Abstract
Although investment in IT is growing rapidly, a number of studies have shown that the full benefits of this technology are not realised. One explanation is that these benefits exist, but have not been correctly measured. The other explanation is that they have not been achieved because of lack of management attention. Experience and common sense supports the latter assumption. Hence, the aim of this research is to develop and to trial a method which assists in the realisation of the benefits of a particularly problematic technology, executive information systems (EIS).The proposed method introduces the concept of generic benefits models into previous research on IT benefits management. This construct improves existing methods of benefits management in three respects. First, it employs reuse of benefits models in order to speed their development. Secondly, it allows these methods to be employed at any stage of a project, not just at the outset, in order to extract benefits. Thirdly, it provides a conceptual object which serves as the focus of organisational learning. The major focus of the research is on the development of generic benefits models of EIS. This requires an understanding of the role which information plays in executive work and how EIS technology may affect the manner in which executives use information. Unfortunately, the literature on executive work does not make this clear. Moreover, it seems that there is no agreed definition of EIS. As a result, it is necessary to create different generic models for different theories of executive work and different forms of EIS. The methodological approach adopted is pragmatism, in particular the experimentalism proposed by John Dewey and implemented by Donald Schon. The justification for this is that the primary objective of the study is to demonstrate the utility of the proposed method rather than its ability to explain, predict, or provide insight. The study includes six experiments conducted within the office of the Malaysian Prime Minister. Different types of EIS, with different degrees of benefits already realised, and at different stages of development are the subjects of these experiments. The results largely affirm the proposed method, but they do suggest some refinement of the original benefits models. They also indicate simplification of the models is possible. In addition the executives who participated in these experiments favour the proposed method. However, the IS practitioners in the organisation are less enthusiastic. The study proposes a solution to this problem which includes both changes to organisational structure and to the education of IS practitioners. The research also produced a number of supplementary findings. It reaffirmed the findings of Mintzberg's study of executive work. It demonstrated once again that executives rarely use EIS directly and that they mostly depended on subordinates to access information. When executives do employ EIS themselves, they are more likely to use it to enhance their learning than to support decision-making.