More than ever, educators are expected to implement evidence-based interventions to improve student outcomes. This is often easier said than done, as illustrated by a recent study by Mathematica Policy Research in Oakland, California. To help secondary students who were several years behind in reading, the district piloted an intensive program for struggling readers that had been proven to work in early grades. The study showed that the intervention was difficult to implement in secondary schools and actually did more harm than good. The findings highlighted the importance of considering context and implementation, in addition to evidence of effectiveness, when choosing an intervention program.