New technologies offer many promises to improve student learning, but efforts to bring them to the classroom often fail to produce improvements to student outcomes. A notable exception to this pattern is one-to-one laptop programs. While early evaluations of these programs have been encouraging, they are costly to implement, and no study has investigated the impact of a one-to-one technology program implemented on a large scale over a multiyear period. With administrative school data, this paper uses a differences-indifferences strategy to evaluate the impact of a one-to-one laptop program implemented in a midsize school district. We find that while short-term impacts of the program were modest, math scores improved by 0.15-0.17 standard deviations in the medium term (4-5 years post-implementation). We also investigate heterogeneity in impacts on test scores and the impact of the program on several measures of student behavior.