Little is known about the effectiveness of a psychosocial behaviour management programme on home‐dwelling people with dementia. We developed a Behaviour Analytics & Support Enhancement (BASE) programme for care managers and professional caregivers of home care services in Japan. We investigated the effects of BASE on challenging behaviour of home‐dwelling people with dementia. Methods A cluster‐randomized controlled trial was conducted with home care providers from 3 different districts in Tokyo. Each provider recruited persons with dementia aged 65 years or older to receive home care in the BASE programme in August 2016. An online monitoring and assessment system was introduced to the intervention group for repeated measures of challenging behaviour with a total score of the Neuropsychiatric Inventory. Care professionals in both the intervention and control groups evaluated challenging behaviour of persons with dementia at baseline (September 2016) and follow‐up (February 2017). Results A majority of persons with dementia had Alzheimer disease (59.3%). One‐hundred and forty‐one persons with dementia were included in the intervention group and 142 in the control group. Multilevel modelling revealed a significant reduction in challenging behaviour in the intervention group after 6 months (mean score, 18.3 to 11.2) compared with that of the control group (11.6 to 10.8; P < .05). Conclusion The implementation of the BASE programme resulted in a reduction of challenging behaviour of home‐dwelling people with dementia. Future research should examine the long‐term effects of behaviour management programmes on behaviour, nursing home placement, and hospital admission of home‐dwelling people with dementia.