The most dangerous idea in international development right now is held almost universally - questioned almost never - that if something works, you should do more of it.
We call it scale, and sure, scale has done extraordinary things.
In 2004, people were dying of HIV at a rate that defied comprehension. The emergency response — get testing out, get treatment out, fast, everywhere, now — was the only moral response available. People came back from the edge of death. Practitioners who were there still describe it in terms that sound almost religious: the Lazarus effect, they called it. You watched someone who had been bedridden, wasting, unreachable — and then they came back.
Scale did that.
But here is what that type of scale could not do:
It could not address why those same people were vulnerable in the first place: The economic conditions, the gender dynamics, the systems of power that limited someone's ability to make decisions about their own body and health. Those things required time, and relationship, and the willingness to stay in one place long enough to understand it.
There are two different types of scale:
Scaling Wide means reaching more people with the same model. You build something that works, you prove it, you replicate it. This is what most funders mean when they ask "can it scale?" — and it's the logic that has dominated international development for the past two decades. Bigger is better.
Scaling Deep means staying with a community long enough to address not just the presenting problem but the conditions that sustain it. It means accepting that the solution for a community in one catchment area may not transfer as near as 15 kilometers west. It requires building community assets, leader capacity, and organizational resilience that can outlast the funding relationship.
Here is what the difference between Scaling Wide vs deep looks like in practice:
The opioid crisis in the United States has killed over half a million people in the last two decades. The acute phase of the response looked like scale-wide thinking at its best: naloxone distribution, emergency room protocols, hotlines. Get the antidote out. Save lives immediately. It worked, in the sense that people didn't die who would have otherwise.
But the opioid crisis is not an acute problem. The conditions driving it are irreducibly local, which means they require a solution that scales deep.
What drove addiction in rural Appalachia is not what drove it in suburban New England or in Native American communities.
The groups that have made the most durable progress are the ones where someone stayed. A recovery coach who grew up in the same town. A church that became a de facto support network. A county health department that spent five years building trust with the same families.
Twenty years into the crisis, overdose deaths are still near record highs — even as naloxone access has never been better. The antidote to the emergency was deployed. The antidote to the conditions was not.
The difference between Scaling Wide and Scaling Deep shows up everywhere we try to solve things at a distance.
…And why we’re not Scaling Deep enough right now
If we want to address the root cause of an acute problem, it makes sense to scale deep, but still, we're not funding frequently enough in that direction. Here's why:
Scaling Wide gives us clean numbers: Cost per beneficiary. Lives reached. Replication sites. These are metrics that translate across contexts, and that a funder can evaluate.
Scaling Deep produces outcomes that are harder to name. Using the HIV crisis, Scaling Deep is less like counting pills distributed and more like asking someone who found a support group after years of managing HIV alone, how their life has changed.
Beyond the metrics, there is the problem of time. Scaling Wide is built for the three-year grant cycle; it is a sprint toward a number. Scaling Deep is a generational marathon. It requires a "risk of time" that most funding structures aren't designed to hold. When we fund for "impact now," we inadvertently punish the slow, quiet work of building trust and local resilience—the very things that ensure a solution survives once the outside funding stops.
The development sector is in its own acute phase right now.
USAID funding has collapsed, and we are scrambling to replace what was lost.
But acute responses solve for the emergency. They don't solve for what created it.
The chronic problem is this:
We've been measuring what funders can count, not what communities say matters.
Scaling Deep means going back to communities and asking: how would you know if this worked? And then building the measurement framework around that answer.
It requires everyone to agree that "we don't have the metric for this yet" is a legitimate place to start a conversation - one that works toward a new measurement vocabulary.
This was inspired by a podcast conversation with Nadia Kist, Executive Director of Blood:Water. It is part of an ongoing series of talks with practitioners who are rethinking what funding can look like. More to come!
