BACKGROUND: Local authorities need to find new ways of collecting and using data on social care users' experiences to improve service design and quality. Here we draw on and adapt an approach used in the healthcare improvement field, accelerated experience-based co-design, to see if it can be translated to social care. We use loneliness support as our exemplar. OBJECTIVES: To understand how loneliness is understood and experienced by members of the public and characterised by social care and voluntary sector staff; to identify service improvements around loneliness support; to explore whether accelerated experience-based co-design is effective in social care; and to produce new resources for publication on Socialcaretalk.org. DESIGN AND METHODS: Discovery phase: in-depth interviews with a diverse sample of people in terms of demographic characteristics with experience of loneliness, and 20 social care and voluntary staff who provided loneliness support. Production of a catalyst film from the public interview data set. Co-design phase: exploring whether the accelerated experience-based co-design approach is effective in one local authority area via a series of three workshops to agree shared priorities for improving loneliness support (one workshop for staff, another for people with experience of local loneliness support, and a third, joint workshop), followed by 7-monthly meetings by two co-design groups to work on priority improvements. A process evaluation of the co-design phase was conducted using interviews, ethnographic observation, questionnaires and other written material. RESULTS: Accelerated experience-based co-design demonstrated strong potential for use in social care. Diverse experiences of participants and fuzzy boundaries around social care compared to health care widened the scope of what could be considered a service improvement priority. Co-design groups focused on supporting people to return to pre-pandemic activities and developing a vulnerable passenger 'gold standard' award for taxi drivers. This work generated short-term 'wins' and longer-term legacies. Participants felt empowered by the process and prospect of change, and local lead organisations committed to take the work forward. CONCLUSIONS: Using an exemplar, loneliness support, that does not correspond to a single pathway allowed us to comprehensively explore the use of accelerated experience-based co-design, and we found it can be adapted for use in social care. We produced recommendations for the future use of the approach in social care which include identifying people or organisations who could have responsibility for implementing improvements, and allowing time for coalition-building, developing trusted relationships and understanding different perspectives. LIMITATIONS: COVID-19 temporarily affected the capacity of the local authority Project Lead to set up the intervention. Pandemic work pressures led to smaller numbers of participating staff and had a knock-on effect on recruitment. Staff turnover within Doncaster Council created further challenges. FUTURE WORK: Exploring the approach using a single pathway, such as assessing eligibility for care and support, could add additional insights into its transferability to social care. TRIAL REGISTRATION: This trial is registered as Current Controlled Trials ISRCTN98646409. FUNDING: This award was funded by the National Institute for Health and Care Research (NIHR) Health and Social Care Delivery Research programme (NIHR award ref: NIHR128616) and is published in full in Health and Social Care Delivery Research; Vol. 12, No. 27. See the NIHR Funding and Awards website for further award information.