The Founder and Chief Executive Officer of Mathesis Analytics Inc., Winston Osuchukwu, has urged Nigerian financial institutions to replace blanket interest rates with data-driven pricing models that reflect the specific risk profile of each borrower.
Osuchukwu said traditional lending systems often group customers with materially different financial behaviours into broad risk categories, resulting in identical interest rates for borrowers who do not present the same level of default risk.
He made the argument in an opinion article titled “Pricing Risk in the Dark: The End of Blanket Interest Rates in Retail Lending.”
Using the example of two small businesses seeking ₦5 million term loans, Osuchukwu explained that companies with similar revenues and years of operation could have significantly different financial positions.
While one business may depend heavily on extended supplier credit to maintain its working-capital cycle, another may sell inventory quickly and maintain stable cash flows.
According to him, a conventional credit scorecard could approve both applicants and offer them the same interest rate, despite the second business demonstrating stronger financial sustainability.
ALSO: FairMoney Hits 30 Million Registered Users, Deepens Digital Banking Push in Nigeria
“The issue is not that lenders cannot identify risk. It is that traditional lending architecture groups materially different borrowers into broad risk segments,” he said.
“When lenders have access to richer information about how customers actually behave financially, treating materially different risks the same is inefficient.”
Osuchukwu argued that the next stage of retail lending should move beyond a simple decision to approve or reject an application.
He said financial institutions should use predictive analysis to determine how much to lend, the appropriate repayment period and the interest rate that reflects the borrower’s financial behaviour.
Lower-Risk Borrowers Pay for Riskier Customers
The Mathesis Analytics CEO said blanket pricing assumes that borrowers within the same segment are sufficiently similar to justify receiving identical interest rates.
In reality, he said, such lenders price the average customer rather than the individual borrower.
This creates what he described as an “invisible micro-subsidy,” in which lower-risk borrowers are charged more to compensate for the expected defaults of higher-risk customers.
At the same time, viable businesses that fall slightly outside rigid credit thresholds may be denied access to financing.
“The result is an inefficient system where good borrowers overpay and viable borrowers are excluded,” Osuchukwu stated.
Behavioural Data Can Improve Credit Decisions
Osuchukwu said improved risk visibility would enable lenders to make more accurate credit and pricing decisions.
Nigeria’s financial ecosystem generates behavioural data through bank transactions, merchant activity, mobile money usage, utility payments and supplier settlements.
When analysed appropriately, he said, this information could help lenders quantify default risk more accurately than broad revenue figures and static historical records.
Daily cash-flow data can reveal volatility and provide a clearer assessment of a company’s operational health beyond its headline revenue.
According to Osuchukwu, continuously updated information can also alert lenders when a borrower’s transaction activity declines or cash inflows improve, allowing the institution’s risk assessment to reflect changing financial circumstances.
Personalised Interest Rates
With improved visibility, Osuchukwu said lenders could move from fixed pricing categories to more flexible risk-pricing models.
Algorithms could analyse a borrower’s consolidated financial footprint, combining internal account history with external indicators such as payment and transaction patterns.
The estimated level of risk could then be translated into a personalised interest rate, while lenders retain their cost-of-funds baselines and risk limits.
Osuchukwu said this would allow financial institutions to offer competitive rates to lower-risk customers while providing appropriately priced credit to borrowers with higher but manageable risks.
He argued that precision pricing could improve lenders’ margins while creating a fairer credit market for consumers and small businesses.
“Clean transactional records become bankable assets that actively lower the cost of credit, freeing resilient borrowers from subsidising the defaults of their peers,” he said.
Risk-Based Pricing Could Expand Financial Inclusion
Osuchukwu added that predictive models could help lenders distinguish between businesses that lack conventional credit histories and those that present genuine repayment risks.
This distinction, he said, could expand access to productive credit for underserved individuals and businesses without weakening lenders’ risk-management standards.
“By mathematically distinguishing between businesses that lack a conventional credit history and those that are genuinely risky, the system naturally extends productive credit to underserved segments,” he stated.
The Mathesis Analytics founder predicted that competition would eventually push financial institutions away from static rate cards because of the value lost through inaccurate pricing.
He said banks and other lenders capable of using multiple data sources and algorithmic models to price individual risks would be better positioned to lead the next phase of retail lending.
For consumers and SMEs previously restricted to rigid products or excluded from formal credit, Osuchukwu said the transition could deliver loan pricing that more accurately reflects their financial circumstances.
