Clay and Schaffer’s book ‘Room for Manoeuvre’ in 1984 described agricultural policy processes in developing countries as “... a chaos of purposes and accidents ... not at all a matter of the rational implementation of decisions through selected strategies." That may be putting it extremely, but there is growing recognition that policy processes are complex, multidimensional and unpredictable and there is an urgent need to find mechanisms to promote the use of research-based and other forms of evidence in development policy. Theoretical and case-study research and practical work carried out by ODI's RAPID (Research and Policy in Development) programme and the GDN Bridging Research and Policy project over the past three years has led to the development of a practical approach designed to help with this. The approach includes an analytical framework to help unpack the complex range of factors which can influence research uptake including the 'political context', the credibility of the 'evidence' and the 'links' between policy and research communities.