Evidence-based policymaking (EBP) relies on an epistemological assumption that evidence from randomized controlled trials (RCTs) is the finest evidence for policy formulation, while expert testimony is the poorest one. This paper argues that while RCTs are a valuable source of empirical evidence for policy interventions, they are not sufficient on their own to support evidence-based policy formulation. Through the lens of the INUS framework of causation, we demonstrate that the effectiveness of a policy is influenced by a complex interplay of contextual factors, which RCTs alone cannot capture. We advocate for the integration of contextual and qualitative knowledge, including testimonies from experts and community members, to supplement RCT findings. This additional knowledge provides insights into the social, cultural, and subjective dimensions of the target population, addressing motivations, preferences, and other factors that can significantly impact policy success. By comparing reductionist and non-reductionist perspectives on the use of testimony in evidence-based policy, we argue for a balanced approach that values credible testimonies as essential to understanding context. Ultimately, this paper underscores the importance of a multifaceted evidence approach in crafting effective, context-sensitive public policies.