Background: Digital Social Prescribing (DSP) has expanded rapidly, particularly following the Covid-19 pandemic, as digital technologies increasingly support mental health and wellbeing interventions. Despite this growth, DSP remains under-researched, with limited understanding of its opportunities, challenges, and impacts. This project aimed to map current DSP practices, explore stakeholder perspective, and test a novel digital evaluation methodology to assess DSP outcomes. Methods: A mixed-methods design was used in two phases. First, a UK-wide survey and semi-structured interviews with stakeholders to map current DSP practices and identify challenges and enablers in digital delivery. Second, a co-designed feasibility study tested a digital ecological momentary assessment (EMA) approach, combining continuous physiological data from smartwatches with twice-daily self-reported wellbeing measures via a bespoke mobile application. Thirteen adults engaged in DSP activities (e.g. online yoga) participated, supported by two participatory workshops co-designing the protocol and providing qualitative feedback. Results: Over half of survey respondents (55%) reported delivering DSP, most commonly arts, crafts, and exercise-based sessions. Reported benefits included improved accessibility and cost-effectiveness, while challenges involved building rapport remotely, the digital divide, and potential reinforcement of isolation. In the feasibility study, physiological data suggested reductions in stress and improvements in sleep during DSP participation, although correlations with self-reported wellbeing were inconclusive due to the small sample size. Qualitative feedback indicated positive experiences with both the DSP activities and the digital EMA process, despite technical and data-integration challenges. Conclusion: DSP shows significant potential to enhance accessibility to mental health support. Combining real-time self-reporting with physiological monitoring via smart devices offers a promising, scalable method for evaluating DSP interventions. Future research should use larger samples and automated data systems to strengthen evidence, ensure equity, and optimise implementation.