The world has the science to transform food systems. The next frontier is scaling it
Global food systems face climate, nutrition and market crises, yet proven solutions stay stuck in pilots. Scaling must become a core scientific discipline, asking early who will deliver and fund innovations, what regulations and incentives apply, and how systems must change for lasting adoption.
SummaryCGIAR’s Scaling for Impact Program shows that the bottleneck in food systems is not a lack of technology but weak scaling capacity. By validating proven tools, linking them to major investments and public delivery systems, and planning for long-term uptake, the programme illustrates how disciplined scaling can turn pilots into durable outcomes. The world’s food systems face real and urgent challenges. These include climate change, nutrition insecurity, food safety, and unequal access to markets. Research has produced practical solutions to each of these that could benefit hundreds of millions of people. Too few are moved into widespread use. For years, the development sector has flattered itself with pilots. A new tool works in a controlled pilot, a crop variety performs well in a field trial, and a digital advisory service shows promise in early testing. Evidence is written up, a case study or experiment is published, and then comes the familiar refrain: now we need to “take it to scale”. That is the moment the real difficulty begins. Solutions do not spread simply because they are good. They move, or fail to move, through systems where scientific supply and demand for innovative solutions are frequently misaligned. Policy environments are not ready, financing is difficult to mobilise, demand is weak, and markets are not designed to carry promising ideas beyond their pilot phase. These are some of the challenges that have facedCGIAR, the world’s largest publicly funded research-for-development partnership focused on agriculture and food systems. But these challenges are not limited to CGIAR alone; they are common in research for development. The world does not just need more breakthroughs. It needs more organisations that know how to turn scientific advances into adoption, investment and lasting use. Put plainly, it needs stronger efforts to move proven science into widespread use. That sounds abstract. It is not. As director of CGIAR’sScaling for Impact Program, which works with partners across Africa, Asia and Latin America to connect innovations with the systems and investments needed to scale them, I have seen this pattern myself. The evidence from that work consistently points to the same conclusion: scaling must be treated as a core part of the scientific process – built into research from the start, with systems thinking prioritised, not treated as a final phase. What it takes to scale upScaling is about asking different questions earlier in the research process – identifying the challenges that prevent innovations from moving into use, and charting strategies and actions to overcome them. Not just: does this solution work? But: who will deliver it, who will pay for it, what incentive do they have, what regulations apply, what evidence unlocks funding, and what has to change in the surrounding system for uptake to last beyond a project cycle?Those questions are rarely asked early enough in the research process. Yet they determine whether a promising idea becomes a public good or another stranded pilot. One example of what this looks like in practice is a “clearinghouse” created under the African Development Bank’sTechnologies for African Agricultural Transformationprogramme, now integrated into theCGIAR Scaling for Impactprogramme. Its role is not to invent new technologies, but to make proven ones usable at scale: validating them, packaging them with complementary innovations, and linking them to large public investments and agricultural delivery systems. Thatmodel is now positionedto connect agricultural innovations to a US$1. 5 billion AfDB-backed portfolio in 2026, expected to benefit 3. 4 million additional smallholder farmers. In Nigeria, it helped connect heat-tolerant wheat varieties – developed to maintain yields as temperatures rise – to an AfDB-financed programme backed by US$134 million, contributing to a sharp expansion in wheat area over two years. The point is not only that the varieties worked. It is that someone built the bridge between science and investment. Sometimes the real bottleneck is not the innovation itself. It is the absence of systematic scaling support for the organisations working to deliver it. That is why building scaling capacity matters. In 2025, Enabel, Belgium’s development agency, drew on the Scaling for Impact Program’s scaling fund to apply a structured approach to two African innovation projects:Tap & TrackAsset Management in Uganda and theAbalobi Monitorfisheries platform in the Western Indian Ocean. The value of the support, through the Scaling for Impact Program’sscaling fund, was not simply more investment. It was a more disciplined way of thinking about scaling. Enabel found the approach “practical” and “doable” because it surfaced constraints the teams had not previously recognised as part of the innovation system. These included government roles, how systems need to connect, institutional gaps and coordination failures. For Tap & Track, that process fed into a medium-term scaling plan and a long-term ambition to reach 30 utility companies across seven countries. What the support produced most clearly was stronger planning and strategy. And that is precisely the point: scaling capability is itself part of the infrastructure of impact. These examples point to the same conclusion. We spend a great deal of time celebrating innovation and far less time understanding how innovations can be moved into use. But the gap between a successful pilot and a durable outcome is where much of the important and scientifically exciting work sits. It is where solutions are translated into investment cases, fitted into public and private delivery systems, adapted to institutional realities, and made credible to the actors who have to carry them forward. That is why scaling should not be treated as a final phase or a dissemination exercise. It should be treated as a discipline. A capability. A scientific endeavour in its own right. Good science remains indispensable. But it is not self-propelling. The science to address food systems challenges exists. What remains insufficient is the systematic capacity to move proven innovations into widespread use at scale. Building that capacity and treating scaling with the rigour it requires is among the most important tasks facing food and agricultural research today. Timothy Krupnik, Director - CGIAR Scaling for Impact,CGIARThis article is republished fromThe Conversationunder a Creative Commons license. Read theoriginal article.