By Winston Osuchukwu | founder & CEO, Mathesis Analytics
The average Nigerian borrower is widely considered high-risk – a claim repeated in credit committees, priced into retail loans, and largely treated as settled fact. Every credit market accepts that an individual loan may not be repaid; this is ordinary, priced risk. The high-risk claim, however, is applied to whole segments – the informal trader, the gig economy earner whose income is steady but split across several accounts, the remote worker paid by an overseas client into a fintech FX wallet. What the assessment establishes is not whether they are likely to repay, but how they fit into an arbitrary segment. Having spent years building decisioning systems for this market, my thesis is a specific one: “high-risk” does not mean “no credit” – it simply requires that the lender embrace alternative datasets to price the risk appropriately.
This is not a criticism of the institutions that built their frameworks around collateral and documentation; those were rational responses to the tools available at the time. When data is scarce, prudence means defaulting to the status quo. The limitation is not that this approach is wrong, but that it leaves a blind spot – excluding fundamentally sound borrowers whose economic lives simply are not captured on the bank’s ledger. A market trader who has moved consistent, growing volumes of cash through mobile money for three years is not, in any meaningful sense, unknowable. Their financial behaviour is observable and patterned; it simply occurs outside the traditional banking system, rendering it invisible to conventional underwriting.
This is the gap technology is now positioned to close – not by replacing institutional judgment, but by augmenting it. When AI-driven analysis is applied rigorously to the financial behaviour these borrowers generate, a far more complete picture of their repayment ability emerges – and a meaningful share presents a risk profile that compares favourably with segments the traditional system has long considered safe. The “high-risk” label, applied broadly to an entire category of borrower, was never a risk pricing tool so much as the limit of what the available tools could see.
For banks, this is the opportunity to extend capital with confidence beyond the borrowers who fit their stringent criteria. Nigerian banks are highly liquid; the constraint on credit growth has rarely been capital, but the ability to assess and price the borrowers who sit outside the traditional file. Close that gap, and the whole ecosystem strengthens: banks grow their loan books into segments they have long wanted to serve, and the real economy gets the capital it needs to expand.
This is precisely what we focus on at Mathesis Analytics: building AI-powered credit decisioning that gives lenders a fuller, more defensible picture of the individuals long excluded as high-risk when they were simply misjudged. The Nigerian credit gap has never been a non lendable-population problem, but one of incomplete visibility. By unifying varied data sources and partnering with the institutions that hold the capital and scale to move the market, we translate out-of-ecosystem behaviour into reliable, bank-grade risk scores. Closing this gap is one of the clearest, highest-leverage opportunities in Nigerian financial services today.
