Resident consultant cover has long been proposed as a solution to a wide variety of current intrapartum issues: poor clinical outcomes; high intervention rates, particularly increasing rates of caesarean section; middle-grade rota gaps; and decreased training opportunities. As has been the case for many previous well-intentioned, apparently simple and plausible interventions introduced into maternity care, however, there may be little or no benefit when robustly tested. The recent systematic reviews from the National Perinatal Epidemiology Unit and Reid et al. conclude that there is no clear evidence of different intrapartum outcomes and safety of care with 24-hour resident consultant presence on the labour ward. Reid et al. found that increased hours per week of rostered consultant presence significantly reduced the likelihood of emergency caesarean sections and increased the likelihood of non-instrumental vaginal deliveries; however, an effect on mortality and morbidity was not identified. The authors accept that the quality of the existing data was low; however, the introduction of a resident labourist model of care in the USA has similarly failed to demonstrate any tangible benefits in intrapartum process measures or clinical outcomes (Srinivas et al. Am J Obstet Gynecol 2016;215:770.e1–9). Although intrapartum outcomes are important direct markers of patient safety, it is crucial to consider other drivers behind the push for resident consultant cover, particularly the increasing number of middle-grade rota gaps in UK maternity units: heads of school reported 30% rota gaps in a 2014 survey. This is likely to be exacerbated by the recent call for a reduction in training numbers by Health Education England (HEE). In response to these rota gaps, the Royal College of Obstetricians and Gynaecologists (RCOG) have recommended resident consultant working (http://www.rcog.org.uk/globalassets/documents/guidelines/working-party-reports/ogworkforce.pdf), and a recent RCOG commentary concluded that some resident consultant care in most units was inevitable (http://www.rcog.org.uk/globalassets/documents/news/membership-news/og-magazine/December-2016/feature.pdf). RCOG also recognises that there may be difficulties with the sustainable implementation of resident consultant working, however. A Kings Fund review of maternity staffing concluded that the skill mix, experience, and deployment of available staff were of greater importance and were more amenable to change, and that the costs of continuous consultant presence was prohibitive (Sandall et al. The King's Fund, 2011). Moreover, issues with perceived junior status and reduced opportunities to practice independently have been raised by other Medical Royal Colleges. This may be a timely opportunity for a wider review of the structure of maternity units in the UK. Some centralisation of obstetric services may be useful, with a networked approach to provide all models of maternity care in the most efficient manner possible, including formally matching and classifying obstetric units with neonatal units (levels 1, 2, and 3). Certainly, different level obstetric units are likely to use different models of staff provision. We agree that there is unlikely to be a single solution across all maternity units, and more research in this area is imperative; however, we consider that the call for a future cluster of randomised trials may provide too narrow a focus, and there should also be mixed-methods studies investigating the sustainability of different models of care provision, including health economics evaluation, perceptions of patients and doctors, as well as clinical outcomes. Finally, it is the current generation of junior obstetricians that will be most affected by these changes, and their opinions need to be actively sought as part of any sustainable solution for the future. Full disclosure of interests available to view online as supporting information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.