Many countries conduct national assessments of educational performance, a type of standardised test that gives schools diagnostic feedback without having high stakes for students. We examine how publication of national assessment results at the school level affects primary school student outcomes. Using a policy reform in Japan that created a variation across municipalities in the level of aggregation at which the assessment results are disclosed, we show that disclosing school-mean scores increased students' test scores by 0.12 standard deviations without compromising educational equity. The information disclosure worked as a catalyst to induce the schools to utilise the assessment's diagnostic feedback.