How can education systems leverage existing resources to help teachers and school principals make informed decisions and prioritize resources? We study this question with a large-scale randomized evaluation of an early warning system program to prevent dropout across 4,000 schools in Guatemala, where $\sim$30\% percent of children drop out in the transition between primary and secondary school. Specifically, 1,000 schools were randomly assigned to receive a user-friendly guidance manual and half-day training on specific, evidence-based methods to help students transition to lower secondary school. On top of the manual and training, another 1,000 schools were randomly assigned to receive a list of sixth-grade students at high risk of dropping out based on a statistical model using administrative data.  Finally, another 1,000 schools received the manual, the training, the list, and two behavioral nudges to try and keep dropout prevention top-of-mind for principals and teachers. Overall, we find that all three treatment arms reduce dropout by roughly the same amount: 1.5 percentage points, from a base of 35%) Given the low cost of the program (estimated at less than US\$2 per student) and the successful implementation at scale, this program is a promising and cost-effective approach for reducing dropout in resource-constrained contexts.