Background: The present study aims to test out contextually tailored interventions to increase evidence-informed health-enhancing physical activity policy-making in two Danish municipalities. Methods: The study was performed as experiments in natural settings. Based on results from a pre-intervention study defining the needs and contexts of the two settings, the interventions were developed based on logical models. The interventions aimed at increasing the use of knowledge in policy-making, primarily via strengthening intersectoral collaboration. The interventions were evaluated via pre-, post- and 12-month follow-up questionnaires and qualitative interviews were carried out prior to the intervention start. Results: The use of knowledge changed in several ways. In one municipality, the use of stakeholder and target group knowledge increased whereas, in the other municipality, the use of research knowledge increased. In both municipalities, the ability to translate knowledge to local context, the political request and the organisational procedures for use of knowledge increased during the interventions. There was some variation between the two settings, which shows the importance of tailoring to context. Most of the changes were diminished at the 12-month follow-up. Conclusion: Contextually tailored interventions have the potential to increase evidence-informed policy-making on health-enhancing physical activity. However, this finding needs to be tested in larger samples and its sustainability must be strengthened.