Most randomised controlled trials focus on outcomes, not on the processes involved in implementing an intervention. Using an example from school based health promotion, this paper argues that including a process evaluation would improve the science of many randomised controlled trials