People with disabilities experience stark inequities across the spectrum of development priorities. Yet they are rarely prioritised by development programmes. The poor evidence base for ‘what works’ for disability-inclusive development is a major challenge to taking action and advancing the meaningful inclusion of people with disabilities. The Programme for Evidence to Inform Disability Action (PENDA) was established to undertake high-quality impact evaluations to assess whether holistic approaches to disability-inclusive development improve the participation of people with disabilities in key areas of society (health, livelihoods, education). Through PENDA, 13 evaluations were implemented across 13 low-and middle-income countries. Impact evaluations are predominantly through the gold standard approach of randomised controlled trials. Each trial has a complementary mixed-methods process evaluation to better understand the findings and their implications. The article reflects on key considerations for undertaking impact evaluations of disability-inclusive development interventions, illustrated from the PENDA trials, and learnings to advance this research agenda.