Holistic approach to software process improvement in small software enterprises: a Delphi study
For small software enterprises with 50 or fewer employees, using international standards to manage their software development projects has been a major challenge as these standards are more suited to large organisations. This has presented a challenge to small software enterprises (SSEs) since they operate in highly challenging and constantly changing environments and tend to perform their productive, administrative and management processes in a chaotic manner. There is agreement in the literature that SSEs need superior internal systems and structures to help them achieve their project objectives. Despite this, there is limited research into developing a process model that incorporates best practice and is suitable for SSEs. The purpose of this research study is to develop a capability model with a set of essential practices for SSEs. The analysis of the literature of Software Process Improvement (SPI) and Dynamic Capabilities (DC) suggested the need for a wide array of capabilities that account for different aspects of the organisation and the external environment. This research, therefore, is based on the premise that organisations can take a holistic multifaceted approach to SPI programmes. Sound organisational processes based on best practice are needed to improve their internal structures. Embedding dynamic capabilities within these processes will additionally enable the enterprise to address the business challenges in competitive markets. An initial version of the proposed model was developed. This provided the necessary practices to help SSEs meet their perceived project outcomes. The model builds upon multiple software engineering standards, informed by the Dynamic Capability approach, and designed to be feasible for application in SSEs. Using the Delphi technique as the methodology, this study sought the help of a group of experts to assess and enhance the proposed model by removing or adding practices to make it feasible for SSEs. A panel of nine members experienced in software process improvement and software engineering from different geographic regions (South Asia, North America and Australasia) responded to two Delphi rounds; their views about the model were probed through qualitative analysis of members’ feedback on the practices, supported by quantitative analysis using the interquartile range (IQR) and median measures. The outcome of this research is a capability model that includes a set of basic and advanced practices, which represents the model that the participants have collectively approved. This model hopes to be an effective software development framework for SSEs. Further work is still required to verify the model using a case study research approach, in order to compare the Delphi findings against observed data.