This blog post initially appeared on the Center for Education Innovations’ blog in April 2016.

Scaling up innovations to improve outcomes and increase impact is a topic of growing interest to education practitioners, funders, and researchers, as well as to the broader development community. Below, Larry Cooley considers the tools available for assessing innovations’ scaling potential and discusses the role of such tools in informing program design and evaluating readiness to scale. He offers several recommendations on how to think about scale.

Size Matters

Doing things well on a large scale is not just a matter of expanding things that work on a small scale. For one thing, there are lots of institutions that can do things on a small scale but only three that are capable of delivering development outcomes sustainably at scale – governments, markets and, in exceptional cases, large NGOs and faith-based networks.

It is also a fact of life that assuring quality at scale requires different management approaches and different business models than delivering goods and services on a smaller, or pilot basis.

For these reasons and others, innovators and those funding innovation need strategies and tools for designing and choosing interventions with the greatest likelihood of reaching scale and delivering effectively at scale.

Four Pieces of the Puzzle

Efforts to assess scalability usually begin by trying to tease out the factors about an intervention that affect the feasibility of it reaching scale. One useful framework for doing this originated in the health sector under the auspices of ExpandNet. It uses the acronym CORRECT to summarize 7 factors that affect the prospects of an intervention going to scale – Credibility, Observability, Relevance, Relative Advantage, Ease of installation, Compatibility, and Testability.

Building on that foundation and 12 years of applied experience, my colleagues at MSI developed a 28-item Scalability Assessment Checklist that highlights identifiable characteristics of interventions that facilitate or complicate scaling. This tool, the most widely used of its type, has been applied extensively in a wide range of sectors as a means to select among alternative interventions, to design pilot programs with scale in mind, and to support the scaling process.

In her recent book on scaling up in Bangladesh, Colette Chabbott adds to the list of factors affecting scale by suggesting several organizational and contextual factors that influence scalability. These factors focus on the homogeneity of the target population and its receptivity to change; on the characteristics of the organization(s) expected to implement the intervention at scale, and on the nature of the change agents facilitating the change.

Complementing this focus on the characteristics of interventions, populations and organizations that affect their scalability, Hartmann and Linn’s work explores enabling conditions that enhance or limit the prospects of particular innovations going to scale. Specifically, Hartmann and Linn highlight 5 “Drivers” (champions, market demand, community demand, incentives and external crises) and 6 “Spaces” (political, policy, fiscal/financial, institutional, partnership, cultural) that shape the prospects for particular interventions reaching scale.

Scalability, Scaling-up and Operating at Scale

Scalability assessments are only the beginning of the story. They do not provide a “yes”-“no” answer to whether one should proceed with scaling up an intervention. Rather, they provide insight into adaptations to the intervention, changes to the implementing organizations, and changes in enabling conditions that will make scaling up possible.

And no matter how good the idea and how felicitous the setting, scaling of social outcomes rarely happens without support. There is almost always an active agent or “intermediary institution” involved, though these institutions and functions are frequently invaluable and under-appreciated. Finding organizations able and willing to develop and implement explicit scaling strategies based on the results of a scalability assessment is, more often than not, the determining factor on whether good ideas proceed past the pilot phase.

Conducting Scalability Assessments

Experience suggests that a practical way to conduct scalability assessments and to flesh out scalability plans, involves two participatory workshops, each composed of a mix of innovators, potential implementers and funders, separated by a 4-6 week fact-finding exercise. In the first workshop, participants analyze the intervention in depth from a scaling perspective and flesh out a draft scaling plan. During the fact-finding exercise, they close information gaps, conduct stakeholder analyses, and explore institutional alternatives. And during the final workshop, they refine, vet and initiate the actions needed to move concretely towards scale.

Learning and Adapting as an Innovation Scales

Scaling never proceeds in a straight line. During the scaling up process, implementers are likely to have to adapt not only the intervention but also the way it is assessed and how it addresses the enabling conditions. For example, financial constraints may not be binding at a smaller scale, but will often be so at a larger scale. Additionally, political resistance from those who might see their existing privileges eroded by an intervention might become a serious obstacle only as scaling advances. In that sense, scalability assessment is not a one-time activity, but a continuous part of the scaling up process. For these reasons and more, it is critical that implementers not only monitor and evaluate the achievement of impact at greater scale but also that they review their scalability assessment and scaling strategy frequently and test whether the enabling conditions for scaling are falling into place.

What Does All of This Mean for Implementers?

A growing body of experience testifies to the value of approaching innovation with a vision of scale, a realistic assessment of scaling challenges, and a plan for overcoming them. Scalability assessment and the tools noted herein are a good place to start. My colleagues and I are currently at work on an integrated checklist reflecting the four sets of scalability considerations described in this note. Expect a blog post on that soon.

Blog posts on the MSI website represent the views of the authors and do not necessarily represent the views of MSI.