Scaleup Service Practices

The supply side of scaling support has a labelling problem. Many programmes use scaling terminology while delivering standardised incubation. The distinction matters because the interventions are different in kind: early-stage incubation addresses product-market fit, business model validation, and founder capability. Scale-up support addresses organisational design, management professionalisation, governance systems, growth capital architecture, and the operational challenges that arise when a venture with proven product-market fit attempts to replicate its model at speed.

Conflating the two wastes founder time, misallocates scarce expert capacity, and produces the cycle of training fatigue: ventures that have cycled through multiple generic programmes without receiving the support that their actual stage requires. The substantive treatment of the structural reasons this misallocation persists - donor architecture incentives, programme-rich/capability-thin equilibrium, role collapse - sits in (Stalled) Acceleration, Feedback Loops, and The Political Economy of the Ecosystem.

“We're scarce on people who are supporting startups at that scaling stage with the right balance of customisation, support services, and that balance between strategy versus on hands on execution.” - interviewee

"They don't need someone to say, here's a recipe that you have no money to execute on. And here's a bunch of consultants that you can't afford to pay, who've never done this before in a real market. That's what's usually happening."* - interviewee

The demand side has shifted materially since 2022. Ventures that survived the correction period are more sophisticated consumers of support - more likely to have experienced the cost of generic provision and less willing to accept it again. The market for genuinely expert, contextually grounded, execution-oriented support is larger and better-informed than it was. The supply of that support has not kept pace.

What the evidence says about effective delivery

The empirical evidence on what works in scale-up support is now substantial. The Argidius Foundation's SCALE framework synthesising nearly a decade of BDS evidence, the GALI ten-year synthesis covering 23,364 enterprises across 369 programmes, and Spring Impact's analysis of scaling delivery patterns are treated substantively in Ecosystem Characteristics, (Stalled) Acceleration, and 15 Acceleration Goals.

The convergent finding across the three evidence bases: the ventures that benefit most from acceleration are those that enter programmes with stronger teams and more developed businesses; programmes most likely to add value are those with intensive, tailored, on-demand support rather than fixed curricula; programmes that do not screen rigorously end up providing expensive support to ventures that are not ready for it, with negligible performance effects; and the support typically ends at the point where sustained engagement is most needed - during implementation. These findings have been available for years. Their adoption across the African support ecosystem has been slow.

The provider landscape has changed materially since 2022. Two changes are most consequential.

The most institutionally sophisticated scaling support offering on the continent prior to 2024 was Founders Factory Africa - milestone-governed, monthly progress-tracked, with portfolio engagement extending beyond programme completion. Its rebrand to 54 Collective and subsequent liquidation in 2025 was the institutional failure made visible at scale. The substantive treatment of the venture-studio funding-architecture lesson - and the distinction between funding-model collapse and support-model failure - sits in What Working Support Infrastructure Looks Like.

Endeavor remains the most evidence-backed network-anchored, high-intensity scaling support model in operation on the continent. The substantive treatment of Endeavor's brokerage-network architecture and the Multiplier Effect mechanism that produces its measurable outcomes also sits in What Working Support Infrastructure Looks Like. The accessibility constraint - support primarily available within Endeavor's selected portfolio, and selection deliberately rigorous - remains a limitation on ecosystem-wide impact.

As far as we are aware, no government in Africa provides dedicated scale-up support funded at the level and quality of international models such as Tech Nation in the UK or Scale-up Holland in the Netherlands. This is among the most consequential structural gaps in the ecosystem.

The most notable new entrant at the early-stage end of the support spectrum is Accelerate Africa - launched in 2024 by Iyinoluwa Aboyeji and Mia von Koschitzky-Kimani. Designed to fill the gap created by Y Combinator's withdrawal from Africa - which backed no African startups in its most recent summer batch and only three in each of the three preceding batches - it ran eight (now 12)-week, in-person cohorts of ten startups across Lagos and Nairobi, with optional investment of $250,000–$500,000 from Future Africa available post-programme subject to diligence. Importantly, it is an early-stage acceleration programme, not a scale-up programme. Its value is in the contextual grounding and the founders who lead it. Its gap is the same gap that existed before it launched: there is no commercially structured, expert-delivered, sustained support offering for ventures at the scale-up stage in most African markets.

The most significant development in scaling support provision since 2022 is not a new programme or institution. It is the emergence of AI tools that materially change what scaling support can be delivered, at what cost, and to how many ventures simultaneously. Support that previously required expensive human capital to deliver at reasonable quality can now be partially automated - enabling the scarce pool of genuinely expert practitioners to focus on the highest-value interventions while AI handles knowledge-diffusion, diagnostic, and curriculum functions at scale. The substantive treatment of AI in ecosystem capability, and in the knowledge-commons architecture more broadly, sits in What AI changes about African scaling and Data, Insights and Knowledge.

The shared-service and affordable online learning models long predicted as the structural solution to the cost problem are now architecturally achievable in ways they were not in 2022. AI-powered diagnostic tools can assess a venture's operational state - across financial management, governance, people systems, customer data practices, and market positioning - at a fraction of the cost of human assessment. AI-powered learning platforms can deliver genuinely personalised curriculum adapted to each venture's specific gaps, sector, and market context. The institutional design question - who builds and maintains these platforms, how they are funded, and how quality is assured - remains to be answered. The 54 Collective collapse makes this question more urgent: the dominant funding model for scaling support - large donor grants to institutionalised intermediaries - has demonstrated its fragility in a way the ecosystem cannot ignore.

The economics literature has named precisely what the willingness-to-pay debate is about. Akerlof's foundational Quarterly Journal of Economics paper “The Market for Lemons" established the analytical framework: in markets with information asymmetry between buyers and sellers about product quality, the market unravels - the rational response to uncertainty about quality is to refuse to pay for it, which drives high-quality providers out and leaves the market populated by low-quality offerings. Spence's Quarterly Journal of Economics paper on market signalling identified the resolution: where direct quality measurement is impossible, costly signals - credentials, fees, demonstrated track records - function as quality screens that buyers rationally treat as informative.

The scaling-support market in Africa is structurally a lemons market. Ventures cannot evaluate scaling-support quality before consumption. The default response - refusing to pay - has produced exactly the unravelling Akerlof predicted: high-quality providers find the market commercially unviable, exit or never enter, and the market is dominated by low-quality, often donor-subsidised offerings whose persistence depends on donor relationships rather than venture demand.

Darby and Karni's Journal of Law and Economics paper on credence goods extends the analysis. Credence goods are those whose quality cannot be evaluated even after consumption - medical care, legal advice, expert consulting. The scaling-support market is structurally a credence-goods market: the venture cannot definitively determine whether the support it received caused the subsequent outcome, or whether the outcome would have occurred regardless. Resolving the credence-goods problem requires institutional infrastructure: documented outcome data, third-party evaluation, professional standards, reputation mechanisms that operate independently of the provider's own marketing. None of this exists at scale in the African scale-up support market.

Charging is therefore not just a revenue mechanism. It is a quality signal in a Spence sense and a commitment device in a credence-goods sense. The Argidius SCALE framework finding that charging improves both selection and outcomes is the empirical confirmation of these mechanisms at the BDS-provision level.

"Founders coming into the system need to understand that to succeed, they really need quality support. And this support costs money. There is a justifiable business case for it, but this narrative has not been adequately articulated." - interviewee

"At the end of the day, you get what you pay for. If something is free, you've got to expect a certain level of service." - interviewee

The causal story - expert support contributes to operational resilience, which contributes to survival through capital contractions - is becoming more legible as the natural experiment of the correction period unfolds. Ventures that received the right support at the right stage, and paid for some of it, are disproportionately in the cohort that survived. The collapse of donor-funded support institutions adds the structural argument: commercial sustainability of support provision is not a preference. It is the only architecture that ensures the support is there when a venture needs it, rather than when a donor relationship permits it.

"There is a recognition of the need, but there isn't the conviction, and therefore willingness to pay. Even private equity companies which have value creation arms haven't been fully able to put together strong business cases." - interviewee

The gap between recognition and conviction is closing. It has not closed. Closing it - through better evidence, better articulation of the return on support investment, and better designed commercially structured provision - is among the most important near-term tasks for the support ecosystem. The Akerlof-Spence-Darby-Karni framework names what would actually close the gap: documented outcomes that function as credible signals, institutional infrastructure that resolves the credence-goods problem, and pricing that filters for both venture readiness and provider quality simultaneously. The scaling-support market will become commercially sustainable when these conditions are met, and not before.