This study presents results from a conceptual replication of a promising early reading intervention. The On Track program, previously found to be effective in a researcher-driven and tightly controlled setting, was implemented in a larger sample of schools. To investigate the robustness of the original findings, specific aspects of the implementation were deliberately altered from the original study. Specifically, (i) local school authorities were in charge of the practical implementation and (ii) schools and teachers had greater autonomy in decisions regarding which student received the intervention and for how long. All schools (n = 42) in two Norwegian school districts were randomly assigned, for the first grade, either to a condition involving implementation of the On Track program for students at risk of reading difficulties or to a control condition. Results indicated that the intervention significantly reduced the number of at-risk students by the end of first grade. However, effect sizes in the replication study were reduced compared to the original try-out.