This dissertation contributes to the existing body of knowledge on how we design computer systems, particularly multiuser software for knowledge sharing and creation in globally diffused companies. This is achieved by conducting a work place study of a global industrial engineering conglomerate which has the strategy of working with knowledge in the form of "best practices" meant to boost performance. The thesis explores the situation that workers are in, since they are meant to share and develop "best practices" knowledge in a portal based Knowledge Management System (KMS). The study indentifies a set of problems that prevents knowledge sharing from taking place to the degree to which management was specifically aiming. It was explored whether these problems could, to some degree, be mitigated by employing persuasive design, which is a new stance towards design where the aim is to directly seek to change the user's behavior, i.e., persuading more knowledge sharing. The main contribution is an indication of an anomaly with regards to the strategic approach towards knowledge management, where knowledge sharing is seen as an effort by which companies can gain a competitive advantage by working with knowledge in a structured fashion. The issue is that the descriptions found in literature on strategic knowledge management do not address the many issues uncovered when conducting prolonged fieldwork among workers who engage in the activities that the literature seemingly takes for granted. Thus, many practical problems were uncovered that would need some level of mitigation before a company could hope to gain a strategic advantage from working with knowledge. This challenges the "stock" approach towards knowledge management, which seems to address only the management level of the organization. A contribution is also made in exploring the state-of-the-art of the emerging field of persuasive design.
- brugerdreven innovation