"Portfolios of Control in Outsourced Software Development Projects"
Fon Sundaravej Portfolios of Control in Outsourced Software Development Projects By Vivek Choudhury and Rajiv Sabherwal This article examines the evolution of portfolio of controls over the duration of outsourced information systems development (ISD) projects. The study posits mechanisms constituting the portfolios of control in outsourced ISD projects, evolution of controls change during the course of an outsourced ISD projects, and factors influencing the initial composition and subsequent evolution of the portfolio of mechanisms used to control an outsourced ISD project. Five case studies, measured by qualitative method, demonstrate that the portfolios of control in outsourced projects are dominated by outcome controls. Behavior controls are less used but often added later in the project. Clan control is difficult to achieve in outsourced ISD projects because of the distance between vendor and client’s locations. Self-control is internally implemented by the vendor or controllee organization. Additionally, interactions between clients and vendors during the course of a project affect their perceptions on choice of control modes and mechanisms. Project size and controller’s project-related knowledge to facilitate behavior control are found insignificant factors on the choice of controls in outsourced ISD projects. Chaudhury and Sabherwal (2003) retrieve a concept in portfolios of control modes from Kirsch (1997). Kirsch (1997) examines the function of control as an aspect of management processes. Control ensures that individuals in organizations behave in a consistent manner to meet organizational objectives. Control is categorized into formal (behavioral and outcome) and informal (clan and self) modes. The study explains how control modes are implemented during information system development projects and why information systems and users implement particular combinations of control modes. Four case studies of systems development projects reveal that users and information systems managers play an important role in controlling systems development projects. Users and managers implement a portfolio of control modes that include both formal and informal modes. In addition, the same mechanism is often used to implement various modes of control. Throughout the process of control modes construction, the choice of particular control depends on task characteristics, role expectations, and project-related knowledge and skills. Two differences between the studies of Chaudhury and Sabherwal (2003) and prior study by Kirsch (1997) are that (1) in the case of outsourced projects, the controller and controllee are members of different organizations, and (2) the controller and controllee may not be single individuals but teams of individuals representing their respective organizations. Additionally, the findings from this study demonstrate a different assumption from Kirsch (1997) in terms of the knowledge and skills of the controller. Kirsch (1997) argues that the lack of specific project-related knowledge prevents the use of behavior control. In contrast, Chaudhury and Sabherwal (2003) find that the controller’s project-related knowledge and skills becomes insignificant because controllers may believe that their knowledge is not superior to the controllee’s. This implies that controls in different contexts may be used in distinctive ways and do not predict a success of the result of using particular controls. IS 7890: IS Research Seminar Spring 2006