The ESO Funding Crisis in Africa: After the Aid Collapse

The February 2025 dismantling of USAID eliminated roughly 90% of American foreign-aid contracts and triggered the deepest funding crisis the African entrepreneur-support sector has ever faced: an estimated $100 million startup-funding shortfall in Kenya alone, with quiet wind-downs of accelerators and venture builders across the region (1, 2, 3). But the crisis is not fundamentally about the money leaving. It is about what the money concealed: a sector whose unit economics never worked, sustained by grant cycles rather than by the value it created. The organizations that survive this reckoning will be the ones that stop behaving like projects and start behaving like businesses: earning from entrepreneurs, contracting with governments, and getting paid for outcomes.

Our answer to the funding crisis is a self-sustaining model; see how the AVODA Institute is built.

Key Takeaways

• The 2025 USAID dismantlement cut ~90% of American foreign-aid contracts; Kenya faces an estimated $100 million startup-funding shortfall, and preliminary estimates project its startup economy could shrink 15% within three years (1, 2).
• The fragility predates the cuts: over 40% of African innovation hubs cited funding as their biggest challenge before USAID collapsed, and research finds 40-75% of typical ESO funding comes from governments, donors, or foundations (4, 5).
• The evidence problem compounds the funding problem: GALI data on 23,000+ ventures shows acceleration works on average, but a 2025 NBER working paper finds most accelerators add negative value relative to no accelerator, with a small right tail of excellent programs generating the gains (6, 7).
• An October 2025 study of East African ESOs, built on 35 confidential interviews, documents 25 working solutions: service-hub models, equity and success-fee compensation, alumni economics, and outcome-based funding among them (8).
• Governments are now the largest remaining buyer of enterprise support in East Africa; World Bank and OECD evidence says they get better results funding independent operators against verified outcomes than running programs themselves (9, 10).
• Uganda’s NSSF Hi-Innovator program (438 businesses backed, 14,000 new pension contributors generated) shows what a domestic, institutionally anchored funding model looks like when the incentive loop closes (11).

What Actually Happened to ESO Funding?

For two decades, the entrepreneur support organization (the accelerator, the incubator, the innovation hub, the business development service provider) was one of development finance’s favorite instruments. Donors funded cohorts; ESOs ran them; reports counted heads trained and grants disbursed. By 2024, AfriLabs counted over 500 innovation hubs across the continent (4). Then the largest single funder of the system was switched off in a matter of weeks. USAID’s dismantlement in February 2025 eliminated roughly 90% of its contracts, part of a broader Western aid retreat that hit climate, health, and agriculture programming hardest, precisely the sectors where blended finance and grant-linked capital had played the largest role in African innovation ecosystems (1, 3).

The transmission into the entrepreneur-support sector was fast and largely unreported, because ESOs rarely publish their funding structures. Kenya’s estimated $100 million startup-funding shortfall and projected 15% three-year contraction made headlines (2); the quieter story was organizational: accelerators and venture builders across the region wound down operations as US-backed programs were withdrawn, froze hiring, cut cohorts, or pivoted overnight toward whatever donor calls remained (3). Uganda’s hub ecosystem, among the most donor-dependent on the continent, entered what can only be called a sorting event: a separation of ESOs with real business models from those that were donor projects wearing an accelerator costume.

It is worth saying clearly: real harm followed. Capable teams disbanded; entrepreneurs mid-program were stranded; years of accumulated relationships and local knowledge evaporated. Nothing in the analysis that follows minimizes that. But a sector-wide near-death experience deserves a root-cause diagnosis, not just sympathy, and the root cause was not in Washington.

Why Were ESOs So Exposed in the First Place?

Because the sector’s revenue model violated the first rule of institutional design: an organization should be funded, at least substantially, by whoever receives its value. The typical African ESO drew 40-75% of its funding from governments, donors, and foundations (5), and in much of East Africa the share ran higher, while its actual beneficiaries, the entrepreneurs, paid little or nothing, and its other beneficiaries (investors who received screened deal flow, corporates who received innovation pipelines, governments that received jobs and tax revenue) paid nothing at all. Over 40% of African hubs named funding their biggest challenge while the aid was still flowing (4). An institution whose survival depends on a foreign government’s election cycle was never an institution. It was a project with a logo.

Donor dependence did not just make ESOs fragile. It made them worse at their jobs, through three mechanisms. First, it selected the wrong customer: programs were designed to satisfy donor reporting templates (beneficiaries trained, workshops held, gender ratios achieved) rather than venture outcomes, because that is what renewals depended on. Second, it suppressed price discovery: when the service is free, nobody learns what entrepreneurs would actually pay for, which is the single most informative datum any service business possesses. Third, it crowded the field with supply that demand never validated: 500+ hubs is a number that reflects the availability of grants, not the demand for acceleration (4).

The deeper measurement culture made this invisible. As I have argued in examining whether startup acceleration works at all, the global evidence is double-edged: GALI’s decade of data across 23,000+ ventures shows accelerated companies outperform rejected peers on revenue, employment, and capital raised (6), but the 2025 NBER “Beyond Demo Day” analysis finds that most accelerators have negative value-added against a no-accelerator benchmark, with a small right tail of excellent programs producing the average gains (7). Translate that into the funding crisis: donors were paying list price for a sector in which the median product may have been worth less than nothing, and almost no program could prove which side of the distribution it was on. When the money left, there was no outcomes evidence to argue for keeping it.

Is This a Funding Crisis or a Business-Model Crisis?

It is a business-model crisis that a funding crisis made undeniable, and that distinction dictates the response. If the problem were merely lost funding, the answer would be replacement funding: new donors, new grants, better proposal writing. Much of the sector is attempting exactly this, and for most it will fail, because it rebuilds the same structure on a different patron. The European and multilateral funders now being courted face their own fiscal pressures, and the localization rhetoric sweeping the aid industry points the same direction as this article: fund institutions that can eventually fund themselves (12).

The business-model answer starts from a different question: who receives value from a good ESO, and what would each beneficiary rationally pay? Worked through, that question yields three customers, and a test.

The Three-Customer Test: A Framework for ESO Sustainability

An ESO is structurally sustainable when it can pass what I call the Three-Customer Test: it earns meaningful revenue from at least two of the three parties who demonstrably receive its value, and could survive the loss of any single one. The three customers:

Customer 1: Entrepreneurs (earned revenue). If ventures will not pay anything (fees, success-based commissions, revenue shares, equity or quasi-equity in markets where those can be redeemed) that is data about the program’s value, not about African ability to pay. The practical menu is wider than tuition: paid advisory for growth-stage alumni, service-hub models in which the ESO convenes vetted providers and takes platform fees, shared back-office services, and alumni arrangements in which successful graduates mentor or buy services from the institution that built them (8). The October 2025 East Africa ESO study found versions of all of these already piloted in the region (8). Pricing entrepreneurs at zero forever is not solidarity. It is a confession that the product has no measurable worth.

Customer 2: Governments and institutional buyers (contracts). With donors receding, government is the largest remaining buyer of enterprise support in East Africa, through programs like Uganda’s JICA-partnered NINJA accelerator and the NSSF Hi-Innovator scheme (13, 11). The evidence on how this should work is consistent: World Bank/infoDev and OECD reviews both find governments achieve more by competitively contracting independent operators against defined outcomes than by operating programs themselves (9, 10). For ESOs, this is a legitimate, durable revenue line, if they can win tenders on verified results rather than relationships. The full procurement argument (what governments should demand, and how operators should be paid) is the subject of how governments should buy acceleration.

Customer 3: Capital and corporates (success economics). Investors receive screened, prepared deal flow from good ESOs; corporates receive innovation pipelines, supplier development, and distribution partnerships. Both can pay: success fees on capital raised by portfolio ventures, retained scouting and diligence services, corporate-sponsored vertical programs with commercial (not CSR) budgets, and carried interest or equity stakes where local exit markets can actually redeem them. Hi-Innovator is the most instructive East African case because the incentive loop closes inside one balance sheet: Uganda’s national pension fund backs small businesses, the businesses formalize and hire, and the program has already generated 14,000 new pension contributors, so the funder is buying its own future customers (11).

Donors, note, are not on the list, by design. In the post-2025 world, grant funding should be treated the way a disciplined company treats venture capital: an instrument for building capabilities and proving models, never an operating revenue line. This is also where donors themselves must change, as ANDE’s “Sustain Impact” work argues: fund ESO institutions and their transitions to earned revenue (multi-year, flexible, capacity-building money) rather than perpetual project cycles that recreate dependence (12).

The test is deliberately demanding. Most ESOs today earn meaningful revenue from zero of the three customers. Passing it typically takes years, which is exactly why the work must start during the crisis, not after it.

How Do ESOs Earn Revenue Without Corrupting the Mission?

The standard objection deserves a direct answer: won’t charging entrepreneurs, chasing contracts, and taking success fees distort an ESO toward serving whoever pays? Three responses. First, the status quo already distorts, toward donors, whose templates shaped programs far more than any entrepreneur’s needs ever did. Earned revenue does not introduce incentive distortion; it redirects the distortion toward the people the mission claims to serve. Second, instrument choice manages the tension: success fees and revenue shares align the ESO with venture outcomes; flat fees for premium services preserve free or cheap access at entry level; cross-subsidy (corporate and government contracts funding early-stage open programs) is how universities and hospitals have managed identical tensions for centuries. Third, selection and values do real work here: programs built around a coherent founder community and explicit values, the case examined in values-based accelerator design, sustain the trust that makes alumni economics and word-of-mouth recruitment function, which is itself a commercial asset no grant can buy.

What does corrupt the mission is institutional death. An ESO that closes serves no one. The moral hierarchy is not purity versus commerce; it is existence versus absence.

For the director reading this mid-crisis, the ninety-day version is concrete. Price one service this quarter (a paid diligence package for investors, a premium advisory tier for alumni, a corporate scouting retainer) and learn what the market says. Build the outcomes file: two-year survival, revenue growth, and capital raised for every graduating cohort you can reconstruct, because every customer worth having will ask for it. Map the government and institutional tenders in your market for the next eighteen months and decide now which two you intend to win. And renegotiate your remaining grants toward capability building (measurement systems, earned-revenue pilots, reserves) rather than another undifferentiated cohort. None of this requires permission. All of it compounds.

What Will the Sector Look Like in 2030?

Smaller, and better. The honest projection is consolidation: fewer than half of today’s hubs likely survive in recognizable form, and that is an acceptable outcome if the survivors are the right tail the NBER data describes (7). Expect four convergent features. First, hybrid revenue stacks as the norm: a typical sustainable East African ESO in 2030 might run 30% government and institutional contracts, 25% corporate programs, 20% earned services and fees, 15% success-based economics, 10% philanthropic capability grants. Second, radical measurement: survivors will publish survival rates, revenue growth, and follow-on capital for their portfolios, because outcome-paying customers demand it, so the reporting culture donors never required becomes the price of admission to government tenders. Third, specialization by sector, founder type, or values community, because depth is what each of the three customers actually pays for. Fourth, regional consolidation and shared infrastructure (back offices, measurement systems, even merged institutions) as the WDI study’s systems-level solutions anticipate (8).

The sector panic of 2025-26 will, in retrospect, mark the moment African entrepreneur support stopped being a donor program category and started becoming an industry. Industries have customers, prices, measurable products, and competitive survival. The ESOs that internalize this fastest will not merely survive the aid collapse. They will inherit the field it cleared, and they will serve entrepreneurs better than the subsidized version ever did. That is not optimism as consolation. It is the standard pattern by which subsidized sectors, everywhere, finally grow up.

Frequently Asked Questions

What is an ESO?
An entrepreneur support organization: any institution (accelerator, incubator, innovation hub, or business development service provider) whose core product is helping ventures start, survive, and grow. Africa had over 500 innovation hubs by 2024, most funded primarily by donors, governments, or foundations rather than by the value they create (4, 5).

How badly did the USAID collapse hurt African entrepreneur support?
Severely. The February 2025 dismantlement cut roughly 90% of American aid contracts; Kenya alone faces an estimated $100 million startup-funding shortfall with a projected 15% startup-economy contraction over three years, and accelerators and venture builders across the region wound down as US-backed programs were withdrawn (1, 2, 3).

Do accelerators actually work?
On average, yes. GALI data across 23,000+ ventures shows accelerated companies outperform rejected peers. But a 2025 NBER paper finds most individual programs add negative value, with a small excellent minority generating the average gains. The question for any ESO is which side of that distribution it occupies (6, 7).

How can an African ESO become financially sustainable?
By earning from the parties who receive its value: entrepreneurs (fees, revenue shares, success commissions, alumni services), governments (outcome-based contracts won through competitive tender), and capital or corporates (success fees, scouting retainers, sponsored programs). Grants should fund capability building, never permanent operations (8, 9, 12).

Should governments run their own accelerators?
The evidence says no. World Bank and OECD reviews find governments achieve better results competitively contracting independent operators against verified outcomes (survival, revenue, jobs at 24 months) than operating programs themselves. The government’s highest-value role is demanding customer, not operator (9, 10).

Related Reading

• How governments should buy acceleration: a procurement manifesto
• Does startup acceleration work? What the GALI and NBER evidence actually says
• Values-based accelerator design: the industry’s most underrated choice
• The post-accelerator valley of death in East Africa

Sources and Evidence

1. Oxfam America, 2025. “What did USAID do and what are the effects of USAID cuts?” https://www.oxfamamerica.org/explore/issues/making-foreign-aid-work/what-do-trumps-proposed-foreign-aid-cuts-mean/. Major international NGO’s documentation of the scale of contract terminations (~90%).
2. Michigan Journal of Economics, 2025. “Op-Ed: The Devastating Impacts of the USAID Pullout on Africa.” https://sites.lsa.umich.edu/mje/2025/05/13/op-ed-the-devastating-impacts-of-the-usaid-pullout-on-africa/. University economics publication; source for Kenya’s $100M shortfall and 15% contraction estimates (clearly labeled preliminary).
3. Launch Base Africa, 2025. “Revisiting the 2025 Predictions for African Tech.” https://launchbaseafrica.com/2025/12/19/revisiting-the-2025-predictions-for-african-tech/. African tech analysis outlet documenting accelerator and venture-builder wind-downs following US program withdrawals.
4. AfriLabs, 2024. 2024 Impact Report: Over 500 Innovation Hubs. https://www.afrilabs.com/afrilabs-releases-2024-impact-report-over-500-innovation-hubs-280000-lives-reached-and-a-1-trillion-vision-for-africas-digital-future/. The continent’s largest hub network; source for hub counts and the 40%+ funding-challenge finding.
5. Ducker, M. (Ecosystem Mike), 2024. “4 questions for new Entrepreneurship Support Organizations (ESOs) on financial sustainability.” https://www.ecosystemmike.com/post/4-questions-for-new-entrepreneurship-support-organizations-esos-on-financial-sustainability. Practitioner analysis; source for the 40-75% donor-funding share and the finding that seed-investment gains cannot sustain most ESOs.
6. Global Accelerator Learning Initiative (GALI / ANDE / Emory University). “Does Acceleration Work?” https://andeglobal.org/publication/does-acceleration-work/. The largest longitudinal dataset on acceleration outcomes (23,000+ ventures); methodologically rigorous with comparison groups.
7. National Bureau of Economic Research, 2025. “Beyond Demo Day” (Working Paper 35063). https://www.nber.org/papers/w35063. Peer-reviewed-track academic analysis finding negative median accelerator value-added with a high-performing right tail.
8. William Davidson Institute, University of Michigan, October 2025. “Reimagining the Future of Enterprise Support Organizations in East Africa.” https://wdi.umich.edu/wp-content/uploads/WDI-Reimagining-the-Future-of-ESOs-in-East-Africa_Final-Report_HRes.pdf. 35 confidential interviews with ESOs, investors, entrepreneurs and donors; documents 25 solutions including service-hub, equity-compensation, alumni, and outcome-based models.
9. World Bank / infoDev. “Business Incubation: Lessons for Policymakers.” https://documents1.worldbank.org/curated/en/981161468331855750/pdf/700230ESW0P11100Business0Incubation.pdf. Foundational multilateral evidence on public funding of incubation through third-party operators.
10. OECD, 2024. “Start-up globalisation through incubation and acceleration.” https://www.oecd.org/content/dam/oecd/en/about/projects/cfe/incubation-and-acceleration/Start-up-globalisation-through-incubation-and-acceleration.pdf. OECD review of incubation/acceleration policy; supports outcome-contracting over state operation.
11. ImpactAlpha, 2025. “Pension fund in Uganda readies a $100 million fund-of-funds to create jobs, and savers.” https://impactalpha.com/pension-fund-in-uganda-readies-a-100-million-fund-of-funds-to-create-jobs-and-savers/. Impact-investment publication; source for Hi-Innovator’s 438 businesses and 14,000 new pension members.
12. ANDE, 2024. “Sustain Impact: Donor practices to grow enterprise support organizations.” https://andeglobal.org/publication/sustain-impact-donor-practices-to-grow-enterprise-support-organizations/. Sector body’s analysis of how donors should fund ESO institutions rather than project cycles.
13. Ministry of Trade, Industry and Cooperatives (Uganda). “Launch of JICA NINJA Acceleration Program for Growth Stage in Uganda.” https://www.mtic.go.ug/launch-of-jica-ninja-acceleration-program-for-growth-stage-in-uganda/. Official government source on bilateral acceleration procurement.