BACKGROUND: South Africa did not meet its Millennium Development Goals with regard to the reduction in maternal and under-5 mortality. Furthermore, many birthing women do not receive intrapartum care with empathy and endure disrespect-ful and abusive care. OBJECTIVES: To implement a multicomponent, context-specific intervention package to change the complex interplay between preventable maternal and perinatal mortality and morbidity and poor clinical governance and supervision in midwife-led labour units. METHODS: A mixed-methods intervention study was conducted in Tshwane District, South Africa, in 10 midwife-led obstetric units (MOUs), from which a purposive sample consisting of five units was selected for the intervention. The intervention took place in three phases: (i) baseline measurement; (ii) implementation of the so-called 'CLEVER' intervention package in the five intervention units, based on the results of the first phase; and (iii) a review of health systems improvements and perinatal outcomes. The intervention had three pillars: (i) feedback of the baseline measurement to the intervention units to raise awareness and solicit participation; (ii) health systems strengthening; and (iii) intensive weekly engagement for 3 months, with further monthly support afterwards. Observation of barriers during baseline activities contributed to the health systems strengthening and improvement strategies during implementation. RESULTS: Perinatal outcome indicators for the year before the intervention were compared with data for the year in which the intervention took place and the year after the intervention. Significant declines were observed in in-facility fresh stillbirths, meconium aspiration and birth asphyxia in the intervention MOUs from 2015 to 2017. The control group showed some decline during the period owing to support from district clinical specialist team members. CONCLUSIONS: CLEVER as a context-specific, multicomponent, clinically focused intervention package may have contributed to improved perinatal morbidity and mortality rates in MOUs.