This study investigates the causal impacts of two iterations of a large-scale nutritional intervention to promote the self-development and empowerment of adolescent girls in India. We use a difference-in-differences strategy to evaluate the program’s effects on education and health outcomes. Our findings from the first program iteration reveal that while it successfully reduced the prevalence of anemia among adolescent girls by 2.4%, it also had unintended adverse consequences on their educational attainment. Specifically, the intervention led to an 8% increase in school dropout rates for girls and widened the gender disparity in school enrollment by 6.4%. These negative outcomes are primarily attributed to a program component that conditioned additional nutritional supplements on being out of school. Heterogeneity analysis indicates that these effects were driven by families residing in villages without secondary schools and with low levels of maternal education. In contrast, our evaluation finds that updates made to the scheme during the second iteration have no significant impact on girls’ school dropout rates.