Introduction: Governments in different countries have committed to better use of evidence from research in policy. Although many programmes are directed at assisting agencies to better use research, there have been few tests of the effectiveness of such programmes. This paper describes the protocol for SPIRIT (Supporting Policy In health with Research: an Intervention Trial), a trial designed to test the effectiveness of a multifaceted programme to build organisational capacity for the use of research evidence in policy and programme development. The primary aim is to determine whether SPIRIT results in an increase in the extent to which research and research expertise is sought, appraised, generated and used in the development of specific policy products produced by health policy agencies. Methods and analysis: A stepped wedge cluster randomised trial involving six health policy agencies located in Sydney, Australia. Policy agencies are the unit of randomisation and intervention. Agencies were randomly allocated to one of three start dates (steps) to receive the 1-year intervention programme, underpinned by an action framework. The SPIRIT intervention is tailored to suit the interests and needs of each agency and includes audit, feedback and goal setting; a leadership programme; staff training; the opportunity to test systems to assist in the use of research in policies; and exchange with researchers. Outcome measures will be collected at each agency every 6 months for 30 months (starting at the beginning of step 1). Ethics and dissemination: Ethics approval was granted by the University of Western Sydney Human Research and Ethics Committee HREC Approval H8855. The findings of this study will be disseminated broadly through peer-reviewed publications and presentations at conferences and used to inform future strategies.