Abstract: This discussion paper aims to offer an overview and working definition of translational research, appropriate to health. Methods: Using scholarly and applied literature, the paper first identifies key challenges in achieving evidence-based policy and practice. It highlights international policy interest in new approaches to evidence translation and the barriers to achieving sound evidence translation. The paper offers an explicit definition of translational research and explains why it is important to have such a definition. It then elaborates on this definition by identifying and exploring seven distinctive research practices that could be associated with translational research. Findings and conclusions: Translational research is research with a sense of place. Its defining feature is excellence in evidence for a specific context or sphere of action, whether that is health policy for the World Health Organisation or service design for a local non-government organisation. If research is to be translated at all, it needs to be meaningful to many specific contexts, including small and regional contexts. The best promise that translational research offers is of exciting new techniques to achieve rigour and systemacy for such localised ‘real world’ policy, service and practice contexts.