The logic of traditional credit underwriting is circular in a way that most lenders rarely stop to examine. To get credit, you need a credit history. To build a credit history, you need credit.
For borrowers who grew up in a system that extended credit to them early — student cards, retail accounts, car loans tied to their first job — this loop is invisible because they stepped into it naturally. For everyone outside that pipeline, the loop is a wall.
Not everyone outside the group is small. More than 9 million Canadians are either credit unserved, underserved, or credit invisible, meaning they have no credit file or insufficient credit information to generate a score.
In the United States, roughly 28 million adults lack conventional credit scores, meaning their credit files are too thin to be scored. These are not marginal populations. They are large, economically active segments that most underwriting models are systematically set up to miss.
The practical problem for lenders is not philosophical but operational. When a creditworthy applicant presents a thin file, most systems either decline them automatically or route them to expensive manual review.
Both outcomes cost revenue. The applicant either goes to a competitor that found a way to say yes, or the lender absorbs processing costs for a decision that often ends in rejection.
Neither option is sustainable at the scale this population represents, and that scale is growing, driven in large part by newcomer immigration patterns and the structural data gaps they create.
Why The Score Misses What it Should be Measuring
The credit score was designed as a shorthand for repayment behavior, a compressed summary of how someone has handled debt over time. It works well for borrowers with long, varied credit histories across multiple product types. It sometimes fails badly when the borrower's financial behavior doesn't map onto the kinds of accounts that bureaus track.
For decades, traditional data — basic information on an individual's current and historical credit accounts — has served as the basis for consumers' credit scores. However, alternative data could bolster, or sometimes substitute for, traditional data, with subsequent benefits and risks for consumers.
The Federal Reserve's October 2025 analysis of alternative data is explicit about this: the same behaviors that a credit score seeks to infer — income stability, spending discipline, consistent payment patterns — are directly observable in cash flow data and often more accurately than any backward-looking bureau file.
This distinction matters practically. Cash-flow data can uncover so-called invisible prime borrowers, increasing credit access for customers who would otherwise be categorized as subprime based solely on traditional credit reporting.
The invisible prime concept is important because it entirely reframes the risk question. The lender is not being asked to take on more risk by serving thin-file borrowers; they are being asked to use better data to see the risk they were previously unable to measure. Those are different requests, and conflating them is expensive.
Consider who actually falls into the thin-file category. Young professionals who have avoided debt by habit. Immigrants who built repayment histories in another country. Gig workers whose income flows through digital platforms rather than a formal payroll. Entrepreneurs whose financial life runs through business accounts rather than personal credit products.
A thin credit file is usually a reflection of modern economic realities, not a sign of poor financial behavior. Many of these borrowers have never missed a payment in their lives; they simply never had the type of account that bureaus are built to track.
The Data that Fills The Gap
The move from score-only underwriting to a broader data picture has been building for years and is now crossing from early adoption into mainstream practice. 90% of lenders in Nova Credit's 2024 survey believe that greater access to alternative data would help them approve more worthy borrowers, including 41% specifically interested in reaching recent immigrants and 39% interested in reaching other thin-file applicants.
The interest is clearly there. What has lagged is the infrastructure to act on it consistently. The data categories now available for thin-file underwriting fall into a few practical groups, and understanding what each one measures and what it doesn't is what separates well-designed models from ones that introduce new blind spots.
Cash flow and banking behavior
This is the most direct substitute for bureau data for measuring repayment capacity. Rather than inferring future behavior from past debt management, cash-flow underwriting looks at what is happening in a borrower's bank account right now.
That is, how often income arrives, how consistent it is, how expenses behave relative to income, and whether the account regularly runs into stress. For example, Experian's partnership with Plaid is specifically designed to build cash-flow credit scores for lenders working with borrowers who don't have extensive credit records, with the explicit rationale that banks are requesting this data because it has a significant impact on financial inclusion.
Similarly, VantageScore launched a model in 2024 combining traditional credit data with bank account information for exactly this purpose. These are not fringe experiments; they are mainstream product decisions by the largest credit infrastructure companies in the market.
Rent and utility payments
These are perhaps the most intuitively fair additions to thin-file underwriting. A borrower who has paid rent on time every month for five years has a five-year repayment history that is, by any reasonable measure, relevant to credit risk. Yet for most of that time, it contributed nothing to their score.
Fannie Mae estimated that about 17% of first-time homebuyers who were initially ineligible would have been approved if they had maintained a consistent 12-month rental payment history. Similarly, Freddie Mac made the same move in 2022, incorporating rental payment data into its automated underwriting system. The impact on newcomers and first-time applicants is direct and measurable.
Cross-border credit history
This, specifically for immigrant populations, is the category where the data gap is most acute, and the infrastructure is most underdeveloped, particularly for African newcomers.
For example, a Nigerian software engineer who arrives in the country with a job offer, a decade of clean repayment history, and documented financial responsibility, but the system designed to measure these qualities cannot see any of it.
The data that would establish their creditworthiness in their home country, including bureau files, BVN-linked repayment records, and mobile money transaction histories, does not connect to Canadian or US credit infrastructure in any standardized way.
The result is that a prime borrower is processed as credit-invisible and either declined or offered secured products years after they no longer need them.
Telecom and subscription payment data
The data rounds out the picture for borrowers whose formal financial footprint is genuinely thin. With over 95% of Canadians using mobile internet, leveraging telecom payment history can be an excellent example of consistent payment behavior not found in traditional credit reporting.
Equifax Canada has incorporated telecom data into its Score Complete product for this specific purpose. It is not a replacement for bureau data, but as a supplementary signal for newcomers or young borrowers with limited account history, it adds meaningful predictive value.
What the Model Needs to Do Differently
Using alternative data well is not simply a matter of adding new inputs to an existing model. The underwriting framework has to be designed around a different question. Traditional credit models ask: How has this person managed debt?
On the other hand, alternative data models ask: What does this person's financial behavior tell us about their ability and likelihood to repay? These are related but not identical, questions, and the distinction has real consequences for which borrowers are approved.
Traditional credit models perform well for borrowers with conventional credit histories, but they leave many creditworthy consumers behind whose income or employment doesn't fit the standard mould.
As work becomes more flexible and the gig economy continues to develop, a growing share of consumers are falling into this category. The practical implication is that lenders who rely exclusively on score-based underwriting are not protecting themselves from risk. They are outsourcing a risk decision to a model built for a population that no longer represents the full applicant pool.
When advanced analytics and expanded data sets are combined, 96% of applicants can be scored, including an estimated 65% of the credit invisible population. This represents millions of borrowers who are visible to a better-designed model and invisible to a traditional one.
For lenders competing for volume in a managed immigration market like Canada's, where newcomer demographics are growing and concentrated in specific source countries, that gap is as much a revenue question as an inclusion question.
How Zeeh Powers Access for Thin-File Newcomers
For African immigrants specifically, one of the fastest-growing newcomer segments in Canada, the thin-file problem is not a product of limited financial behavior. It is a product of data borders.
The credit history exists, but it does not connect to the lender's workflow in any way that the lender can act on. Zeeh addresses this through three products that work together across the underwriting gap.
ZeehID handles identity first, which has to come before any data retrieval is meaningful. It verifies African government-issued documents (BVN, NIN, Cameroonian national IDs, Ghanaian voter cards, Kenyan national identity documents) against primary government and bureau databases.
This step is not separate from the credit workflow, but structurally part of the same pipeline, so that by the time a credit query goes out, the link between the applicant and their data is confirmed.
Then, CreditBridge retrieves the applicant's bureau data from African sources, including CRC and FirstCentral in Nigeria, Metropol in Kenya, XDS in Ghana, and equivalent bureaus across Uganda, South Africa, Cameroon, Zimbabwe, Zambia, and Rwanda.
It then normalizes it into formats that Canadian underwriting systems can read directly. This is the translation problem that most lenders haven't solved. It is not just accessing the data, but converting it into score equivalents, trade line formats, and risk attributes that fit a Canadian credit model without requiring lenders to rebuild their decisioning infrastructure around it.
Insights fills the remaining gap for borrowers whose formal bureau file is thin but whose transaction behavior tells a richer story. Using open banking data, including income flows, expense patterns, and account stability, Zeeh generates a cash-flow risk profile that functions as a repayment capacity signal.
For African newcomers who operated through mobile money systems or informal savings structures before arriving, this layer often captures financial discipline that no bureau record reflects.
The combined output of all three gives a lender a complete picture of an applicant they would otherwise have to decline or process manually at high cost. The borrower's full financial behavior becomes readable within the lender's existing workflow via a single API integration.
Bottom Line
Many underserved Canadian consumers would likely be high-performing and profitable borrowers, but because of their limited credit history, lenders are reluctant to extend credit to them. This is not a risk warning, but a growth signal. The underserved population actively seeks credit, migrates to full credit activity when given access, and becomes a sticky customer segment once served well.
The thin-file borrower is not a risk problem that alternative data helps manage. It is a visibility problem that better data solves. The score misses these borrowers not because they are risky, but because the score was designed around a different population's financial life.
Expanding the data picture, through cash flow, rent history, telecom data, and for immigrant populations, verified cross-border credit history, does not change who is creditworthy. It changes who the lender can see.
And right now, for African newcomers, the gap between who is creditworthy and who a lender can actually see is measured in tens of thousands of applicants per year. That is not a niche use case, but a structural market gap, and the infrastructure to close it is available today.
Zeeh provides financial institutions with verified identity, cross-border credit history, and cash-flow underwriting data for African newcomers. Thus, make thin-file African borrowers fully visible to Canadian lenders.
Ready to explore how it works? Book a demo with our team to get started.
