This paper studies the impact of a ‘textbook’ ad-valorem export subsidy on firm-level export performance. The Cash Incentive Scheme for Exports (CISE) program offered by the government of Nepal offers a cash subsidy to firms exporting a select group of products to countries other than India. Using customs transactions data combined with subsidy disbursements at the firm level from 2011 to 2014, we estimate the impact of the subsidy on exports of targeted and non-target product-destination combinations and their extensive and intensive margins. We employ a range of doubly-robust matching estimators to control for the non-random selection of exporters into the scheme. We find that subsidized firms increased their exports of targeted product-destinations relative to firms in the control group and that this rise is fully accounted for by the extensive margin: a higher number of targeted products exported and foreign markets served. We do not find any significant changes along the intensive margin nor among non-targeted product-destination combinations. While our results show that the CISE scheme fomented export diversification, its limited impact on total exports and high fiscal cost call into question its effectiveness.