Guiding principle 2: Optimal scale 

22 January 2021
SERIES The 2020 scalingXchange: building a field from the Global South 6 items

Optimal scale challenges the “bigger is better” scaling model.  Scaling produces a collection of impacts and to determine optimal scale, we have to consider the trade-offs between different types of impact.

An introduction to Guiding principle 2: optimal scale

Guiding questions

  • How have you gone about identifying the optimal scale in your work? What opportunities and/or challenges have you encountered?
  • What approach(es) have you used to understand and balance the magnitude, variety, equity and sustainability of impacts?
  • Have you encountered situations where there are different perspectives on the “right scale”, and how have you managed this?
  • What types of trade-offs have you had to consider in the process of identifying optimal scale?

General conclusions from the discussion

The conversation around the optimal scale principle confirmed the importance of recognising that “bigger is not always better”, and noted that there is still a problematic tendency to focus on magnitude as the predominant measure of successful scaling. Advisors also noted that sufficient information is critical to understanding optimal scale, but that lack of resources to pursue it are another challenge. Additionally, advisors raised the need to consider “optimal speed” of scaling processes – acknowledging that particularly in cases where we need to influence behaviours as a part of a scaling process, the speed at which we attempt to do so will likely affect the chances of success.

The following statements do not reflect all the opinions or reflections presented during the session. In some cases they reflect the ideas presented and shared in their own working groups. We have kept the name of the person who shared this particular idea during the session. 


[Petronella Chaminuka, Agricultural Research Council]

  • How you arrive at an optimal scale depends on what you are scaling. For example, an innovation for the market versus a public good.
  • Sustainability for market goods versus public goods is very different. For the public good it links back to political context and social welfare goals, whereas for private goods the emphasis is usually profitability and efficiency.
  • Insufficient information can be a big constraint to deciding optimal scale. Investment is needed in ongoing evaluation, research and multi-stakeholder engagement.
  • Sometimes knowing what optimal scale is isn’t enough to get you there. You might reach consensus on what is the optimal scale, and then not have the resources/budget/time to achieve it.

[Blanca Llorente, Fundacion Anaas]


  • Optimal scale is a mix of evidence and political analysis on the viability of scaling – especially when talking about public policy. Sometimes the political context may change and what is achievable in terms of scaling also shifts. 
  • Change may need to be incremental. In other words the demonstration effect. First you work with those who are willing to take risks because they share your vision. Then you show others that it can work.
  • Sometimes more isn’t better. But sometimes more is better and if you don’t ‘scale up’ the effects will be lost. For example, work on tobacco control: you won’t see the benefits if just one region adopts the protocol. You have to coordinate – that’s complex and requires technical resources. Every context is different.
  • As well as optimal scale, we also have to think about optimal speed of scale. If you’re talking about affecting social behaviour, the speed at which you try to do it matters in the long term.
  • Magnitude is still the predominant measure of scaling and impact. There’s a tendency to separate technical from social aspects, with technical impacts measured more than the social (quantitative versus qualitative data). Some funders want more sustainability, but for government funders magnitude seems to weigh more and other dimensions get left behind.
  • Achieving optimal scale doesn’t end with a research project. We have to ask how we can use our research and data beyond our funded project cycles.
  • Good evaluation protocols and asking the right questions at the right time are important. Principles are useful as a tool to discuss with funding institutions how to include them in evaluation processes, to make impact sustainable, to ask the right questions and come up with our own answers for the context we’re working in. 

[Johannes Linn, The Brookings Institution]

  • In my experience, the bigger problem is that often there’s no scaling at all – or it does not go far enough because we are too focused on getting a project implemented. So, we may be going wrong and overshooting the optimal scale but the more common error is to not scale [From Fishbowl 3 on Coordination].
  • Optimal scale can change over time and circumstances. Political constraints may get removed, or may shut down an optimal scale process. Monitoring matters – not just monitoring for impact but optimal scale and how that changes over time too.

[John Gargani,  Gargani + Company]

  • It’s a fallacy to think that to have a big impact we need a big action – they often get confused. Sometimes you can do more with less. For example, the way technology spreads, or we get better at doing something. 
  • Sometimes you overshoot the optimal scale and have to scale back. That’s what makes it really hard. But it’s also the important part.

[Rob McLean, IDRC]

  • This discussion reinforces the principles approach to scaling – there are no ‘rules’, we are constantly searching for optimal scale and optimal scale is constantly changing. We need to be OK with that. Be transparent. How can IDRC communicate that uncertainty is OK, and support an iterative process? Keeping the focus on impact and impact at optimal scale is part of this.