Background/aim: Children with autism spectrum disorder (ASD) frequently demonstrate impaired play skills and poor quality social interactions compared to typically developing peers. Complex interventions to improve play skills should be investigated with randomised control trials (RCT) where possible to support evidence-based practice for occupational therapists. Prior to a RCT, multiple feasibility studies are recommended to identify barriers to the trial. The aim of this study is to adapt a complex intervention to improve play skills in children with ASD and investigate the feasibility of conducting a RCT. Methods: Participants were 10 children with ASD paired with typically developing playmates chosen by their families. Pairs attended 10 intervention sessions involving video modelling, therapist- and peer-mediation and free play. Participant numbers needed for future investigation of effectiveness were calculated and time and cost considerations were reviewed. The Test of Playfulness (ToP) was the primary outcome measure; both parent- and child-report outcome measures were evaluated for appropriateness. Results: To determine effectiveness of the intervention 34 pairs in both control and intervention groups will be needed to conduct a RCT over an 18-month timeframe. ToP scores showed a positive, but not statistically significant trend from pre- to post-intervention. These improvements were maintained at follow-up in both clinic and home environments. The Piers-Harris 2 and the Parent Relationship Questionnaire were identified as appropriate secondary outcome measures; additional parent- and teacher-report outcome measures are recommended for the RCT. Conclusion: A RCT to investigate effectiveness of this intervention for improving play skills of children with ASD and their playmates are feasible.