Viewing organizations as political coalitions has a long history in management studies and has generated a great deal of research on the roles of organizational politics and power in decision making. Information technology (IT) researchers generally have followed the general trend by providing structural-functional accounts of politics, power, and control strategies. Although such studies underscore the importance, nature, and complexity of politics in IT implementation, they are constrained by the lack of both a general theory of human behavior as well as a model of power that bridges the individual actor and the wider interpersonal/organizational setting. In this paper we present a case study of power and politics in a large financial transactions company, but seen through the lenses of the action science perspective of Argyris and his associates and the model of power developed by A.A. Berle. More specifically, we examine the power vacuum in the IT effort that resulted from the senior executive abrogating his power and failing to delegate it to his subordinates. The resulting political havoc, in turn, caused competition and confusion, wasted effort, and a significant lack of productivity with regard to the implementation of a multimillion-dollar IT effort. We conclude with several hypotheses about how principles of action science and interpersonal power must be used by executives to manage the politics of IT design and implementation.
Information technology (IT) implementation is likely to be a complex and difficult process involving both the technical and social systems of an organization. Much of the theory and practice of organizational IT implementation assumes that organizational decisions--whether technical or social--are consequences of individuals and organizational units objectively collecting, evaluating, and applying information in a rational manner to make choices on behalf of the organization. However, recent data from case studies on IT implementation suggest that rationality may be the exception, rather than the rule. One important limitation, then, of traditional research and systems analysis methods is that they fail to account for the human factors that impact IT implementation. As a result, there is no systematic way to incorporate what we might learn from individual action and practice into the model. A theoretical perspective and research methodology that successfully encompass and expand the normative approach in mainstream science is "action science." Action science is a theory of action that helps explain how, and why, individuals behave as they do; and how their actions impact their organizations. In this paper, our primary goal is to employ an action science perspective to understand how a large financial services firm was inhibited from implementing a major IT effort in a timely, effective, and error-free way because of the thought and action routines of those charged with designing and implementing the system. In particular, we look for disagreements and other, unproductive action routines among these individuals, and evidence of their inability to surface and discuss them. We also review two published studies of IT implementation to show that the level of analysis required by action science is implied, but not made explicit in them. We conclude with a discussion of what constitutes a descriptively adequate account of an IT implementation and the barriers to the production of such an account. We also suggest a competing framework, based on action science, for designing and managing organizational IT change.