PurposeThis article illustrates the use of mixed methods in the development and evaluation of the Positive Online Weight Reduction (POWeR) programme, an e‐health intervention designed to support sustainable weight loss. The studies outlined also explore how human support might enhance intervention usage and weight loss.MethodsMixed methods were used to develop and evaluatePOWeR. In the development phase, we drew on both quantitative and qualitative findings to plan and gain feedback on the intervention. Next, a feasibility trial, with nested qualitative study, explored what level of human support might lead to the most sustainable weight loss. Finally, a large community‐based trial ofPOWeR, with nested qualitative study, explored whether the addition of brief telephone coaching enhances usage.ResultsFindings suggest thatPOWeRis acceptable and potentially effective. Providing human support enhanced usage in our trials, but was not unproblematic. Interestingly, there were some indications that more basic (brief) human support may produce more sustainable weight loss outcomes than more regular support. Qualitative interviews suggested that more regular support might foster reliance, meaning patients cannot sustain their weight losses when support ends. Qualitative findings in the community trial also suggested explanations for why many people may not take up the opportunity for human support.ConclusionsIntegrating findings from both our qualitative and quantitative studies provided far richer insights than would have been gained using only a single method of inquiry. Further research should investigate the optimum delivery of human support needed to maximize sustainable weight loss in online interventions.Statement of contributionWhat is already known on this subject?There is evidence that human support may increase the effectiveness of e‐health interventions.It is unclear what level of human support might be optimal or how human support improves effectiveness.Triangulation of quantitative and qualitative methods can be used to inform the design and implementation of interventionsWhat does this study add?This paper demonstrates the value of a mixed methods approach when developing and evaluating an intervention.Qualitative methods provided complementary insights into the optimal level of human support.Brief human support is valued by some and may enhance usage and outcomes of an e‐health intervention for wei