School improvement programs that work in some places don't work in others. School improvement programs that work with some students don't work with others. Programs that appear to have positive effects in the hands of some teachers don't work for other teachers. If we can't be confident that a program will travel well, from one school or district or state to another, what's a school leader to do? Authors Berliner and Glass cite three problems with transferring programs that seem to work: there's the problem with findings (how strong were they?); the problem with replicability (there are few replications of program effects in the field of education); and the problem with fading effects. The authors suggest three considerations that can increase the chances that experimentation will lead to improvement.