PurposeThis paper aims to explore the determinants of maternal healthcare services (MHS) utilisation in Ghana.Design/methodology/approachIn this paper, logit and negative binomial regressions were used to model data from a recent nationally representative survey (the 2017 Ghana Maternal Health Survey). The analysis is observational, without causal implications. The authors measure MHS utilisation by four indicators: antenatal care (ANC) use, the number of ANC clinic visits, choice of health facility delivery and postnatal care (PNC) use.FindingsAge, parity, education, marital status, wealth, residence and health facilities concentration proved to be significant predictors of MHS use in Ghana. Specifically, older, married and educated women; women of lower parity; those living in urban areas and women from wealthier households were more likely to use MHS. The authors also find that health facilities and personnel predicted higher MHS use. Lastly, women with frequent stillbirths had higher MHS use.Practical implicationsThe results for health facilities and personnel suggest that increasing the supply of health centres and workers may enhance MHS use. The authors conclude that women of lower socio-economic status have worse MHS use, meaning empowering such women might increase such women's MHS use.Originality/valueExisting data suggest falling MHS use in Ghana. Yet, the extant works are based on relatively old data or lack external validity (besides using limited MHS indicators). This paper provides recent and generalisable evidence for public health policies. Additionally, this paper tests the statistical significance of some supply-side factors that have yet to be emphasised in the literature.Peer reviewThe peer review history for this article is available at: https://publons.com/publon/10.1108/IJSE-03-2022-0199.