I am writing this from Accra, where the conversation among the organizations I meet has changed in tone over the past year. The question used to be which call for proposals to chase. Now it is a quieter one, asked more carefully: who will we report to next year, and what will they want to see?
What changed
In September 2025 the United States Department of State published the America First Global Health Strategy. The core structural move is a shift toward direct bilateral agreements between the United States and recipient governments. As of early August 2026, 34 countries had signed. The stated aim is to reduce the share of health assistance consumed by technical assistance, program management, and other overhead, and to transition responsibilities to national governments, the private sector, and faith based organizations. The strategy makes its case with a number: less than 40 percent of health foreign assistance reaches frontline supplies and healthcare workers, while about 60 percent goes to technical assistance, program management, and other overhead. It reports that implementing partners spent 678 million dollars on program management in fiscal year 2024.
Reasonable people disagree about that framing, and this article is not the place to settle it. My interest is narrower and more practical: what happens underneath, to the organizations that were never party to those agreements but were always at the end of them.
The layer that thins
Health assistance has moved through a layered structure. A prime implementing partner holds the award. Beneath it sits a wide field of sub recipients, and those sub recipients are overwhelmingly local organizations doing the delivery. The strategy itself gives a sense of the crowding. In fiscal year 2024 Uganda had 57 implementing partners and more than 300 sub recipients. Tanzania had 38 implementing partners and more than 350 sub recipients. Read those numbers again with an operational eye. Hundreds of local organizations in a single country, each attached to a prime, each delivering against someone else's award.
Here is the part that matters and that I rarely hear discussed. When a prime carries the reporting, the organization underneath never has to build that capability. The prime designs the indicators, runs the data quality checks, assembles the narrative, and answers the questions when they come. The local organization delivers the work and hands over what it is asked for. That arrangement can be efficient. It is also a kind of dependency, and it is invisible until it ends.
Remove or thin the prime layer and the reporting obligation does not disappear with it. It lands. Under the bilateral agreements, performance is monitored across four areas: service delivery, epidemiology, supply chain, and the recipient country's own investment. Performance against those metrics is required to trigger future year United States funding. The strategy also recommends off the shelf private sector solutions for data systems rather than systems designed and managed entirely by governments, which tells you how central the data question is to the whole design. Money will follow evidence, and it will follow it faster and more directly than before.
What the data says about readiness for that
This is where our own numbers become uncomfortable. Across 428 assessments from 413 organizations in 36 countries, the NGO Readiness Assessment produces a consistent shape. On a five point scale, organizations average 4.15 on Mission and Strategic Clarity and 4.02 on Programs and Program Design. They average 3.86 on Systems, Governance and Readiness, 3.62 on Evidence, Data and Storytelling, and 3.40 on Stakeholders and Positioning. Clarity of purpose is not the problem. Proving what that purpose produced is.
Three caveats belong right here, next to the numbers, not in a footnote. These scores are self reported. The sample is self selected, made up of organizations that chose to assess themselves, which likely skews toward the more reflective end of the sector. And some country samples are small, Tanzania most of all at 11 organizations. Treat what follows as a signal worth investigating, not a measurement of a national sector.
With that said, the pattern is hard to ignore. Forty three percent of organizations score below 3.5 on Evidence, Data and Storytelling. Sixty percent score higher on mission clarity than on evidence, and for 32 percent of them the gap is a full point or more. Now place the country figures beside the strategy's own examples. Evidence scores average 3.17 in Tanzania across 11 organizations and 3.39 in Uganda across 25. Ethiopia sits at 3.31 across 14. Ghana averages 3.52 across 48, Kenya 3.70 across 59, and Nigeria 3.82 across 70. The two countries the strategy singles out for the most crowded implementing partner layer are among the weakest on evidence in our dataset. That may be coincidence in a small sample. It may also be exactly what a thick prime layer produces over two decades: delivery capability that is real, and evidence capability that was never asked for locally because someone above was supplying it.
The gap between what organizations ask for and what they need
One more finding sits underneath all of this, and it is the one I keep turning over. In the same dataset, 84 percent of organizations name fundability and proposals as a focus area. Systems and governance trails at 63 percent. Organizations ask for help raising money far more often than they ask for help proving outcomes. That is understandable. Cash is urgent and evidence feels like homework. But in a bilateral environment where continued funding is tied to performance metrics reported by the recipient side, the two things stop being separable. The proposal skill that won you a sub award may not be the skill that keeps you in the system.
The open question
So here is where I land, which is short of a conclusion. The transition could produce exactly what its architects describe: national systems that own their own data, local organizations that build capabilities they were previously allowed to skip, and a larger share of money reaching the people it was raised for. That would be a genuine gain, and the localization argument has wanted something like it for years. Or it could move a reporting burden onto the organizations least equipped to carry it, with the same delivery expected and less support behind it, and the quiet result would be consolidation toward whoever already had the systems.
I do not know which of those we get. I suspect it will not be uniform, and that the deciding variable will be unglamorous: whether local organizations build evidence capability in the next 18 months, before the layer above them finishes thinning. That is a capacity question, a funding question, and a design question all at once, and it is being answered right now, mostly by default and mostly in silence. Anyone who wants to see where their own organization stands can take the NGO Readiness Assessment free at readiness.alleviate.org.
Robert Okine is Founder and CEO of Alleviate Technologies, and has spent fifteen years building digital systems for civil society across 60+ countries.
