Objectives: To understand feedback from participants in Paired PLIÉ (Preventing Loss of Independence through Exercise), a novel, integrative group movement program for people with dementia and their care partners, in order to refine the intervention and study procedures. Method: Data sources included daily logs from the first Paired PLIÉ RCT group, final reflections from the second Paired PLIÉ RCT group, and responses to requests for feedback and letters of support from Paired PLIÉ community class participants. All data are reports from care partners. The qualitative coding process was iterative and conducted with a multidisciplinary team. The coding team began with a previously established framework that was modified and expanded to reflect emerging themes. Regular team meetings were held to confirm validity and to reach consensus around the coding system as it was developed and applied. Reliability was checked by having a second team member apply the coding system to a subset of the data. Results: Key themes that emerged included care partner-reported improvements in physical functioning, cognitive functioning, social/emotional functioning, and relationship quality that were attributed to participation in Paired PLIÉ. Opportunities to improve the intervention and reduce study burden were identified. Care partners who transitioned to the community class after participating in the Paired PLIÉ study reported ongoing benefits. Conclusion: These qualitative results show that people with dementia and their care partners can participate in and benefit from community-based programs like Paired PLIÉ that include both partners, and focus on building skills to maintain function and quality of life.