The Perception of Virtual, Inter-Team Collaboration Within a Large, Multi-Team Project

Table of Contents

Introduction

1.1 Background of Study
1.2 Research Purpose and Research Question
1.3 Thesis Outline

2 Literature Review
2.1 Collaboration in Organizations
2.2 On Virtual Teamwork
2.3 On Virtual Inter-Team Collaboration
2.3.1 Shared Meaning
2.3.2 Trust
2.3.3 Boundary Objects and Team Boundary Spanning
2.4 Summary of Literature and Research Gap

3 Methodology
3.1 Research Context
3.2 Research Philosophy
3.3 Data Collection
3.3.1 Semi-Structured Interviews
3.3.2 Observations
3.4 Limitations and Creation of Quality
3.5 Data Analysis

4 Presentation of Empirical Data
4.1 Collaboration as Everything (and Nothing)
4.1.1 Glorification of Collaboration
4.1.2 Importance of Big Picture
4.1.3 Communication as Key
4.2 Inter-Team Collaboration Experienced as Burden
4.2.1 Inter-Team Collaboration as a Challenge
4.2.2 Inter-Team Collaboration Increases Work Pressure
4.2.3 Inter-Team Collaboration Through Representatives
4.3 Inter-Team Collaboration as ‘Us versus Them’
4.3.1 Importance of Trust
4.3.2 Different Priorities
4.4 Summary of Findings

5 Discussion
5.1 Collaboration as an Empty Word
5.1.1 Discourse Detached from Practice
5.1.2 Working Together Is Not Necessarily Collaboration
5.2 Inter-Team Collaboration as Competition
5.2.1 Collaboration Infrastructure as an Enabler of Inter-Team Competition
5.2.2 Boundary Objects as Enablers for Coopetition
5.3 Creation of Subgroups in Inter-team Collaboration
5.3.1 Competing In- and Out-Groups
5.3.2 'Collaboration Spanners' Across Subgroups

6 Conclusion and Outlook
6.1 Brief Study Recap
6.2 Main Findings
6.3 Theorizing Virtual Inter-Team Collaboration
6.4 Practical Contributions

Abstract

Title: Virtual inter-team collaboration within a large, multi-team project

Authors: Lisa Prigge and Lena Kreibel

Purpose: The purpose of this thesis is to achieve a more nuanced understanding of inter-team collaboration and of how collaboration between teams is experienced by individual team members.

Methodology: Our research includes a qualitative study, which was conducted by using an abductive approach. Throughout our research, we adopted an interpretative perspective.

Methods: The empirical material was gathered in thirteen semi-structured interviews with members of five different teams of a large, multi-team project as well as four observations of inter-team meetings.

Conclusion: In general, collaboration seems to be experienced in many different ways by the interviewees, resulting in it becoming an empty word as no common understanding of what collaboration comprises is prevailing. In practice, inter-team collaboration is perceived as a burden. The virtual infrastructure has a major influence on this perception, not only as an enabler for inter-team collaboration, but also as a restraining factor for inter-team collaboration. Further, inter-team collaboration involves competitive thinking between the teams. The feelings of team belongingness, trust, and personal relationships play a major role in the establishment of competitive thinking through the emergence of in- and out-groups in and among teams. The simultaneous existence of collaboration and competition leads to the emergence of coopetitive structures. Many boundary objects increase the workload for employees so that collaboration with other teams is passed on to certain individuals of respective teams. We define these individuals as ‘collaboration spanners’. The latter, in addition to competition as an integral part of collaboration, potentially provide areas for further research.

Keywords: Inter-team collaboration, virtual teamwork, competition, coopetition, boundary spanning, multi-team project, perception, in- and out-groups

Acknowledgements

We would like to express our gratitude to our supervisor for her support, assistance, and valuable feedback throughout the writing process.

Moreover, a special thanks goes to our research participants and their interest in participating in our study. We thank all our interview partners from Beneal Bank for letting us participate in their meetings and their daily work. Without their openness and time investment, this project and paper would have never succeeded.

Thank you!

List of Tables

Table 1: Interviewee Sample 28Table 2: Observation Sample

List of Figures

Figure 1: Different Ways of Working Together (Adapted from Bedwell et al., 2012; Elliott, 2006; Ghobadi & D’Ambra, 2012)

Introduction

1.1 Background of Study

Collaboration as a concept in the workplace is nothing new. Today seemingly more than ever, it is omnipresent in explaining the success of businesses. The internet is riddled with articles titled ‘ The Importance of Collaboration in the Workplace ’ (Boyer, 2018) or ‘ 5 Reasons Why Collaboration Is Essential in Today's Business Environment ’ (Nixon, 2014). The common denominator in these articles is that collaboration is required to a greater extent today than in the past due to the evolution of organizational designs.

Traditional organizational designs are, according to Fjeldstad, Snow, Miles, and Lettl (2012), described and explained through hierarchy. In this context, hierarchy represents a method of organizing work, with which goals are set and fulfilled, resources are allocated, and interdependencies are managed. In this context, higher-level organizational members possess, compared to lower-level ones, the authority and capability related to control and organization to resolve conflicts at lower levels within the organization (Fjeldstad et al., 2012). This is based on the historical notion of the hierarchical concept, stating that higher hierarchical levels ought to have a broader view of the organization and its environment which the lower hierarchical levels are not in need of (March & Simon, 1958).

The global proliferation of communication, financial, and logistics services during the twenty-first century resulted in a more complex, global business environment and the need for immediate responses of organizations to external opportunities and challenges (Fjeldstad et al., 2012). To cope with such new business environments, organizations were redesigned because the usefulness of hierarchy as primary mechanisms of control and coordination decreased. As a result, new and more collaborative organizational designs emerged (Fjeldstad et al., 2012). Common to these designs is that more organizational actors have access to more resources that allow them to self-organize and form suitable collaborative relationships (Benkler, 2002). According to Fjeldstad et al., (2012), organizational control and coordination are based on direct exchanges among organizational actors themselves rather than on hierarchical planning. An example of this is, what practitioners call, the Agile Methodology. The latter is an organizational framework based on self-organized teams that are required to collaborate to achieve a common goal (see Research Context, section 3.1 for detailed explanation). Thus, teams and individuals are increasingly responsible to collaborate to get work done by managing their common resources and goals (Ostrom, 1990). This explains the increasing importance of the concept of collaboration in today’s business environment and the growing need for it to be subject to scientific research.

1.2 Research Purpose and Research Question

Our research purpose is twofold and comprises theoretical and empirical purposes. Theoretically, we aim to contribute to the clarity of the concept of collaboration by conducting research on virtual inter-team collaboration. Many scholars have investigated team collaboration, either without specifically differentiating intra- and inter-team collaboration or focusing only on intra-team collaboration. Intra-team collaboration means collaboration taking place within the context of one team, whereas inter-team collaboration describes the interaction process between several different teams (Longoria, 2005). For instance, many researchers described elements of collaboration for effective work on an intra-team level (Horwitz, Bravington & Silvis, 2006; Bjørn & Ngwenyama, 2009). Others studied the impact of collaboration on intra-team conflicts, knowledge sharing, and trust (e.g., Weingart & Jehn, 2015; Webster & Wong, 2008; Alsharo, Greggb & Ramirez, 2017). Although collaboration has also been studied specifically on an inter-team level by some scholars, it was only done in relation to different leadership designs and how these influence the collaboration between teams (e.g., Ogren, 2016; Cha, Kim, Lee & Bachrach, 2015). Thus, in the field of organizational research, we need to advance our knowledge on inter-team collaboration, as it is a way of working that becomes more and more popular and widespread (Ostrom, 1990; Bodrozic & Adler, 2018; Chudoba, Wynn, Lu & Watson-Manheim, 2005).

Our practical aim is to understand collaboration in such emerging business contexts. With more resources available, including technological tools, for individual actors and teams, new ways of collaboration become apparent. For instance, organizations operating in several geographical locations often need teams to work on one project. The resulting virtual teamwork leads to employees experiencing collaboration differently than in co-located work settings (Gratton & Johns, 2013). We want to contribute to a better understanding of collaboration in the modern sense by investigating how inter-team collaboration is experienced by employees working in the same organization. This may include identifying potential distinct factors in the perception of inter-team collaboration compared to intra-team collaboration or determining the impact of the virtual work environment on the perception of collaboration across teams. We especially want to highlight that although the individuals under study are separated into different teams, they still belong to the same organization and work towards a common goal. We believe that being assigned to different teams while still being part of the same organization is especially interesting to investigate.

We do this by studying the collaboration between different teams in a large, financial services company called Beneal Bank (the company name has been changed). The studied organization resembles one of the previously illustrated new organizational designs. More than 80 interdependent teams, including geographically dispersed team members, work on an intra-organizational project to develop a new core-banking system. Thus, our study site has certain characteristics that justify it as a site in which new organizational designs are applied and in which new ways of collaboration are required. Firstly, the organization employs an Agile Methodology, implying that close and continuous collaboration is needed between different, interdependent teams. As members of different teams, all individuals belong to the same company and strive for an overall goal, which will be described further in the Methodology (see Research Context, section 3.1). Secondly, collaboration mainly takes place virtually as team members are geographically dispersed. An increasing number of companies in today’s business environment can be characterized by these two features, which is why our aim is to study collaboration in the context of a virtual, multi-team project setting.

Considering the above explanations, the purpose of our research results in the following research question:

How do employees experience virtual inter-team collaboration within a large, multi-team project?

1.3 Thesis Outline

The research question is answered throughout the course of the following five chapters: Literature Review, Methodology, Findings, Discussion, and Conclusion. Following this introduction, the literature review aims to provide an overview of existing research on the topic. Moreover, it supplies background knowledge which is referred to in the discussion part, in order to position our findings within existing research. We explain collaboration as a concept in which two or more social entities actively and reciprocally engage in joint activities to achieve a common goal. The related concepts of coordination and cooperation are also illustrated, as well as the opposing concept of competition. In this context, we review the concept of coopetition, which is related to competition and collaboration. The literature review is then tailored to our research purpose, by reviewing collaboration in the context of virtual work. Shared understandings, trust, and boundary spanning are important concepts that we review. Chapter three, namely the Methodology, presents how the research has been conducted. At the beginning of this section, a detailed description of our study site is given to better understand the research context. More specifically, it is described how our interpretative stance guided this qualitative study. Furthermore, this chapter reflects upon the limitations of the research and on how these may have impacted the findings.

In chapter four we present the findings from the data collection process. Quotes taken from interviews are used to exemplify the findings. Firstly, we portray the diverse understandings of collaboration among the employees. Collaboration is almost ‘glorified’ by our interviewees, giving the impression that the concept seems to be detached from practice. They further perceive two aspects, of which they talk about on a rather superficial level, as most important for inter-team collaboration: having a big picture of the overall project and communication. Yet, we find that inter-team collaboration seems to be perceived as a burden which increases the workload for employees. To deal with the burden of collaboration and facilitate it, we identify that different representatives are used to collaborate with other teams. Lastly, we highlight the feeling of ‘us versus them’ among the employees. Factors enabling this feeling are the lack of trust, team belongingness, and differing priorities of the individual teams.

Based on the findings and our analysis, as well as the literature review, chapter five seeks to analyze and discuss our findings on a broader level, identifying how our research fits into the existing literature of collaboration. In this chapter, we begin with unrolling how and why collaboration is an empty word in the reality of employees in contemporary organizations. Further, we elaborate on how inter-team collaboration turns into the very opposite: inter-team competition, which leads to the emergence of coopetitive structures. Lastly, we discuss how certain individuals develop into the role of collaborators on behalf of the respective teams. We introduce and define these individuals as ‘collaboration spanners’. To sum up the research, the practical and theoretical implications of the main findings are condensed in the final, concluding chapter six. We suggest that especially competition, its development, and impact on inter-team collaboration may provide opportunities for further research with the aim of enhancing the clarity of the concept of collaboration.

2 Literature Review

In line with the evolution of different management designs, the understanding of collaboration as a tool for organizing has changed over time. The literature review presents how different researchers understand the concept of collaboration, followed by a description of the related concepts of coordination, cooperation, competition, and coopetition. The chapter ends with a presentation of what collaboration means in a virtual context. In this context, the concepts of boundary objects and team boundary spanning activities become important and are elaborated on.

2.1 Collaboration in Organizations

Bedwell, Wildman, DiazGranados, Salazar, Kramer, and Salas (2012, p.130) define collaboration “as an evolving process whereby two or more social entities actively and reciprocally engage in joint activities aimed at achieving at least one shared goal”. Graham and Barter (1999) stress that collaboration consists of interpersonal interactions and relationships that change over time. This is supported by Gray (1989, p.15) who adds that collaboration is an “emerging process rather than a prescribed state”. This process occurs between different entities including individuals, teams, units, departments, functional areas, and organizations (Longoria, 2005). For entities to collaborate, active, mutual engagement is required from all involved parties (Longoria, 2005). For the purpose of our study, we focus on collaboration that takes place between teams or individual members of different teams in an intra-organizational setting.

The elements of the actor-oriented scheme, that is introduced by Fjeldstad et al. (2012), can be considered as preconditions for effective collaboration to happen on an intra-organizational level. According to Fjeldstad et al. (2012), the actor-oriented scheme composes three elements which influence the way in which collaboration takes place in organizations: (1) actors who have the capabilities and values to self-organize, (2) commons that the actors share, for instance resources, knowledge, and understanding of the situation, and (3) protocols, processes, and infrastructures that enable multi-actor collaboration. Infrastructures in this context are systems that allow actors to connect with one another and access the same information, knowledge, and resources (Fjeldstad et al., 2012). Protocols mean codes of conduct used by organizational actors in their exchange and collaboration activities (e.g., searching for collaborators, advertising problems) (Fjeldstad et al., 2012). Bedwell et al. (2012) also state that researchers argue that sharing resources to work towards a common goal is the main aspect separating collaboration from other ways of working together.

Other ways of working together include coordination and cooperation, which are often used interchangeably with the term collaboration (Bedwell et al., 2012). Cooperation can be seen as an enabler for collaboration because it is about the attitude towards being concerned about the overall goal rather than individual goals (Salas, Sims & Burke, 2005). Elliott (2006, p.40) puts it differently and describes that “collaboration is cooperation with the addition of collective creativity”. He further explains that cooperation includes parties being involved in some procedure without generating new or alternative processes, ideas, or objectives. Conversely, collaboration means that parties collectively generate multiple solutions which must be developed and selected by everyone involved (Elliott, 2006).

Coordination is different from collaboration in the sense that there is no need for active and reciprocal participation by entities (Salas, Burke & Cannon-Bowers, 2000). It is about ensuring “that tasks are integrated, synchronized and completed within established temporal constraints” (Salas, Burke & Cannon-Bowers, 2000, p.342). According to Elliott (2006), coordination does not include any form of problem solving and creative process. It is rather “a fundamental enabling requirement for all collective activities” (Elliott, 2006, p.40).

Collaboration as a broader concept encompasses both, coordination and cooperation, where each is necessary for one another (Bedwell et al., 2012; Elliott, 2006). As collaboration means sharing resources and knowledge, as well as establishing dependencies with others, there is a risk that the collaboration partners may develop competitive mindsets and behave opportunistically in striving for own successes (Ghobadi & D’Ambra, 2012). Competition describes “a situation in which someone is trying to win something or be more successful than someone else” (Cambridge Academic Content Dictionary, 2018). Therefore, competition and collaboration can be seen as two opposing ends on a continuum illustrating ways of working together, with coordination and cooperation in between (Figure 1).

Abbildung in dieser Leseprobe nicht enthalten

Figure 1: Different Ways of Working Together (Adapted from Bedwell et al., 2012; Elliott, 2006; Ghobadi & D’Ambra, 2012)

Chen (2008) argues that collaboration and competition shall be viewed as interrelated and interdependent concepts. To account for this interplay, coopetition has been introduced as a concept describing situations in which different organizational teams are required to cooperate. At the same time, they are likely to experience tension caused by diverse professional philosophies and competing goals from different, cross-functional representatives (Ghobadi & D’Ambra, 2012). According to Le Roy and Fernandez (2015), organizations are increasingly encouraged to adopt coopetitive strategies due to, for instance, fast-paced developments in IT (information technology) and shorter product life cycles that require companies to work more closely together to achieve competitive advantage over others. Fernandez, Le Roy, and Gnyawali (2014) identify sources of resulting tensions at three levels: inter-organizational, intra-organizational, and inter-individual. The intra-organizational tensions are relevant for our study as they describe the competition among internal teams over human, technological, and financial resources. On an inter-individual level, coopetition leads to the dilemma of either following collaborative principles or competitive ones in decision-making. Therefore, coopetitive structures can create challenges in the interaction between different teams and individuals (Fernandez, Le Roy & Gnyawali, 2014).

2.2 On Virtual Teamwork

With changing organizational designs, new forms of working become apparent in today’s business environment, requiring different forms of collaboration between entities (Fjeldstad et al., 2012). Tasks are becoming more computer-mediated, which allows work to be carried out over computer networks and reduces the need for co-locations (Chudoba et al., 2005). According to Gratton and Johns (2013), the evolution of work models can be illustrated in three waves of virtual work. Beginning in the early 1980s, the first wave was marked by the emergence of virtual freelancers. The increased use of emails and the resulting connectivity allowed individuals to set up a ‘one-person shop’ and to provide services such as translations, design or report writing from outside an organization. This connectivity also allowed marginalized groups, for instance, stay-at-home parents, caregivers, retirees, or students, to enter the labor market (Gratton & Johns, 2013). The second wave of virtual work brought forth an increasing number of virtual corporate colleagues. In times of globalization, companies required employees to work unusual hours and to work on projects with international colleagues and customers (Gratton & Johns, 2013). The third wave of virtual work, that we witness today, is marked by virtual teamwork (Gratton & Johns, 2013). It is, also, according to other researchers, a current topic in the literature on global organizations (Chudoba et al., 2005).

Virtual teams can be described as “groups of geographically ... distributed participants who collaborate towards a shared goal using a combination of information and communication technologies (ICT) to accomplish a task” (Bjørn & Ngwenyama, 2009, p.228). Orlikowski (2002) adds that this includes people working across several boundaries, namely across time zone differences, internal business units, and across cultures. Lipnack and Stamps (1997) as well as Jarvenpaa and Leidner (1999) identify three particular dimensions that define virtual teams: the degree of dispersion, the degree of homogeneity, and the level of continuity. The first determines the geographical distance between team members, the second is based on the extent to which team members feel and are similar to one another (e.g., by belonging to the same organization), and the last is determined by the time that the virtual team has already been established. High degrees of dispersion, for instance, may lead to the fact that team members never meet face-to-face and thus comprise ‘pure’ virtual teams (Chudoba et al., 2005). In other cases, so-called ‘semi-virtual’ teams may consist of some co-located as well as some remote members (Webster & Wong, 2008).

Especially in large organizations operating in different geographical locations, virtual inter-team collaboration becomes important (Chudoba et al., 2005). Today, organizations increasingly realize that more virtualization results in less natural team collaboration and a lack of community feeling (Gratton & Johns, 2013). This may lead to the finding that especially in semi-virtual teams, co-located members form subgroups in which a community feeling is established, which the following section deals with in further detail. Virtual teams must manage the ongoing challenge of collaborating at the very center of teamwork (Bjørn & Ngwenyama, 2009).

2.3 On Virtual Inter-Team Collaboration

This chapter highlights important aspects influencing inter-team collaboration in virtual environments. These aspects, which are described in the following sections, include the importance of creating 1) a shared meaning, 2) trust, and 3) boundary objects or team boundary spanning activities.

2.3.1 Shared Meaning

Marheinekea, Velamurib, and Möslein (2016, p.1109) understand virtual collaboration as “an intertwined process comprising a technical and a social dimension”. It includes the social interrelationships between the collaborating people and the technical infrastructure that enables collaboration. Most literature focuses on the social dimension of virtual collaboration. For instance, Horwitz, Bravington, and Silvis (2006) studied the perceptions of managers as well as team members regarding team effectiveness when being geographically dispersed but connected through information technology. The study (Horwitz, Bravington & Silvis, 2006, pp.474) finds that especially “social dynamic risks associated with building and sustaining team and organisational commitment, team relationship difficulties like trust, cross-cultural differences, member conflict, role ambiguity, and complex decision-making issues including interpretive problems relation to decisions” are apparent for people collaborating virtually.

Most scholars agree that the result and purpose of collaboration in the context of virtual work is about the creation of a shared meaning and understanding. Schrage (1995, p.33) states that collaboration in the context of virtual teamwork is meant to “create a shared meaning about a process, a product, or an event ... It can occur by mail [or e-mail], over the phone lines, and in person”. Most of the literature states that the establishment of a shared meaning and understanding is a major part of virtual team collaboration. Schrage (1995, p.33) describes virtual collaboration as “two or more individuals with complementary skills interacting to create a shared understanding that none had previously possessed or could have come to on their own”. Gibson and Cohen (2003, p.23) define shared understanding as “a collective way of organizing and communicating relevant knowledge, as a way of collaborating”. They further state that similar backgrounds, shared experiences, opportunities to learn about each other over time, communication, information sharing, and development of a team spirit helps to create shared understandings.

Bjørn and Ngwenyama (2009) mention similar features that provide the basis for shared meanings in virtual teams. They categorize these features into three analytical levels: (1) the lifeworld; (2) organizational structures; and (3) work practices. The lifeworld includes taken-for-granted background knowledge and assumptions (Bjørn & Ngwenyama, 2009). It, thus, serves as an implicit point of reference for the behavior of individuals, shaping their interpretations and giving meaning to events and situations. Organizational policy, norms, and procedures comprise the organizational structure through, for instance, symbolic artefacts, ritual activity, and patterned behavior (Bjørn & Ngwenyama, 2009). Forming of virtual teams can be described as creating new organizational structures that need to be negotiated and understood. Moreover, people develop different work practices, in terms of, for instance, the usage of language and speech based on previous workplace experiences which need to be synergized. All three analytical levels can contribute to establishing shared meaning by providing common stocks of knowledge, rules, and resources which team members can use to interpret each other’s actions (Bjørn & Ngwenyama, 2009).

According to Bjørn and Ngwenyama (2009), the concept of translucence can be used in uncovering important but invisible social clues to enable distributed individuals to interpret each other’s actions during collaboration. It is a “feature of collaborative technologies, and should be provided in a low-effort, seamless way that does not interfere with the user’s primary task” to limit communication breakdowns (Bjørn & Ngwenyama, 2009, p.232). These breakdowns become visible in delays or interruptions of current activities. Three characteristics are important in providing translucence: Accountability, visibility, and awareness. The fact that actions are visible and can be tracked by others in the virtual collaboration environment should lead to an increased awareness of and adaptation to each other’s actions and thus to an increased shared understanding (Bjørn & Ngwenyama, 2009).

2.3.2 Trust

Besides a shared understanding, many researchers highlight trust in relation to knowledge sharing and the establishment of personal relations as a crucial factor in virtual team collaboration. For instance, a study by Brown, Poole, and Rodgers (2004) shows that trust affects the willingness to collaborate and the sustainability and productivity of the collaboration within virtual teams. Kanawattanachai and Yoo (2002) argue that virtual team members rely more on cognition-based trust than on affect-based trust, basing their decision to trust someone on perceptions of evidence of trustworthiness and not on genuine caring and emotion. Cognition-based trust is built around the elements of reliability, integrity, competence, and responsibility; whereas affect-based trust is based on the concern about the welfare of the other and sensitivity about the other’s needs (Kanawattanachai & Yoo, 2002). Alsharo, Greggb, and Ramirez (2017) argue that knowledge sharing represents an important behavior determining to what extent trust is developed in any kind of virtual team. By sharing knowledge, team members can show their ability, integrity, and benevolence to others, serving as evidence for their trustworthiness and as a precondition to develop trust (Alsharo, Greggb & Ramirez, 2017). A link can be made here to Bjørn and Ngwenyama’s (2009) concept of translucence; in translucent collaboration systems, accountability is established, which ensures cognition-based trust to be established between different parties.

Affect-based trust is likely to be established by regular face-to-face contact and establishing personal relations (Kanawattanachai & Yoo, 2002). Therefore, trust is more likely to be developed between co-located employees. Because of this, Webster and Wong (2008) argue that co-located employees in semi-virtual teams may form ‘in-groups’, separating them from the other, remotely working employees, which become ‘out-groups’. In-group members develop more trust in and identify more strongly with each other. This is because information is exchanged more easily, which reduces uncertainty (Webster & Wong, 2008). The formal categorization into certain subgroups (e.g., into different teams) can, according to Kramer (1999), already create a climate of distrust between these subgroups. The creation of in- and out-groups increases the likelihood of developing favoritism towards certain individuals and/or subgroups, including the likelihood of collaborating more frequently and effectively with certain parties and less so with others (Webster & Wong, 2008).

2.3.3 Boundary Objects and Team Boundary Spanning

In virtual collaboration, boundaries appear when new, complex information is exchanged and the need for crossing the barrier of the boundaries emerges. Marheinekea, Velamurib, and Möslein (2016) highlight the importance of boundary objects in such situations. Boundary objects are “any form of an artefact individuals use to collaborate with each other with the aim of overcoming the mutual distance of knowledge domains” (Marheinekea, Velamurib & Möslein, 2016, p.1108). Examples of boundary objects are shared information systems, project management tools, and 3D modeling tools (Marheinekea, Velamurib & Möslein, 2016). In the social dimension, which constitutes the social interrelationships between collaborating actors, boundary objects influence knowledge creation of people and their relationships with each other. In the technical dimension, boundary objects influence the fulfillment of tasks and the technology used (Marheinekea, Velamurib & Möslein, 2016).

This interplay is exemplified by Gibson and Cohen (2003), who say that although available technology in virtual teamwork enables knowledge sharing, there is a need for social capital and a social network which the knowledge can be shared with and that individuals can employ to acquire knowledge. By aligning the social dimension, that is the people and structures, and the technical dimension, meaning the technology and task, of the collaboration process, boundary objects transfer knowledge between differing social domains. Through this knowledge transfer, they “negotiate meaning to establish a shared understanding between the two formerly dispersed domains” (Marheinekea, Velamurib & Möslein, 2016, p.1110).

The article of Marheinekea, Velamurib, and Möslein (2016) also highlights that certain boundary objects are more important than others. Boundary objects to filter information are especially necessary for virtual teamwork as technology can lead to information overload for employees. Moreover, boundary objects offering visual representations positively influence the creation of shared understanding. For effective knowledge transformation, a mix of boundary objects such as PowerPoint slides combined with video- and audio-conferencing is useful (Marheinekea, Velamurib & Möslein, 2016).

The concept of team boundary spanning can be used to specifically describe the collaboration between different teams. Team boundary spanning includes “the team’s actions to establish linkages and manage interactions with parties in the external environment” (Marrone, 2010, p.914). Actions can, for instance, include representing the team to outsiders (e.g., updating and seeking feedback on team progress), searching for information (e.g., contacting others for project-related expertise), and coordinating task performance with other teams (e.g., communicating plans and delivery deadlines with other interdependent teams). Boundary objects, thus, influence how virtual team boundary spanning takes place. Often, individual team members take over the role as a collaborator on behalf of the team through team member boundary spanning.

Research found that this role may be perceived as challenging for individuals as they continuously balance between internal and external demands and face different internal and external expectations, leading to role stress (Marrone, 2010). Member self-efficacy, role responsibilities, and team external focus can explain why individuals become team boundary spanners and others not. Marrone (2010) also points out that there are few studies to date that have investigated the benefits of and the reason for a certain number of boundary spanners in a team (e.g., the use of a single boundary spanner versus shared boundary spanning responsibility across all members).

2.4 Summary of Literature and Research Gap

To conclude this literature review, a brief summary is given of the existing research on the concept of collaboration, and specifically in the context of virtual, inter-team work. Teams are increasingly responsible for collaborating with each other in today’s business environment. Collaboration means to “actively and reciprocally engage in joint activities aimed at achieving at least one shared goal” (Bedwell et al., 2012, p.130). If viewed on a continuum, the concept of competition is often placed on the other end, with cooperation and coordination as other types of working together in between. Research agrees that collaboration is especially challenging in the context of virtual work. Virtual or semi-virtual teams must manage the challenge of creating shared understanding, building trust, and transferring knowledge within and between teams (Bjørn & Ngwenyama, 2009; Gibson & Cohen, 2003; Brown, Poole & Rodgers, 2004; Alsharo, Greggb & Ramirez, 2017). Especially co-located employees in semi-virtual teams may form ‘in-groups‘, which impacts collaboration because favoritism is established (Webster & Wong, 2008). Furthermore, trust is an important aspect in virtual collaboration; especially cognition-based trust (Kanawattanachai & Yoo, 2002). Boundary objects and team boundary spanning are identified as useful mechanisms to overcome different knowledge domains, and thus collaboration barriers among different teams (Marheinekea, Velamurib & Möslein, 2016; Marrone, 2010).

This literature review shows that most researchers either do not distinguish between levels of collaboration or focused solely on collaboration on an intra-team level. As previously outlined, our research fills this gap by investigating virtual collaboration between teams specifically. With an increasing number of large organizations operating in several geographical locations and employing teamwork as main modes of organizing, it is important to more clearly differentiate what collaboration means in this context. We see the need for organization studies to advance knowledge about how different teams of an organization work together, as our literature review highlights several aspects that influence this. We are particularly interested in identifying the experiences of involved employees and how these influence the process of collaborating with other teams. Therefore, our study aims to answer the research question of how virtual inter-team collaboration is experienced by employees as members of the same organization working in a multi-team project.

3 Methodology

In this chapter, we present our methodological approach and our method – based on a qualitative case study – in order to answer our research question. In the following, the research context of this case study is described, followed by the research philosophy, which encompassed and directed the research approach and strategies. The methods of data collection and analysis are presented, ultimately leading to a description of how the quality of the research was ensured.

3.1 Research Context

To investigate collaboration between virtual teams – an emerging way of working in many organizations – we chose a bank as the site of our qualitative case study. This bank (Beneal Bank) is part of a large financial services group in Europe which operates in 17 countries. In light of the digital revolution and changing customer behavior in the banking industry, the company initialized a five-year program aimed at simplifying the banking experience for customers. This simplification program includes different components. A key component of the program is the replacement of the ‘Core Banking System’ (CBS). The CBS processes customer information, savings, deposits, transaction, and loan products in all business areas. The goal of the CBS program has been to replace the old banking platforms, reduce IT complexity, and transform the related processes, products, governance, and culture. 747 employees within more than 80 teams work on this project.

Our research is embedded in the CBS program, which is executed with, what practitioners call, an Agile Methodology. This concept of working ‘agile’ developed through communities of practice within the field of software development, in order to find new, more efficient ways of getting the work done (Agile Methodology, 2013). Incremental, iterative work sequences known as sprints, which usually last for two weeks, are used to achieve certain sub-goals of the project (Agile Methodology, 2013). At Beneal Bank, six sprints comprise a Program Increment (PI). A PI is a timeframe during which the teams work on a certain product, which should be working and tested at the end of the PI. The goals and corresponding activities that are to be achieved by each team are tracked in a Sprint Backlog. At the start of a PI, a co-located planning meeting of all involved teams takes place to define the team’s Backlogs. Furthermore, each team conducts daily huddles which are about quickly getting together to form a plan for the day. After each PI, a Retrospective takes place to review the process for further improvement.

The teams in the CBS program are divided into different ‘delivery streams’, that are responsible for certain tasks within the project. The delivery streams include a Stream Leadership Team, which ensures the flow and alignment within the stream, and different Cross Functional Feature Teams. These teams include a Scrum Master, who does not act as a typical Project Manager in the sense of telling the team what to do. A Scrum Master rather facilitates and advises the team so that the goal can be achieved. Even though the agile way of working does not include defined team roles, all development teams in the CBS project have a Tech Lead and a Test Lead. The Tech Lead is responsible for the software development activities of the team and the Test Lead organizes and oversees the testing of the developed software. Additionally, the CBS program includes business analysts, who translate business requirements into technical requirements towards IT. The team members of the different teams in the CBS program are distributed among different offices of Beneal Bank but are also permitted to work remotely. In most of the teams, there are some team members that are co-located with one or more members of their team; therefore, the teams of the CBS project can be labeled as semi-virtual teams.

The teams employ information and communications technologies (ICTs) to connect with the own and other teams. The ICTs mostly used by the employees of the CBS program are Skype for Business and Confluence. Skype for Business is used for video-calls and -conferences. It enables the users to share their desktops with other participants to view, for example, slides. Confluence is a collaboration software in which users can create pages to document their work. The pages can be commented on and edited by all members of one team and can be accessed by all employees of the CBS program.

For sprints to be completed, the teams often depend on each other’s work. Due to the enormous size of the project, Beneal Bank employs agile methods as described above on a large scale. Our study site can thus be described as a multi-team system in which “all teams within the system, while pursuing different proximal goals, share at least one common distal goal; and in doing so exhibit input, process and outcome interdependence with at least one other team in the system” (Mathieu et al., 2001, pp.289–313). Thus, our study site is characterized by the need of close collaboration between members of different, dispersed teams and therefore qualifies as having great potential to answer our research question.

3.2 Research Philosophy

There are several philosophies that researchers can adopt in accordance with the research paradigm. Guba (1990) characterizes research paradigms as including ontology, epistemology, and methodology. While the first two create a holistic perspective of how knowledge is viewed and how we see ourselves in relation to this knowledge, the latter determines how we un- and discover it.

Ontology is concerned about the nature of reality. It determines what people believe to be a fact, in other words, what they believe to be reality (Blaikie, 2010). Epistemology asks the question of what and how people are able to know about reality in relation to their ontological stance (Blaikie, 2010). Positivism and interpretivism are two paradigms within ontology. Positivism asserts that social phenomena and their meanings exist independently of social actors (Bryman, 2012). On the contrary, interpretivism argues that social actors continually accomplish social phenomena and their meanings (Bryman, 2012). According to Saunders, Lewis, and Thornhill (2012), interpretivism regards how humans interpret their social roles and how they make sense of the world around them. In contrast to the positivist philosophy, in which the resources are embraced to develop knowledge, interpretivism is concerned about the feelings that the researcher is adopting (Saunders, Lewis & Thornhill, 2012). From an epistemological standpoint, this perspective implies that phenomena should be interpreted in the given context as each situation is different from another as it is found in different contexts.

Symbolic interactionism is a variant of interpretive scholarship which we applied for our research. According to this approach, all social phenomena are symbolic since they hold meanings for individuals and evoke different emotions and responses to them (Prasad, 2005). The social phenomenon in our research is the collaboration between virtual teams in a large, multi-team project. We intended to study the meanings that this social situation has for the team members. Conducting research within the symbolic interactionism tradition enabled us to identify multiple meanings in the social situation of collaboration, to explore the nature of roles and relationships of this particular context and to investigate how different realities are produced (Prasad, 2005). Central to the symbolic interactionism approach is, according to Prasad (2005), that the researchers enter the specific context of the individuals studied to understand the phenomena from their perspective. This approach allowed us to understand how inter-team collaboration works within the specific context, how it is experienced by team members, and how their experiences influence collaboration with other teams. A main disadvantage can be found in its subjectivity, as this philosophy leaves room for bias in interpretation (Saunders, Lewis & Thornhill, 2012). However, several actions were taken to limit the influence of bias (see Limitations and Creation of Quality, section 3.4). In general, our ontological and epistemological stances impacted the choice of research methods, including our research approach and research strategy.

Our research approach can be described as abductive. According to Bryman (2012), the research process in abductive approaches starts with surprising facts or ‘puzzles’, which the research process is devoted to explain. Although literature may provide different alternatives for an explanation, encountered empirical phenomena cannot be fully explained by existing literature. Adopting an abductive approach, our aim was, thus, not to verify an existing hypothesis, which is the research purpose when using a deductive approach (Saunders, Lewis & Thornhill, 2012), but to be open-minded about aspects that have been revealed throughout our study (Alvesson & Sköldberg, 2018). However, our research approach cannot be denoted as inductive because we acquired relevant theoretical knowledge of our research area before starting the data collection (Saunders, Lewis & Thornhill, 2012). According to our abductive approach, we considered our research question as a ‘moving target’, as suggested by Styhre (2013).

There are several strategies that can be used for different research purposes. Case studies are a common strategy in qualitative research as they comprise the empirical investigation of a contemporary phenomenon within its actual context by employing multiple sources data collection (Robson, 2002). In line with this definition, our research involved the empirical investigation of inter-team collaboration at Beneal Bank. We considered conducting a case study as the appropriate design for this research as it enabled us to perform an in-depth analysis of our specific research context and to capture ‘real-life’ occurrences. This also allowed us to develop and increase our knowledge about the topic of inter-team collaboration (Flyvbjerg, 2006). In addition, the case study approach facilitated triangulation, which means applying two or more data collection techniques within one study (Saunders, Lewis & Thornhill, 2012). Using dissimilar sources for our data collection (see Data Collection, section 3.3) contributed to the validity of our findings. A disadvantage of the case study design may be that the detailed investigation of a particular context limits generalization to a certain extent (see Limitations and Creation of Quality, section 3.4).

Adopting the interpretivist philosophy mainly includes qualitative data collection with small samples but in-depth data collection (Saunders, Lewis & Thornhill, 2012). For our research, we “entered the social world” of three different teams by conducting interviews and observations. Therefore, our research was qualitative (Styhre, 2013). Conducting qualitative research allows researchers “to make sense of the subjective and socially constructed meanings expressed by those who take part in the research about the phenomenon being studied” (Saunders, Lewis & Thornhill, 2012, p.546). Qualitative data contains information about behaviors, needs, desires, personal characteristics, and other information that quantitative data cannot display (Saunders, Lewis & Thornhill, 2012). Thus, in comparison to quantitative methods, interviews and observations allowed us to better understand how collaboration is experienced and to get deeper insights into how inter-team collaboration takes place.

Moreover, this study followed an exploratory research approach. According to Saunders, Lewis, and Thornhill (2012, p.171), it is used to “discover what is happening and gain insights about a topic of interest”. It is a flexible research approach implying that the research direction can be changed if new findings appear. Ways of doing exploratory studies include the reviewing of literature, interviewing experts, and conducting in-depth, individual interviews (Saunders, Lewis & Thornhill, 2012).

3.3 Data Collection

Individual in-depth interviews served as our main data collection technique, which was supplemented by observations. This section explains the reasons and usage of both techniques, including the sampling.

3.3.1 Semi-Structured Interviews

Interviews allow researchers to explore the views, experiences, beliefs, and motivations of individuals (Saunders, Lewis & Thornhill, 2012). We conducted semi-structured interviews, in which a set of broad questions served as a flexible guide (Saunders, Lewis & Thornhill, 2012). This enabled us to pursue the interview in accordance with the interviewees’ answers. This means that, based on the interviewee´s answers and especially the unexpected ones, we were able to investigate new areas that potentially unfold as important and worth analyzing. Kvale (1996, p.7) argues that the advantage of interviews is that researchers can capture “the multitude of subjects’ views of a theme” and to therefore illustrate a “controversial human world”. This strength of interviews also benefited our study as we investigated the perceptions of employees of virtual inter-team collaboration. Both researchers functioned as interviewers, one actively conducted the interview and the other took an observing role. As the social world of our research subjects was mainly virtual, we collected most of the data via Skype. The design and direction of the interviews was partly based on the interviewee’s answers and partly based on findings of the literature review. Please refer to appendix A to review our interview protocol, including the interview design and questions.

Interviewees were selected using non-probability sampling “in which the … probability of each case being selected is not known (Saunders, Lewis & Thornhill, 2012, p.676). Common within the research strategy of case studies, as described by Saunders, Lewis, and Thornhill, (2012), purposive sampling was employed to select appropriate interviewees. The selection was done in cooperation with Beneal Bank and was based on the employee’s ability to answer the research questions and on their involvement with other teams. In total, we interviewed 13 people from five different teams (see Table 1).

[...]