The potential use of mathematical models in programme science will be reviewed. The adoption, planning, implementation, and evaluation of programmes in global health should be an iterative process where the collection and analysis of data plays a significant role in planning and evaluation. Mathematical models provide a framework for the integration of data from multiple sources, predicting the impact of programmes based on efficacy data for the range of interventions combined and providing counterfactuals to estimate effect sizes in evaluating impact. Mathematical models describing the impact of alternative interventions are central in health economic analyses. Models can usefully be combined with theories describing why programmes should have an impact in the design and evaluation of the programmes. Synergies in interventions can be considered at multiple levels: in the individual both in enhancing behaviour changes and combining to reduce risks; in populations changing the epidemiological context; in programme activities; and in creating environments where interventions can succeed. Models explain what we can expect from these synergies and help us identify how to integrate new technologies into programmes.