Physical activity (PA) is crucial for preschool-aged children's health and development. However, limited evidence exists regarding the feasibility of implementing home-based interventions and how program components influence parent cognitions and practices and child PA. This study evaluated the feasibility and potential efficacy of a family-based PA intervention on objectively measured PA, fundamental movement skills (FMS), parental efficacy, support, goal setting and parent–child co-activity. Guided by social cognitive theory, an 8-week cluster randomised controlled trial was conducted in Hong Kong. The trial included parental workshops, FMS training, PA homework, social media activity sharing and exercise equipment provision. Data were collected at baseline (Time 1; April 2019) and at the end of the intervention period (Time 2; approximately 2 months later) from 108 parent–child pairs in five preschools. The intervention led to increased moderate-to-vigorous PA and FMS in children, along with improved parental self-efficacy, goal setting, supportiveness and co-participation. However, parental PA did not show significant changes. Parents expressed high satisfaction, supporting the need for tailoring interventions to address the unique needs and preferences of young children and their parents. Reinforcing the parental role and providing informative materials and training can promote healthy lifestyles in early childhood.