BACKGROUND: Burkina Faso, like many low and middle income countries, has been taking a range of actions to address its poor maternal and neonatal health indicators. In 2006 the government introduced an innovative national subsidy scheme for deliveries and emergency obstetric care in public facilities. This article reports on a complex evaluation of this policy, carried out 5 years after its introduction, which examined its effects on utilisation, quality of care, equity and the health system as a whole, as well as its cost and sustainability. METHODS: The evaluation was carried out in six purposively selected districts, as well as at national level, using a case study approach. Data sources included: national and district routine and survey data, household interviews with women who had recently given birth, data extraction from hospital and medical records, and key informant and health worker interviews. RESULTS: The underlying secular trend of a 1% annual increase in the facility-based delivery rate (1988-2010) was augmented by an additional 4% annual increase from 2007 onwards (after the policy was introduced), especially in rural areas and amongst women from poor households. The absence of baseline quality of care data made it difficult to assess the impact of the policy on quality of care, but hospitals with the best level of implementation of the subsidy offered higher quality of care (as measured by health care near-misses), so there is no evidence of a negative impact on quality (as is often feared). Similarly, there is little evidence of unintended negative effects on untargeted services. Household payments for facility-based deliveries have reduced significantly, compared with payments before the policy, and the policy as a whole is affordable, costing about 2% of total public health expenditure. Concerns include that the amounts paid by households are higher than the rates set by the policy, and also that 7% of households still say that they cannot afford to pay. Wealthier women have higher utilisation of services, as before, and the policy of fully exempting indigents is not being put into practice. CONCLUSIONS: These findings highlight the importance of maintaining the subsidy policy, given the evidence of positive outcomes, but they also point out areas where attention is needed to ensure the poor and most vulnerable population benefit fully from the policy.