["Health Expectations, Volume 29, Issue 2, April 2026. ", "\nABSTRACT\n\nIntroduction\nCo‐design is increasingly used to develop research interventions and community services, highlighting the need for a deeper understanding of how this method drives change. This study evaluates the contextual factors and mechanisms shaping a co‐design process intended to foster local collaboration between social services, primary care and the third sector.\n\n\nMethods\nA realist process evaluation was conducted in two stages: (1) formulation of an initial programme theory (IPT) consisting of Intervention‐Context‐Actor‐Mechanism‐Outcome (ICAMO) configurations, and (2) testing and refinement of the IPT. The IPT drew on comprehensive documentation from the planning and implementation phases of the co‐design process. To test and refine the IPT, six group interviews were conducted using realist interviewing and analysed deductively using IPT‐derived categories.\n\n\nResults\nThree interconnected ICAMO configurations were identified. The first concerns stakeholders' assessment of the potential value of participating in the co‐design process and their decision to engage. The second highlights how emerging trust enabled respectful and constructive dialogue. The third focuses on the facilitation of ideation and prototyping of goals and terms for collaboration. Key mechanisms include sensing an opportunity for change, building trust, and experiencing group agency.\n\n\nConclusion\nSuccessful co‐design requires facilitation that not only considers contextual factors but is also perceived as impartial, as such neutrality is vital for establishing trust. Constructive participation depends on representatives who combine relevant practical knowledge, a sense of opportunity for change, and the mandate or organisational conditions to act on outcomes. Continuity in participation, ensuring that the same individuals remain involved throughout the process, is also crucial for maintaining competence and for building the relationships and trust needed to develop shared goals and collaborative structures.\n\n\nPatient or Public Contribution\nPatients and the public were not directly involved in the design, conduct or analysis of this study. However, this evaluation examined a co‐design process involving third‐sector organisations, including older adults. A public advisory board contributed to the broader participatory action research project.\n"]