Talking points: Insights on evidence and its implementation

The questions, challenges and dilemmas raised at the biennial Evidence and Implementation Summit (EIS) don’t end with the event. Common threads arise – across sectors, professions and places – that continue to resonate long afterward. 

Over the next few weeks, we'll share some of what has stayed with us: the ideas, tensions and insights from two days of some of the sharpest thinking in our field. 

First up: scaling. Why is it so challenging, and what do we know about making it work? 

Scaling: How can we make greater impact for more people, more quickly? 

Millions of people around the world could benefit from evidence-informed solutions, yet few promising innovations ever make the leap to widespread impact. The challenges of scaling up social innovations are common across sectors, making this a lively topic of debate at EIS and in the months since.  

For funders and those commissioning big initiatives, there are significant questions: decisions on what to support, how best to do this, for how long, with which partners, and when and what to evaluate. At EIS, we heard from leaders in implementing organisations, philanthropy, intermediary organisations and government about how scale looks from their various perspectives. 

Across all these actors, there is clear alignment that scale must be our aim if we are to improve the lives of people facing adversity. CEO of Veddis Foundation, Murugan Vasudevan, made a compelling call for the “dire, urgent need for funders, researchers, policymakers and philanthropy to… solve problems at scale now with evidence we already have.”  

The scaling challenge 

Scaling is more than expansion or replication, and different from testing a promising program at more sites. It is an intentional process to maximise reach and effectiveness for sustained impact, so everyone who could benefit from an innovation can experience it, and benefit is optimised for all. 

The challenges of scaling are huge: adapting for different contexts, building the necessary political and system alignment, ensuring equity in delivery, and maintaining expected benefits in real-world conditions. Scaling is intentional: it isn’t about doing more of the same, it’s about adapting to significant complexity as an intervention moves into different contexts and reaches different user groups. 

Or, as Executive Director of J-PAL Europe, Cillian Nolan, observed in his invited talk: all real-world projects require continual “thinking about messy conditions and what goes wrong. But as we scale things up, many more things will go wrong.” 

Scaling science 

Even so, the challenges are not insurmountable. There is a science to scaling and there are frameworks, factors and conditions that contribute to success. CEI’s projects with partners such as Education Endowment Foundation and Movember on scaling approaches are evidence that funders are seizing the science and putting it to work to support their initiatives.  

EIS speakers reinforced this. For example, Dr Benjamin Tan from Innovations for Poverty Action, noted that the science is not new, and we should draw on what is already known about how to scale. “The basic process is fairly well understood,” he said. “We know the most recurrent challenges and themes that are going to come up in relation to implementation or scalability, depending on what stage you’re at.”  

The research reinforces that there are multiple phases in a scaling effort, including real-world testing and refinement of the delivery model, identifying the core effective components, and how to adapt them for new contexts in the most cost-effective, simplest way. In tandem with this, it is vital to work within an enabling policy environment, while exploring the partnerships and cultivating the demand that will support scale. 

Affordability is also a key consideration: the initiative must be viable for a payer to sustain over time, whether in the public or private sector. Viji Iyer, Director of JPAL’s ASPIRE scaling unit noted the key role of cost-effectiveness. Organisations must, she advised, explore how to “change elements of the program to shrink the cost.”  

Context and collaboration are key 

A core truth of implementation science also holds for scaling: how things are done is often as important as what is done. Working collaboratively and responding to context are key to making a difference. 

“It’s clear that the simplicity of the innovation is vital, as is its adaptability,” says CEI Director, Dr Cheryl Seah. “But very often the effort stalls or fails due to a lack of enabling conditions: are you building effective partnerships, is the right connective infrastructure in place, is there community trust and demand for your innovation? This is the hard work of scaling. And it’s very much a team sport, with multiple actors working together.” 

Associate Professor Harriet Koorts from Deakin University pointed to the importance of finding ethical, equitable and responsible ways to scale: “When you’re scaling one intervention, you may require simultaneous down-scaling of others… We can only scale so much and decision-makers can only back so much… [And] just because we have evidence doesn’t necessarily mean we should scale… Scaling can widen inequities.”  

Multiple EIS speakers also drew attention to intent. “Scaling up” the enabling policy and institutional and environment is different to “scaling out” the program or innovation, and different still from “scaling deep” hand-in-hand with the community. In a panel discussing experiences of the Movember Scaling What Works Fund, Melissa Abu-Gazaleh, Founder and Managing Director from Top Blokes Foundation, said: “While we were scaling out… we also really tried to prioritise scaling deep… building the right foundations [and] embedding ourselves in the community.”  

Moreover, experience shows that scaling is not a problem to be solved once. It is an ongoing adaptive “dance” of alignment and realignment, that can navigate changing systems, context, relationships and politics.  

This is work the CEI team returns to often, with partners across systems, sectors and geographies. We are currently exploring how to bring together these and other rich insights on scaling in tools and resources that will advance work in the sector.  

If you’re also keen to dig in more on scaling, we would welcome a conversation. And if you want more on any of the ideas raised here, these resources are a good place to start: