Every serious investor in African fintech has read the headline numbers. 57% of adults unbanked. Mobile money transaction volumes growing at 25% CAGR. A $65bn revenue opportunity by 2030. The problem is not that these numbers are wrong. The problem is that everyone with a fund deck has the same numbers, drawn from the same three sources, framing the same surface-level thesis.

What those decks do not model is what we call the Infrastructure Dependency Trap — the structural condition in which African fintech growth is not independently driven but is instead a derivative of telco infrastructure decisions made in Johannesburg, Nairobi and Lagos boardrooms that have nothing to do with fintech.

In markets where USSD remains the primary fintech delivery channel — which includes Ghana, Tanzania, Uganda, Zambia and most of francophone West Africa — the cost of every fintech transaction is a telco toll. USSD session charges are not regulated uniformly. Telcos treat them as margin lines.

The structural issue is this: when MTN Nigeria, Safaricom Kenya or Airtel Africa renegotiates its USSD pricing or shifts strategic priority toward its own mobile money subsidiary, every third-party fintech built on that channel absorbs the impact immediately. This is not a risk that appears on standard due diligence checklists because it is not a credit risk or a regulatory risk. It is an infrastructure sovereignty risk.

Agent banking and agent-assisted cash-in/cash-out networks are routinely presented in pitch decks as distribution moats. The data says otherwise in most markets.

Agent churn in Nigeria runs between 35 and 50 percent annually at the non-exclusive tier. This means that a fintech reporting 80,000 active agents in Q1 may have rotated a third of that network by Q4 while reporting nominally flat headline figures. The underlying economics make exclusivity uneconomical for most fintechs below Series B: agents must be incentivised to stay, and the commission structures needed to ensure loyalty compress margins precisely when investor models demand margin expansion.

The fintechs that have genuinely solved this — M-Pesa in Kenya being the canonical example — did so by becoming infrastructure themselves, not by building on top of infrastructure someone else owns. That distinction is the entire game.

The infrastructure dependency trap is manageable, but only if investors model it explicitly rather than assuming it away. Three variables that belong in every African fintech financial model:

USSD cost sensitivity: What does a 30% increase in USSD session cost do to unit economics? If the answer is 'material' and management has no contractual protection, that is a risk line item not a footnote.

Agent network effective reach: Not the headline agent count but the number of agents who have processed a transaction in the past 60 days, segmented by geography. This is the real distribution number.

Infrastructure stack ownership ratio: What percentage of the fintech's critical infrastructure does it own versus lease or access through a third party with its own competing product? This ratio is a proxy for strategic durability.

A significant portion of African fintech value creation in the 2015 to 2022 period was, in plain terms, regulatory arbitrage. Fintechs moved faster than regulators. Products launched in grey zones. Licensing frameworks did not exist for the products being sold. This created real returns for early investors. It is also a model that is structurally expiring across the continent's four most investable fintech markets.

The Central Bank of Nigeria's posture toward fintech has undergone a genuine structural shift that most investor narratives still have not caught up with. The 2021 PSB licensing framework, the revised PSP guidelines, and the 2023/2024 FX restriction episode collectively signal that the CBN has moved from a tolerance model to an architecture model.

Under a tolerance model, regulators allow innovation to happen and clean up later. Under an architecture model, regulators define the permissible shape of innovation before it happens. The practical implication for investors: the grey-zone premium that early Nigerian fintech valuations were partly built on is no longer available in the same form. What is available instead is a regulatory compliance moat — for the fintechs that have built robust compliance infrastructure, the new architecture creates barriers to entry for new challengers that older vintage thinking often misses.

Kenya and Tanzania are running opposite regulatory experiments in real time, and the outcomes have direct implications for capital allocation across East Africa.

Kenya's CBK has moved toward proportionate regulation — a framework where the regulatory burden scales with the systemic risk profile of the licensed entity. This has enabled a more dynamic mid-market fintech sector. Tanzania's Bank of Tanzania has taken a more restrictive stance, maintaining tight capital requirements and a narrow licensing typology. The result is a market where well-capitalised players have entrenched positions but smaller innovators face a genuine capital barrier to entry.

For investors, this divergence means that East Africa is not one investment thesis. A fintech strategy built on Kenyan market logic will not translate directly to Tanzania without a fundamental regulatory repositioning. Deals that underwrite pan-East-African growth on a single regulatory assumption are mispriced.

WAEMU — the West African Economic and Monetary Union covering Senegal, Ivory Coast, Mali, Burkina Faso, and four other markets — operates under BCEAO monetary authority. Fintech licensing in the WAEMU zone has historically been accessible in a way that attracted operators from anglophone markets looking for regional scale.

What is changing: the BCEAO published revised EME (Electronic Money Institution) regulations in 2022 that significantly tightened capital and operational requirements. The transition period ends in late 2026. Fintechs operating in the WAEMU zone on pre-2022 authorisations face a compliance cliff. Due diligence on any WAEMU-exposed fintech must include an explicit assessment of whether their current authorisation survives the transition, and at what capital cost.

African fintech pitch decks have become sophisticated at presenting unit economics in ways that look compelling at the headline level but contain structural compression pressure that only appears at scale. This section identifies the three most common unit economics presentation patterns that mask real margin risk.

Customer Acquisition Cost calculations in African fintech frequently exclude the cost of trust-building — the promotional credits, zero-fee windows, and high-interest promotional savings rates used to onboard customers in markets where brand trust must be earned through demonstrated benefit rather than assumed.

A fintech that onboards 500,000 customers in Year 1 at a reported CAC of $4.50 may have a true economic CAC of $11–18 once promotional costs, first-90-day subsidy activity, and the compliance cost of KYC at that scale are properly allocated. The gap between reported CAC and economic CAC is not fraud — it is a categorisation choice that is largely industry-standard. But it means that the LTV/CAC multiples being presented to investors are calculated on the reported number, not the economic number.

Many African fintechs generate a meaningful share of revenue from FX spread, remittance margin, or dollar-denominated service fees. In investor models, this revenue line is typically projected forward at or near current rates. This is a structural modelling error in markets with managed exchange rates.

The Nigerian experience of 2023/2024 is the stress test. Fintechs with USD revenue lines that were not hedged experienced 30–40% effective revenue compression on naira-reported financials when the CBN moved rates. Fintechs that had built product architecture assuming naira/dollar stability lost a compounding advantage that had, until that point, looked like a durable margin line.

The model that actually holds: treat FX-derived revenue as a separate revenue segment with explicit exchange rate assumptions, sensitivity tables at +/-20% and +/-40% FX movement, and a clearly articulated product response strategy for each scenario. Fintechs that cannot produce this analysis have not seriously stress-tested their own business.

Float income — the interest earned on customer deposits or transaction float — has become a meaningful revenue line for payment and wallet fintechs as African central banks raised base rates in response to post-COVID inflation. In Nigeria, the MPR rose to 24.75% by mid-2024. In Ghana, the policy rate peaked above 29%. In Kenya it reached 13%.

Float income earned at these rates is structurally transient. It is rate-cycle income, not business model income. Fintechs that grew into profitability partly on the back of float income at peak rates are running a different business than their 2026/2027 projections assume if those projections do not explicitly model rate normalisation.

The three preceding sections identify structural risks that investor models frequently underweight. This section turns the lens to the five structural multipliers — dynamics that tend to produce non-linear returns in African fintech — that standard market analysis equally underweights, in most cases because they are harder to quantify and less amenable to the kind of comparable-based valuation that investment committees prefer.

Formal economic data in Africa structurally undercounts the actual economy. Nigeria's GDP rebasing exercises have repeatedly revealed economic activity — informal trade, cross-border commerce, services — that was invisible to standard measurement. This informality is not simply a measurement problem to be corrected. It is a durable feature of African economic architecture, and the fintechs that have genuinely penetrated it — rather than serving the formal economy adjacent to it — have access to a demand pool that is both undercounted in the addressable market analysis and structurally resistant to the large-bank competition that threatens formal-market fintech moats.

The most interesting African fintechs are not the ones serving banked users better. They are the ones creating financial infrastructure for the informal sector that has never existed before. These are harder to model but structurally more durable.

Africa's commercial infrastructure — logistics platforms, agricultural marketplaces, e-commerce, ride-hailing, B2B trade networks — is in an early but accelerating phase of digitalisation. Each of these digital commerce platforms is a distribution surface for embedded financial products. The fintechs positioned to serve as the financial infrastructure layer for these platforms — lending at checkout, insurance at point of logistics, savings at point of receipt — have addressable market dynamics that compound with the growth of the host platforms.

This is qualitatively different from building a standalone fintech app and competing for consumer wallet share. Embedded finance grows with the commercial layer it sits inside. The investor question is not just 'how large is this fintech's market?' but 'how large is the market of the platforms this fintech is embedded in, and how does financial services penetration of those platforms compound over time?'

In high-trust-deficit markets — which describes the majority of African fintech markets at the customer acquisition stage — trust is the actual scarce resource, more than capital and more than technology. The fintechs that have built genuine trust — through transaction reliability, responsive dispute resolution, and transparent communication during platform outages or regulatory disruptions — have an asset that is genuinely difficult to replicate and that produces compounding returns in the form of referral acquisition and reduced churn.

The risk side of this dynamic: trust destruction is faster than trust construction. A significant platform outage, a fraud incident with poor communication management, or a regulatory action that freezes customer funds even temporarily can erase years of trust-building in days. Investors should ask management to quantify what a major trust event would cost in CAC terms — how many months of organic referral acquisition would be lost, and what paid acquisition budget would be required to replace it.

Consumer-facing African fintech gets the majority of investor attention and the majority of press coverage. The most structurally durable returns in African fintech are being built in the B2B infrastructure layer: payment APIs, core banking replacements, compliance infrastructure, identity verification, credit bureau infrastructure, and treasury management tools for the institutional layer.

B2B fintech infrastructure compounds differently from consumer fintech. Switching costs are real. Customer concentration risk is a genuine consideration, but so is the revenue durability that comes from being embedded in a client's operational stack. The African B2B fintech market is at an earlier maturity stage than consumer fintech, which means both that competitive dynamics are less intense and that the narrative has not yet been arbitraged into valuations.

Africa will add 840 million people of working age between 2020 and 2050. This is the most-cited macro variable in African investment theses and also the most mis-timed. The demographic dividend does not translate into fintech revenue in a linear projection from current curves. It translates into revenue through a sequence: working age population growth, income formalisation, digital access, financial services demand.

The step in this sequence that is most undermodelled is income formalisation — the structural shift from informal to formal employment income, which is the trigger for structured savings, insurance demand, and credit uptake at scale. The fintechs that are positioned to capture the income formalisation transition — particularly in Nigeria, Ethiopia, DRC, and Tanzania, where the demographic trajectory is steepest — are building toward a demand event that is demographically certain but temporally uncertain. The investment question is not whether it happens but whether the capitalisation runway is long enough to get there.

One of the most consistent errors in African fintech investment analysis is treating 'Africa' as a market. It is not. It is 54 regulatory jurisdictions, 8 major monetary unions, at least 6 structurally distinct fintech market archetypes, and an enormous range of infrastructure maturity levels. Capital allocated on a pan-African thesis without market-archetype differentiation is capital allocated with a model error built in.

The following twelve questions are structured specifically for pre-term-sheet due diligence. They are designed to surface the structural risks and multipliers identified in this report. Most of them will not appear in a standard VC due diligence checklist. That is the point.

African fintech is not over-invested. In absolute terms, it remains significantly undercapitalised relative to the size of the financial services opportunity. But the vintage of analysis being applied to most African fintech investment decisions has not kept pace with the sophistication the market now requires.

The infrastructure dependency trap, the regulatory arbitrage window closing, and the unit economics presentation conventions described in this report are not hypothetical risks. They are documented, occurring, and affecting capital returns in real portfolios. The investors generating consistent returns in this market are the ones who have built analysis frameworks that see these dynamics — and price them.

The five multipliers — informality premium, embedded finance surface area, trust velocity, B2B infrastructure undervaluation, and demographic dividend timing — equally represent genuine, structurally documented sources of non-linear return. They are not generically cited growth narratives. They are specific mechanisms with specific investment expressions that can be identified, underwritten, and actively managed.