This is the second post in our series on hard-won scaling lessons (Part 1 on Why Scaling and Systems Change Go Hand in Hand here). This post examines a critical challenge that we faced while learning to integrate systems change into our scaling work: the development sector's incentives often work against the very cost-effectiveness we all claim to prioritize.

What we got wrong: Assuming cost-effectiveness incentives would align naturally

The development sector is filled with calls for cost-effectiveness in our current era of restricted aid funding, alongside many calls for implementation through government systems. However, through a series of puzzling experiences, we learned that the sector doesn't uniformly reward cost-effective scaling. This created a fundamental tension that was undermining our scaling efforts.

When funding incentives work against cost-effective impact

The "too small to fund" problem

Development banks are among the most supportive partners when working to scale effective interventions through governments. Unlike some philanthropies, they can fund governments directly and have investment vehicles built especially for that purpose. This is incredibly valuable, since getting resources to governments for scaling can be an enormous tangle of red tape. Influencing the direction of these investments towards cost-effective interventions is a high-leverage way to put evidence to work.

Yet, we have had a development bank turn down collaboration with one of our government partners for the scale up of an effective program because the program was too low-cost. In essence, this program didn’t involve enough spending, specifically spending on government staff. The program was low-cost, and cost-effective—exactly what we hear we should be emphasizing nowadays. But the development bank couldn’t get involved because the spending entailed by the program was actually too low.

We started to worry: Would responses like these push us and other organizations toward more expensive, less efficient approaches?

The spending pressure trap

In another case, we were scaling an effective program in partnership with a large non-governmental organization (NGO). We found, to our confusion, that the costs of the program seemed to be ballooning as the program was scaling. This extended to changing the design of the program to be more expensive: imagine creating large, in-person training workshops rather than conducting more cost-effective, periodic in-person outreach, as had been done in the original program that was found to be effective.

When we inquired about these decisions, we found out that the NGO felt pressure to pay more attention to spending down their budgets, than to saving money or preserving the original design of the program. From their perspective, this made complete sense: their funders incentivized them to spend, since execution of funds was one of their key performance indicators (KPIs). Once they spent their money, they’d receive more—so, there was no incentive to spend more slowly.

This was destroying the cost-effectiveness that made that intervention promising in the first place.

The scaling cost and investment dilemma

On the opposite side of the spectrum, we’ve seen governments slash investments in programs as they scale. From their perspective, this makes total sense: they are working with limited resources and are constituent-accountable. They often cannot afford the per-person cost of flagship programs, so it makes sense to cut costs where they can.

We’ve wrestled with this tension—for example, when we’ve seen a government’s per-student spending on an education program get cut by 75 percent when it went from pilot to scale. We might be tempted to celebrate the efficiency gain (and assume that some economies of scale are at play). But we worry about risks to the fidelity of the intervention: What critical, impactful program elements are getting cut in order to get the costs low enough?

Kevin Starr’s insights help reframe the challenge: We need to scale what governments can actually pay for. In that light, we’ve learned to ask: How low can costs go while preserving intervention integrity? And, further: how can we ensure that all actors are incentivized to retain effectiveness and cost-effectiveness through the scaling process?

Aligning incentives throughout the ecosystem

Since not all actors are incentivized to keep costs low and interventions effective, we’re learning how to nudge the sector towards aligning these incentives as part of the scaling process. Here are a few recommendations from our work:

• For funders: We've learned to advocate for (a) moving beyond spending targets to impact metrics, and (b) sustainable investment levels that preserve intervention quality. We’re also working on social return on investment (SROI) models to help funders assess the ROI of their investments in terms of impact on the communities they’re serving
• For governments and implementers: We offer technical assistance on effective scaling, to help partners identify which aspects of programs can be streamlined without losing effectiveness. We’re also conducting cost-effectiveness and implementation-focused research to answer the question, “How low can costs go while preserving intervention integrity and effectiveness?”

The incentive alignment challenge taught us that cost-effectiveness isn't just about designing efficient interventions—it's about creating an ecosystem where efficiency is rewarded throughout the scaling process. We've learned that addressing these misaligned incentives isn't a side project; it's central to making scaling work. When funders reward spending over impact, when implementers are incentivized to inflate costs, and when governments are forced to cut so deeply that programs lose their effectiveness, even the most promising interventions will fail to reach their potential. 

This work is ongoing and imperfect—we haven't solved the sector's incentive problems, but we've learned to work within them more strategically. By starting these conversations early, building fidelity monitoring into our partnerships, and helping all parties understand the real costs of effective interventions, we're finding ways to preserve impact while achieving the scale we need. Getting the incentives right from the start has become as important to our scaling work as getting the evidence right.