Cash flow predictions improve small business credit decisions
CADENCE: Distributional Next-Transaction Forecasting for Business Cash-Flow Underwriting
Machine Learning
Summary
Predicting when and how much money a small business will receive regularly can be tricky because payments are often unpredictable and vary in timing. The authors created CADENCE, a compact AI model that treats each transaction like a word in a sentence to forecast the next payment amount and timing more accurately. It uses knowledge about the calendar and descriptions of transactions to improve predictions, especially for businesses with limited payment history. These better predictions help lenders make smarter credit decisions by understanding cash flow patterns more clearly.
What this means in practice
- •For financial institutions: Improve credit scoring models for small businesses by integrating precise cash flow timing and amount predictions to increase approval rates and reduce losses.$Commercial implications: Allows lenders to offer better credit decisions based on enhanced cash flow forecasts, improving financial product offerings for small businesses.
- •For financial software developers: Develop advanced cash flow analytics tools that provide interpretable timing distributions of future transactions for risk assessment and underwriting support.
Authors
Alexander Wu, Jingchao Huang
Abstract
Recurring cash flows are the backbone of cash-flow underwriting: they separate durable revenue from one-off deposits, expose standing obligations, and ultimately determine ability to pay. For small businesses, recurrence is both indispensable and hard to estimate, because much of their income arrives as ambiguous transfers, most payment streams are short, and schedules follow calendar conventions that frequency-domain methods treat as noise. We present CADENCE, Calendar-Aware Distributional Event Network for Cash-Flow Estimation, a compact causal Transformer that treats each transaction as a token and predicts the amount and a full, nonparametric timing distribution of every cash flow's next payment. Inheriting transaction semantics from an external labeler and a frozen text encoder keeps CADENCE compact yet general across the long tail of small-business counterparties, while predicting a calibrated distribution rather than an opaque embedding makes its outputs interpretable enough to feed existing underwriting models and power credit decisions. Ablations confirm the role of each design choice, with counterparty semantics driving recall on short histories, calendar inputs driving timing accuracy, and no-event supervision driving precision. Against the production FFT baseline, CADENCE improves one-day timing accuracy by 93.3% and next-event F1 by 6.7%, while producing calibrated distributions that serve directly as a confidence gate for downstream decisions. In an end-to-end credit evaluation on production applications, replacing FFT recurrence features with CADENCE features raises KS by 4.8%, raises the approval rate at a fixed loss rate by 18.2%, and lowers the loss rate at a fixed approval rate by 13.1%.