What Does It Take to Scale Up Innovations?

Various education innovations are often proposed as solutions to the problems of education in the United States. Moving an innovation from a few schools to a great many, so it can have a regional or national impact, is very challenging, however.

This brief discusses the problem of scaling innovations in education in the United States so that they can serve very large numbers of students. It begins with a general discussion of the issues involved, develops a set of five criteria for assessing challenges of scaling, and then uses three programs widely discussed in the U.S. as examples of the challenges involved: Teach for America (an approach to teacher development), KIPP (a whole-school reform model) and the Harlem Children’s Zone (a school-plus-community model). Five criteria are applied to assess scalability: cost, human capacity, tools and infrastructure, political support, and external or non-school factors.

Many innovations appear to have significant additional costs—for example the additional services offered by HCZ or the longer day and year in KIPP. The non-financial challenges, however, such as being able to find enough highly skilled people, can be just as significant and are often underestimated in discussions of scaling.

One cannot assume that a popular innovation necessarily represents a potential system-wide solution. Every school system should have a way to assess the potential value and challenge in adopting innovations. The goal should be to avoid either excessive enthusiasm or excessive skepticism, but to maintain a reasoned approach that over time yields collective learning. More independent research on the costs and benefits of major innovations is also required.


  • All parties should avoid the temptation to proclaim small-scale innovations as the solutions to large-scale problems in education; rather they should be seen as promising ideas requiring further study before widespread adoption.
  • There should be a more rigorous process for evaluating promising innovations to determine both costs and benefits, involving fair evaluations done by independent and neutral parties.
  • There should be full and open access to data on costs and outcomes of innovations.