We did a cost-effectiveness analysis alongside a cluster-randomised controlled trial of a participatory intervention with women's groups to improve birth outcomes in rural Nepal. The average provider cost of the women's group intervention was US$0·75 per person per year ($0·90 with health-service strengthening) in a population of 86 704. The incremental cost per life-year saved (LYS) was $211 ($251), and expansion could rationalise on start-up costs and technical assistance, reducing the cost per LYS to $138 ($179). Sensitivity analysis showed a variation from $83 to $263 per LYS for most variables. This intervention could provide a cost-effective way of reducing neonatal deaths.