This paper investigates the heterogeneous treatment effects of employment services in Korea, the National Employment Support Program. The empirical analysis reveals substantial heterogeneity in program effects. In particular, positive employment effects are salient among younger individuals, women, and those with relatively limited work experience. Furthermore, the analysis of optimal treatment assignment demonstrates that selectively targeting individuals yields efficiency gains relative to uniform service provision. Overall, the findings highlight the limitations of a one-size-fits-all approach to employment services and provide empirical support for data-driven targeting based on heterogeneous treatment effects.