Evaluation of Conditional Cash Transfer (CCT) programmes in a developing country context indicates a positive impact on increasing enrolment but finds mixed evidence on improving learning outcomes. Using representative rural household survey data, we evaluated one of the largest CCT programmes in India, the Kanyashree Prakalpa, implemented in the state of West Bengal from 2013. We examined the effect on adolescent female school enrolment and tested whether there were concomitant improvements in learning outcomes. Employing double difference, triple difference and synthetic control methods, we found significant positive effects on female school enrolment, even for children from villages without a secondary school. The enrolment increase in heterogeneous settings indicated that transfers from CCT programmes might outweigh supply side constraints like the travel costs of schooling. However, we found that enrolment gains, while associated with improvements in lower-order learning outcomes, were related to a significant decline in higher-order learning skills. Notably, improvements in higher-order learning were only found in schools with lower teacher absenteeism and better physical infrastructure. Our analysis suggests that, without complementary investments in learning amenities, CCT programmes might only redirect adolescent girls to school without greatly enhancing the learning skills that could have a significant impact on potential economic and social (in)security.