Abstract This paper investigates the relationship between human resource practices and educational mismatch. It introduces a novel firm-level measure of educational mismatch, constructed by merging firm- and individual-level data at the sector-firm size-year level. The measure captures both the incidence and the type of mismatch – distinguishing between overeducation, undereducation, and mixed cases – and enables a detailed analysis of how human resource practices relate to skill utilization within firms. Using a rich dataset on Italian firms, we estimate how this measure correlates with different human resource practices, controlling for an extensive set of firm characteristics, as well as year and industry-region fixed effects. The results indicate that on-the-job training is consistently associated with lower levels of educational mismatch. Narrower spans of control and the use of private recruitment agencies are also linked to reduced mismatch, although their effects vary across sectors and types of mismatch. In contrast, public employment services show no systematic association with educational mismatch, while second-level wage bargaining is negatively related to mismatch in firms characterized by overeducation, but positively related in those facing undereducation.