There is a fundamental problem at the heart of small business lending that the traditional credit infrastructure has never adequately solved. The problem is this: the best predictor of whether a small business will repay a loan is not its credit history, not the personal credit score of its owner, and not its collateral position. It is the actual, observable pattern of cash flowing into and out of the business on a daily basis: the rhythm of revenue, the timing of expenses, the consistency of deposit behavior, the seasonal patterns, and the trajectory over time.
Cash flow is the truth of a business’s financial health in a way that credit scores and balance sheets are not. A business can have a poor credit history due to a single difficult period that has since resolved, while currently generating strong and consistent cash flow that makes it an excellent credit risk. A business can have strong collateral and a good credit score while running a cash flow pattern that reveals growing financial stress long before it shows up in formal financial statements. Traditional underwriting, built around static credit metrics and point-in-time financial snapshots, misses both categories systematically.
Carrington Labs was built to address this gap, and the launch of Cashflow Score 2.0, with expanded explainability capabilities, represents a significant maturation of the tool that cash flow underwriting practitioners have been using since the company’s first version.
What the Cashflow Score Does
Carrington Labs’ Cashflow Score translates the complex, noisy data of a business’s bank transaction history into a single, actionable underwriting signal. The system ingests months of transactional bank data, including deposits, withdrawals, transfers, recurring obligations, payroll patterns, revenue seasonality, and applies machine learning models trained specifically on small business cash flow behavior to generate a score that reflects the business’s demonstrated financial health and repayment capacity.
This is meaningfully different from what credit bureaus do. Credit bureaus aggregate historical repayment behavior, focusing on whether debts were paid, when they were paid, and how much credit is in use. This is useful information, but it is backward-looking in a narrow way and says nothing about the current cash generation capacity of the business. The Cashflow Score is built on the actual mechanics of how money moves through a business, which is a much more direct measure of whether there is enough cash to service a new debt obligation.
For small business lenders who have been losing potential good borrowers because their credit scores are insufficient for traditional approval, and who have been taking on borrowers with strong credit scores who subsequently defaulted because the credit score never revealed the underlying cash flow problems, the Cashflow Score 2.0 is an underwriting tool with direct implications for both portfolio quality and addressable market.
The Explainability Expansion: Why It Matters
The version 2.0 launch centers specifically on expanded explainability, the ability to understand and communicate why the Cashflow Score assigned a particular score to a particular applicant. This might sound like a technical refinement, but in the context of lending, explainability has significant practical, regulatory, and ethical dimensions.
On the regulatory side, adverse action requirements in lending mandate that lenders who decline an application must be able to provide specific, actionable reasons for the decision. A score that says “declined” without explainable factors cannot be used to generate the required adverse action notices. Cashflow Score 2.0’s expanded explainability framework generates the specific cash flow factors, whether positive or negative, that most significantly drove the score, in language that can be translated into compliant adverse action communications.
On the commercial side, explainability enables underwriters to have genuine conversations with borrowers about their applications. When a loan officer can say “your score was affected specifically by the pattern of your deposit timing over the past 90 days” rather than simply “your score was insufficient,” the conversation becomes more productive. Borrowers who were declined can understand what would need to improve for a future application to succeed.
The Market Context: Cash Flow Underwriting Goes Mainstream
The adoption of cash flow-based underwriting in small business lending has accelerated significantly over the past three years. The proliferation of open banking frameworks and bank account data APIs has made it practical for lenders to access and analyze transaction-level bank data at scale. Simultaneously, the limitations of traditional credit-based underwriting have become clearer as the economic volatility of the post-pandemic period demonstrated how quickly credit scores can diverge from actual financial health.
The most active adopters of cash flow underwriting technology have been alternative lenders and fintech platforms that built their underwriting models around bank account data from the beginning. But the model is spreading into community banks, credit unions, and regional banks that have traditionally relied on relationship-based underwriting supplemented by credit bureaus, as these institutions recognize that bank account data combined with a well-calibrated analytical model can both expand their addressable market and improve portfolio performance.
References:
- CarringtonLabs.com, Product Overview and Methodology
- Forbes, Cash Flow Underwriting Trends in Small Business Lending 2026
- American Banker, AI in Credit Underwriting 2026 Coverage
- CFPB, Adverse Action Requirements Under ECOA and FCRA
- Fintech Nexus, Open Banking and Small Business Lending Trends 2026
