EDITOR—One way for universities to revitalise academic medicine might be to look outward.1 Research takes money, and competition for funding is fierce. The chances of success are improved by demonstrating an expertise in both subject area and appropriate methods. Collaboration can be one way to achieve this, but how do you know with whom to collaborate? It's not just about the interests or the skills, of course: if successful you have to work with these people.

A recent approach we used was a modified form of "speed dating" to improve our collaborative outlook. The Academic Unit of Primary Health Care and the School for Policy Studies are active in research at the University of Bristol, but in different faculties. We were aware of some overlap in subject areas and methodological expertise, yet to date we have had little experience of working together. We devised "speed networking" as an effective means for members from each unit to meet and identify areas for possible future collaboration.

We introduced the groups to each other through brief oral presentations that gave an overview of the departments. Members from primary health care were then stationed at points around the room and, at three minute intervals, colleagues from policy studies were invited to rotate between each station. After 24 minutes all 16 participants had had the opportunity to meet and rapidly outline their skills and interests. Thereafter followed coffee to allow interested "pairs" to follow up their introductions in more detail.

Other means of networking exist, but so far as we are aware this is the first application of a populist approach to research. Feedback from participants was positive, and we suggest that other departments looking for research relationships speed network as an ice breaker to future collaborations.

• Matthew J Ridd, Alison R G Shaw (BMJ (British Medical Journal)) 24th September 2005

•