BACKGROUND: There is a lack of effective intervention strategies that promote physical activity (PA) in school children. Furthermore, there is a gap between PA intervention research and the delivery of programmes in practice. Evaluation studies seldomly lead to adaptations in interventions which are subsequently evaluated on a wider scale implementation. The stepwise development and study of JUMP-in aims to add knowledge to better understand how, when and for whom intervention effects (or lack of effects) occur. METHODS: This paper describes the stepwise development of JUMP-in, a Dutch school based multi- level intervention programme, aimed at the promotion of PA behaviour in 6 to 12-year-old children. JUMP-in incorporates education, sports, care and policy components. JUMP-in consists of six programme components: 1. Pupil Follow Up Monitoring System; 2. School sports clubs; 3. In-class exercises with "The Class Moves!"; 4. Personal workbook "This is the way you move!"; 5. Parental Information services; 6. Extra lessons physical education, Motor Remedial Teaching and extra care. The process- and effect outcomes of a pilot study were translated into an improved programme and intervention organisation, using the RE-AIM framework (Reach, Efficacy, Adoption, Implementation and Maintenance). This paper presents the process and results of the application of this framework, which resulted in a wide-scale implementation of JUMP-in. RESULTS: The application of the RE-AIM framework resulted in challenges and remedies for an improved JUMP-in intervention. The remedies required changes at three different levels: 1. the content of the programme components; 2. the organisation and programme management; and 3. the evaluation design. CONCLUSIONS: Considering factors that determine the impact of PA interventions in 'real life' is of great importance. The RE-AIM framework appeared to be a useful guide in which process- and effect outcomes could be translated into an improved programme content and organisation.