Abstract
Software development companies have progressively been experiencing the need to migrate their traditional development processes to agile environments. Several frameworks have emerged that support this paradigm, and Scrum is one of the most widely adopted on a global scale. However, Scrum was originally designed to work in small, local team environments. The challenge lies in designing its adoption in large-scale environments with geographically distributed teams. This chapter seeks to explore this challenge by identifying communication and coordination issues that may arise when adopting distributed Scrum. Three case studies are considered to allow comparison of different approaches considering structural and contextual factors associated with each organization.
TopIntroduction
Agile software development methods have acquired great popularity over the past decades. Dissatisfaction with the traditional development model that required a lot of planning and documentation in the early stages of a project, extrapolation of software delivery times, increased costs above initial estimates, and little flexibility regarding changes in requirements along the project are typically pointed out as the reasons for the increase in the number of agile projects in the software industry (Pikkarainen et al., 2008).
At the same time, there is also an increase in distributed software development motivated by the organizational problems faced by companies and the scarcity of highly qualified human resources in software engineering. According to Chakravorty et al. (2014), this model of distributed software development contributes to increasing the development speed through geographically distributed teams. Furthermore, Lee & Xia (2010) report benefits in reducing costs associated with those projects.
Scrum is currently the most agile method used by software companies (Petrova, 2019). Scrum is a simple, lightweight Agile framework used for managing the development of complex products immersed in complex environments. Scrum is empirically based and uses an iterative and incremental approach to deliver value frequently and thereby reduce project risks (Sutherland, 2014). Several studies report its adoption in software development teams (Hajjdiab et al., 2012; Rigby et al., 2016). However, this method was originally designed to be applied in small teams that share the same physical space as recognized in Schwaber & Beedle (2002). Therefore, the adoption of this model in geographically distributed teams is also expected to bring new challenges. Several challenges are pointed out in the literature in areas such as communication between teams, project and process management, knowledge management, and cultural issues (Almeida et al., 2019; Dikert et al., 2016).
Communication and project coordination mechanisms in large-scale environments are more relevant in view of the larger number of developers and their geographical location, which can be quite diverse. In this sense, this study addresses the challenge of exploring the adoption of Scrum methodology in geographically distributed teams through the realization of three case studies within software companies. Firstly, it intends to explore the different levels of team distribution and the causes that led to the choice of the distributed model. Next, an attempt is made to systematize a set of characteristics and practices corresponding to this type of development. Finally, it seeks to identify ways to mitigate the communication and coordination issues that can be found in this distributed model of software development. This work essentially intends to offer practical contributions to software companies that are taking the first steps in the distribution of their development teams. Through this work, these companies will be better prepared to face the challenges posed by the adoption of Scrum in a distributed environment.
The chapter is organized as follows: Initially, a literature review is performed on the software development processes in distributed teams. Next, the methodology and associated methods adopted in this study are presented. It is also in this section that the profile of the software companies participating in this study is presented. After that, the results of the study are presented and discussed considering the relevance of this evidence to the scientific community. Finally, the conclusions of the study are stated. Also, in this section, the theoretical and practical implications, limitations, and some indications of future work are presented.
Key Terms in this Chapter
Sponsor: Person or organization that you represent and who has the “last word” (i.e., the main authority of the project). The sponsor provides the financial resources to pay for the project and its execution, as well as resolving problems, scope changes, and approving important deliverables.
Risk Management: Defining and implementing procedures to assess the risks faced by an organization, with the goal of taking appropriate preventive measures.
Knowledge Sharing: Degree to which the different groups of project members are well informed about each other’s areas of expertise and can disseminate knowledge among them.
Project Manager: Professional responsible for the planning, execution, and monitoring of a project.
Management 3.0: It is a collaborative and humanized management model, which elects the employee as the main asset of an organization in search of better business results.
Stakeholder: Individuals or groups that have an interest or some aspect of rights or ownership in the project, may contribute to the form of knowledge or support, or may impact or be impacted by the project.