. The community manager of RN states
that he is limited by possibilities of the system and budget that he has at his disposal. The choice for implementing suggestions from local banks is among others based on this.
4.4 Information Obtained from Observation
During my internship I had, as a member of the project team, a good view of the activities that were done in order to keep the Rabo Starterscommunity running. This role provided me valuable insights in the field of community management and the cooperation between Rabobank SMG and RN. Moreover, it showed me how important the community was found to be by the community manager, the other members of the project team and the business advisors.
The project had a flying start with meetings that were held on a regular base. In those gatherings the key question was what should be done in order to optimize the Rabo Starterscommunity. A number of ideas were opted to attract and hold members. Some ideas were implemented. In addition, posting blogs by entrepreneurs who already proved themselves in business, posting news items and place statements in the forum. Other ideas were beyond the power of Rabobank SMG to execute. As discussed earlier, there is a certain dependency of Rabobank SMG in relation to RN regarding the development of the website. Therefore, the evaluation about the usability and availability of the Rabo Starterscommunity were fed back to RN. However, despite the fact these suggestions were based on firm ground none of the requests were complied.
Last summer, Rabobank SMG organized an offline meeting in order to discuss the working of the website with the users and enable them to network, which hopefully would be continued online by using the Rabo Starterscommunity. The community manager of RN was present as well and received a number of tips to upgrade the website. However, again these suggestions for improving the online community were not picked up by the community manager of RN. More and more the feeling grew with the project members that they were running up against a wall. The dependency of RN was visible and the project team realized over time, that this was not going to change. In the end their motivation took a knock. Despite the fact that the offline meeting was received with open arms by the starters, not another offline meeting took place since then.
Furthermore, the local community manager, who could be considered as a dedicated community manager, left the project team of the Rabo Starterscommunity. He was succeeded by the team leader of the business unit, which was a new created function. One of her job
responsibilities concerned the Rabo Starterscommunity. However, due to her busy schedule, the community got a low priority. As a result, the community was running less than before. These changes and accompanied observations were obtained after the interviews were conducted. Nevertheless, I believe this is worth mentioning, because this practice showed that when a community lacks a dedicated leader, this is not without any consequences, as was underlined in the literature. This might prove the importance of effective community management in case of the Rabo Starterscommunity.
From the moment the switch of community managers took place the activity on the site ran down. From both, project team ánd users.
From the observations it became clear that the majority of people to whom it concerned, had little faith in the Rabo Starterscommunity. As a consequence the people who are needed to make a success out of the community, like the business advisors and their management and later on the new community manager of SMG, did not (fully) support the virtual community. None of the community managers were able to take away the suspicion of those people towards the community.
4.5 Overview of the Relevant Findings
Results from the survey showed that the success factors did not correlate with perceived usefulness. However, the descriptive did gave insight into how members valued perceived usefulness and the success factors.
The integration of online and offline interaction was positively evaluated with 90 % of the responses clustered around 3,5 – 4,5. This implies that the majority of the members are in need of an offline meeting. This result is confirmed by the observations made. Starting entrepreneurs who attended the offline meeting, were very enthusiastic about the gathering. While both the community managers are aware of the use of offline interaction, an offline meeting only took place once.
Furthermore, members perceived the community as quite useful, despite the rather low and dispersed evaluation of system quality. From the interviews it became clear that system quality is hard to manage in case of the community manager of SMG, due to the dependency towards RN. The community manager of RN is limited by budget and system and any development concerning the website is based on those factors. On the contrary, the community manager of SMG thinks from the members’ perspective by trying to provide a
