tify system borders for” (Alänge et al., 1998: 8). Taken together, these attributes allow a higher level of subjective interpretation on the part of the potential user than is common with technological innovations, which, in turn, increases the importance of the social and political processes fol- lowed by the proponents of the innovation. Second, very few organizations have well- established and specialized expertise in the area of management innovation. A typical large organization might employ tens or hundreds of scientists with technological innovation skills but few, if any, with proven management innovation skills (the closest are organization development consultants, who seek systemic ways of improving the overall effectiveness and health of the organization). This lack of expertise both heightens the uncertainty of management innovation for people across the organization and increases the need for external support. Third, the introduction of something new to the state of the art creates ambiguity and uncertainty for the individuals in an organization. Ambiguity arises because of a lack of under- standing of the intended value of the innovation, and uncertainty arises because of a fear that the innovation will have negative consequences for the individual and/or the organization. If an organizational change is proposed that has already been successfully implemented else- where (eg, the installation of a new IT system), its proponents can allay the concerns of individuals by referring back to those prior successes, but if the change is new to the state of the art, then the task of reducing ambiguity and uncertainty is much harder. Of course, all types of innovation generate uncertainty and ambiguity, but their impact in the case of management innovation is likely to be more far-reaching because of the rest of the attributes identified above. Taken together, these attributes suggest that the management innovation process can potentially require fundamental changes in the routines or DNA of the organization 4 (Argyris & Schön, 1978) that make it very difficult to under- take in an effective manner, and significantly