The most operationally demanding role of my career involved coordinating a single portfolio across more than twenty countries in Africa, the Americas, and Asia — funded in part through government development budgets, delivered through local partner organizations, and reported against a shared evaluation framework that had to make sense in every one of those markets simultaneously. Multi-country partnership delivery is a different discipline from single-market partnership management, and most of what is written about “partnerships” assumes the single-market case. This is the field guide I wish I had going in.
The scale of the system you are operating inside
If your multi-country program touches official development assistance, it is worth understanding the system it sits inside. Net ODA flows from all providers to developing countries totalled roughly USD 255 billion in 2024, with DAC member countries contributing the largest bilateral share. But that share has been shrinking: DAC members accounted for 61% of ODA to developing countries in 2023, down from more than 70% a decade earlier, as multilateral channels and non-DAC donors have grown their footprint.
That shift matters operationally: a multi-country program funded primarily through one bilateral donor a decade ago is now more likely to sit inside a blended funding structure with multilateral co-financing, each layer carrying its own reporting cadence, procurement rules, and risk appetite. Coordinating twenty country offices is hard enough without also reconciling three donors’ definitions of “results.”
Four structural problems unique to multi-country delivery
1. One theory of change, twenty operating realities
A global program needs a single theory of change that survives translation into twenty different regulatory, political, and market contexts. The mistake I see most often is writing that theory of change at headquarters and treating country-level adaptation as an implementation detail. It isn’t — it is the actual work. The stronger model is to write the theory of change at the level of the outcome you want, and let country teams own the pathway, with headquarters holding the measurement framework constant.
2. Currency, procurement, and fiscal-year mismatches compound quietly
A budget approved in one currency, spent across a dozen others, against fiscal years that don’t align with the donor’s own reporting cycle, creates a slow-motion reconciliation problem that rarely shows up until year two or three. Building a single multi-currency, multi-fiscal-year tracking model at the start — however unglamorous — is cheaper than the alternative, which is a finance team reconstructing it retroactively under audit pressure.
3. Local partner capacity is the actual constraint, not funding
In most of the twenty-plus country portfolios I have seen, the binding constraint on program pace was never money. It was the absorptive and administrative capacity of local implementing partners — their ability to manage procurement, safeguarding, and reporting requirements calibrated to a donor used to working with much larger institutions. Programs that invest early in partner capacity-building outperform programs that treat capacity as something partners should already have.
4. Change management has a measurable funding dividend
In the program I coordinated, applying change-management discipline — renegotiating scope with funders when country conditions shifted, rather than quietly absorbing the mismatch — helped attract roughly 30% in additional funding over the program’s life, by demonstrating to funders that the program could adapt without losing accountability. Funders reward visible adaptability far more than they reward the appearance of a plan executed exactly as written.
Why evaluation has to be designed in from day one
Before I moved into partnership and business development roles, I spent several years as an evaluation officer, coordinating program evaluations across more than thirty countries. That experience shaped how I now think about multi-country program design: evaluation cannot be a retrospective exercise bolted onto a program in its final year. It has to be part of the original design, for a simple reason — the indicators you can credibly report on in year three are determined by the data infrastructure you built (or didn’t build) in year one.
Programs that treat evaluation as a compliance requirement typically discover, two or three years in, that they cannot answer the question funders actually care about: not “did we deliver the activities,” but “did the activities produce the change we said they would, and how do we know.” Retrofitting a credible answer to that question after the fact is far more expensive than designing the measurement framework alongside the program logic from the start.
Currency, procurement, and fiscal-year mismatches: a worked example
To make the mismatch problem concrete: a program budgeted in the funder’s currency, disbursed to country offices in a mix of local currencies, with country fiscal years that in some markets start in January and in others start in April or July, and a donor reporting cycle that follows yet another calendar. Left unmanaged, this produces a familiar pattern — a finance team discovers in month 30 that currency movements have created an apparent variance against budget that has nothing to do with program performance, and now has to explain that variance to a donor mid-cycle, using data that was never structured to make the explanation easy.
The fix is unglamorous but effective: a single multi-currency budget model, built at the start, that tracks committed and spent amounts in both the donor’s reporting currency and each country’s local currency, reconciled monthly rather than annually. It costs a few weeks of finance-team setup time. It saves months of retroactive reconciliation under audit pressure later.
Building local partner capacity as core program design
The instinct when a local partner is struggling with procurement or reporting requirements is to add oversight — more headquarters check-ins, more required documentation. This usually makes the underlying capacity problem worse, not better, by consuming the partner’s limited administrative bandwidth on compliance rather than delivery. The more effective approach treats capacity-building as a budget line and a program workstream in its own right: dedicated training on the specific donor’s procurement and safeguarding requirements, delivered early, with a named point of contact the partner can return to with questions, rather than a generic manual and an annual audit.
Programs that fund this properly typically see it pay for itself within the first reporting cycle, in the form of fewer compliance findings, faster disbursement approvals, and partners who can absorb program scale-up without a proportional increase in headquarters oversight.
A practical structure for governance across markets
| Layer | Owns | Cadence |
|---|---|---|
| Global steering | Theory of change, funder relationship, top-line results | Quarterly |
| Regional coordination | Cross-country learning, shared risk register | Monthly |
| Country delivery | Local pathway, partner management, on-the-ground reporting | Continuous |
What I would tell someone taking on their first multi-country portfolio
Resist the temptation to standardize everything for the sake of comparability. The programs that hold together across twenty markets are the ones that standardize outcomes and measurement, and deliberately localize almost everything else. And build your funder relationship as a change-management partnership from day one — the data on where ODA is moving suggests funding structures will keep getting more complex, not less, and the programs that can demonstrate adaptive capacity are the ones that keep attracting capital when budgets tighten.
Template: the country-readiness checklist before launch
Before adding a new country to an already-running multi-country program, I use a fixed readiness checklist rather than relying on general enthusiasm from the country team. It covers five questions: Does a local implementing partner with a track record on donor-compliant financial reporting already exist, or does one need to be developed from scratch? Is there a named in-country focal point who will sit on the regional-coordination calls, not just receive minutes afterward? Has the country’s fiscal year and regulatory calendar been mapped against the program’s existing reporting cycle? Is there a realistic estimate of the capacity-building investment required before this country can absorb its planned share of program funds? And critically, has the risk register been updated to include this country’s specific exposures before launch, rather than after the first incident? Programs that run this checklist before expansion consistently onboard new countries faster and with fewer first-year compliance findings than programs that expand opportunistically and build the checklist’s answers retroactively.
What good looks like by year three
It is worth being concrete about the payoff, because multi-country program design asks funders and country teams to invest heavily upfront for benefits that only become visible later. A program that has done the structural work well by year three typically shows a specific pattern: country teams that can answer donor questions about their own results without escalating to headquarters first; a risk register that has been updated often enough that steering-committee members trust it as a live document rather than a formality; local partners who have grown their own institutional capacity enough to take on adjacent funding from other sources, using systems built for this program; and a funder relationship mature enough that the renewal conversation is initiated by the program team well ahead of the funding cycle’s end, rather than the other way around. None of that is visible in year one. All of it is determined by decisions made in year one.
What actually belongs on a shared risk register
A multi-country program’s risk register is only useful if it is specific enough to act on. Generic entries like “political instability” or “currency fluctuation” tell a steering committee nothing it can use. A working register instead names the specific exposure per country — for example, a named election date that could delay a government counterpart’s sign-off in one market, or a specific local partner whose single largest funder is separately at risk of withdrawing, creating a dependency the program did not originally plan around. Reviewing and updating this register at the regional-coordination cadence, not just at the global steering level, is what keeps it from becoming a static document nobody revisits until something has already gone wrong.
Frequently asked questions
How many countries is “too many” for a single coordinated program? There is no fixed number, but the operational strain tends to jump sharply past the point where a single regional-coordination layer can still hold direct working knowledge of every country team. In practice, that threshold is often somewhere between eight and twelve countries per coordinator; programs spanning twenty or more almost always need at least two layers of regional coordination beneath the global steering level, not one.
Should evaluation be run by the same team delivering the program, or independently? Both roles are necessary but should not be the same function. Program teams need real-time delivery data to manage adaptively; funders and governance bodies need a credibly independent evaluation function to trust the results reported. Collapsing the two into one team under resource pressure is a common and costly shortcut.
What is the earliest sign a multi-country program is heading for a capacity crisis with a local partner? A rising backlog of unanswered procurement or compliance questions from that partner is a far earlier and more reliable signal than a missed reporting deadline. By the time a deadline is missed, the underlying capacity gap has usually existed, unaddressed, for several reporting cycles.
Is it better to standardize the reporting template across every country, or let each country report in its own format? Standardize the indicators, not the format. What headquarters and funders need is comparability on outcomes; how a country team collects and structures the underlying evidence should be flexible enough to fit local data systems, or the reporting burden itself becomes a barrier to the program’s actual work.
Takeaways
- Understand the funding architecture you sit inside — bilateral, multilateral, and blended financing each carry different reporting and risk logic.
- Fix the theory of change at the outcome level; let country teams own the pathway.
- Build multi-currency, multi-fiscal-year tracking from day one.
- Treat local partner capacity-building as core program design, not overhead.
- Practice visible change management with funders — it is a funding advantage, not a risk to be hidden.