Abstract Gender-based violence (GBV) affects one out of three women in the world, with long-term welfare consequences for victims and families, and indirect costs to the health sector, the legal system, and the economy. Yet there has been little rigorous research on the efficacy of interventions that aim to reduce or prevent GBV. As a result, policymaking has been guided by “best-practice” manuals and little robust evidence. An emerging consensus centers on multi-level and gender transformative approaches, but there is a lack of systematic research on the effectiveness of this approach or insight into key components of programming. The goal of this project is to deliver new rigorous evidence to the scientific and policy community by experimentally evaluating the impact of alternative versions of a state-run GBV intervention in Peru. Our evaluation will provide evidence on the effectiveness of distinct program components, assess their cost-effectiveness and potential to scale, and shed light on mechanisms underlying changes in GBV. This project takes advantage of a long-standing partnership with the Peruvian Ministry of Women (MOW) to conduct an experimental evaluation randomized across 250 villages of Leaders in Action (LIA), the MOW’s flagship GBV program that trains local leaders on gender violence and norms. We experimentally assess the impact of its two main components: a household-based module (HT), consisting of household visits by trained leaders, and a group-based module (GT) with education sessions in small gender-segregated groups organized by trained facilitators. We will cross-randomize each approach to assess their efficiency in reducing domestic violence and changing social norms around tolerance toward violence and gender roles. The study disentangles the impact of the two modules separately, as well as their interaction, while explicitly addressing methodological concerns of previous studies. Reporting bias from self- reported measures will be addressed with a fully private self-administrated instrument, and a large clustered sample will address statistical power limitations of previous studies. In addition, comprehensive male and female survey instruments are designed to shed light on the potential mechanisms and theoretical framework underlying GBV. Finally, our experimental design will allow us to disentangle short and long-term effects, and the efficacy of separate delivery modes. This is a unique opportunity to evaluate government programs to guide GBV programming in Peru and worldwide, estimate their cost effectiveness, and bring scientific evidence on GBV reduction and prevention. The MOW is eager to incorporate the results to its programming of the Peruvian National Plan Against GBV. The project includes policy outreach activities to guide MOW’s policymaking through workshops, policy briefs and capacity building sessions. We expect future research to address the accuracy of extrapolating our findings to other Latin America...