Zoom Out · October 2026 · 5 min read

What a Market Systems Diagnostic Should Tell a Donor Before Design Starts

I have worked on three donor programs that set up Business Development Services units inside host institutions. Some units did not outlast the programs. In each case the diagnostic came after the design, so it was asked to confirm the plan, not to test it.

The terms of reference arrive with two attachments. One asks for an assessment of MSME needs in a sector. The other is the list of activities the program will deliver.

I have read that pairing, in different words and different countries, more often than I can count. The diagnostic is asked to find a need. The design is already waiting for it.

A diagnostic that arrives after the design is not a diagnostic. It is a justification with a methodology section.

Three Programs, One Pattern

I have worked on three major donor-funded programs that supported MSMEs by establishing Business Development Services (BDS) units. Millions of dollars went into them. Most of the units sat inside larger host organizations: microfinance institutions, and NGOs with an economic empowerment mission among them.

The design looked at the host. It saw an institution with clients, a mission and a willingness to give the unit a home. It did not see the market barriers around the unit: why these services were not already on offer, who would pay for them, and what would carry the cost once the program stopped. Some of the units did not continue. I wrote about the ones that did in Why Some Business Development Services Units Survive Their Projects.

Zoom out and the picture changes. A unit that closes is the event. The pattern is that a unit built around a willing host can meet every program target and still not outlast the program. The structure behind it is a design agreed before the diagnostic, so the diagnostic was asked to confirm a delivery model, not to test it. The mental model is that a host with the right mission is a ready-made home, and that the market will follow the unit. It does not follow on its own.

A diagnostic carried out before design would not have asked whether the hosts were willing. They were. It would have asked whether a unit would still make institutional and commercial sense to each host in year three, and what the program would have to change in the market, not only in the host, for that to be true. Sometimes the answer is a different host or a different model. Sometimes it is a smaller unit with a price on the service from the first day. Either is far easier to change on paper than after the unit is staffed.

Ask Why the Service Is Missing

“Do MSMEs need this service?” is a question every survey answers with yes. Nobody tells an interviewer they have no use for finance, skills or market access.

The question that moves a design is a different one: why is the service not already provided, and who could provide it? The answer sits with the supplier, the regulator, the buyer, the norm and the incentive, not with the enterprise. A diagnostic that stops at the enterprise has described the symptom in great detail.

For a unit inside a host there is a second question: why would the host keep it? A microfinance institution answers to its portfolio and its regulator. An NGO answers to its donors and its mission. A BDS unit competes inside that organization for management attention, staff and cross-subsidy. Whether it wins depends on the host’s incentives, the price the client will pay and the rules around both. None of that appears in a survey of enterprise needs.

Three Tests Before Design Starts

The DCED guidance on assessing systemic change describes change that lasts as sustainable, scalable and resilient. I use the same three words as a test on the proposed intervention before a single activity is written.

Sustainable: who keeps doing this once the program stops, and what do they gain from it?

Scalable: can it reach beyond the first group without the program in the room?

Resilient: what happens when a price, a regulation or a key supplier changes?

If the diagnostic cannot answer all three for the intervention on the table, the design is premature.

The test also keeps the claim honest. Marcus Jenal’s 2019 stock-taking for USAID’s MESCLA activity found four ways programs talk about systemic change: the spread of an innovation, a change in structure, a change in state and a change in trajectory. The study found little alignment between them. A program that promises structural change and monitors the number of units established has promised one thing and measured another. The diagnostic is where that choice is made, in writing, before the donor is held to it.

What the Donor Should Hold Before Design

  1. A map of who does and who pays for each supporting function, and who could.
  2. The binding constraint, named, with the evidence that it binds.
  3. The three tests applied to the first intervention, including the host’s own reason to keep a unit with and without program money.
  4. The claim of systemic change the program will be held to, and how it will be tracked.
  5. The assumption most likely to be wrong, and the signal that will say so early.

CGAP noted in 2016 that experience with market systems approaches in financial inclusion was still rare compared with agriculture and enterprise development. That is a reason for more diagnostic discipline in finance, not less.

A diagnostic earns its fee on the day it changes the design. If it has not, ask what it was for.

Work With Me on This

I lead market systems diagnostics for donors and implementers across MENA, built to change the design before it is written.

See the Service →

More pieces like this are on the Perspectives page.

Further Reading

← All Perspectives

Organizations I Have Worked With
USAIDGlobal Affairs CanadaWorld BankCentral Bank of EgyptCommercial International BankIFCKfWEuropean UnionHabitat for Humanity InternationalAga Khan Agency for MicrofinanceDrosos FoundationUN WomenUN-HabitatLFS AdvisoryGoss GilroyEnvironment and Development GroupAECOMFINBI