“It takes about 15 years to get a significant idea to scale”

Larry Cooley on the process behind taking an intervention to improve children’s learning and scaling it up to become systemic change
fdecomite, flickr.com, CC BY 2.0
fdecomite, flickr.com, CC BY 2.0

Caroline Smrstik Gentner: As part of Millions Learning, the Center for Universal Education project at the Brookings Institution, you’re working on designing a process to support large-scale change in education. What have you discovered so far?

Larry Cooley: Governments understand what’s needed to deliver services – and to make change – at large scale, and so do companies. But the people who fund and implement 3-5 year projects focused on specific beneficiaries don’t necessarily have the same understanding about doing and sustaining things at scale. What’s needed is a change management process that is quite different from project implementation. That change management process, which is also an adaptive learning process, needs to focus on the transition from “great idea” to “the new normal”.

CSG: This adaptive learning process was named Real-time Scaling Labs. What is its goal?

LC: The team at Brookings had already studied examples of successful scaling during phase 1 of the Millions Learning effort. In phase 2, we’re using Real-Time Scaling Labs to apply those lessons to support the scaling of a range of education interventions in a number of developing countries. These Real-Time Scaling Labs are not physical structures, they are multi-stakeholder processes established for 2-3 years to pursue three objectives.

First, they are intended to help the people who are trying to scale an intervention to do that in a more systematic way. Second, they document that process, knowing that people normally tend to write up the outcomes of their scaling efforts, but not the process they went through to get there. And third, the Labs are designed to provide a focal point for active learning and adaptive management as integral elements of the scaling effort.

CSG: How do you create a scaling plan that’s flexible, but doesn’t lose sight of the end goal or the vision?

LC: Pause and reflect. We’re designing Lab convenings at least every six months. Initial convenings focus on developing an agreed scaling plan and a set of questions or options that need to be investigated. Subsequent convenings begin with a review of the results of ongoing investigation and enquiries, and a discussion of what to do with those answers. This is the way we move forward. It’s easy to say and hard to do – how many governments or even companies do that systematically? The Labs work to instill that whole pause-and-reflect concept and to have participants come to see this as essential to – not as a deviation from – energetic, forthright implementation.

Let me tell you about a project in Tanzania. For about 25 years, the NGO CAMFED has been giving scholarships to young poor girls in three African countries (Zambia, Zimbabwe and Tanzania). Hundreds of thousands of young women in Tanzania have been beneficiaries, but what happens to them after their studies? CAMFED had the idea to take some of these young women and send them back into secondary schools in the localities where they live as volunteers, offering a Life Skills curriculum to young secondary school girls and boys.

After a very promising pilot phase and a second phase that included expansion into 10% of Tanzania’s secondary schools, the focus moved to exploring the feasibility and desirability of the Tanzanian government taking over the program and implementing it nationally. This raised a host of new questions. For example, what about boys as well as girls as volunteers? What incentives would be most effective in recruiting volunteers who are not alumni of the CAMFED scholarship program? How can volunteer training and supervision of volunteers be most effectively provided for large numbers over the long run?

Through a Real-Time Scaling Lab, participants decided on a set of questions to test in “second stage pilots” before a plan was formulated and decision were made for the whole country. In the process, the circle of interested parties grew, leadership began to transition from CAMFED to the government of Tanzania, and what had started as a donor-funded project enhanced its potential of turning into an educational change for the entire country.

“One lesson I’ve learned is that a focus on scale almost always results in a shift from a project mentality to system change mentality. And once that clicks, you can’t un-ring the bell.”

CSG: It sounds like to successfully scale a good education intervention, you need to be in it for the long run. Don’t people get impatient?

LC: In my experience, it takes about 15 years to get a significant idea to scale. A big part of the change strategy is getting everyone on the same page and willing to move ahead: clarity about a vision is only so good as your ability to get people to concur with the vision. In some cases, the governments originate the idea and that’s the simplest since any funding just helps them further their own vision. When an idea originates even partially outside government, you have to find a willing partner on the inside who sees how this can help them advance their career.

I’ve also noticed that people who are good at talking about policy are not necessarily equally versed in the strategic management work needed to translate those policies into practice. Real-Time Scaling Labs are intended in part to help traverse this divide and help people get from one step to the next.

One further lesson I’ve learned is that a focus on scale almost always results in a shift from a project mentality to system change mentality. And once that clicks, you can’t un-ring the bell. People don’t revert to old ways once they’ve had that “Aha!” moment, even if they forget everything else we talked about in terms of what a scaling plan should include. That mindset shift is more profound than I had ever expected.

Larry Cooley is President Emeritus of Management Systems in Arlington, VA, a non-resident senior fellow at the Brookings Institution, and senior advisor to the Real-Time Scaling Labs project.

Weekly newsletter