What A Lender Reads In Your Bank Statements

Where statements or open banking data are supplied, the lender is not glancing at the balance. It is extracting a structured picture of money coming in, money committed and how the account behaves under pressure.

Income is tested for regularity

Credits are grouped and examined for pattern: same source, similar amount, predictable interval. A stable series is treated as income, and an irregular one is treated cautiously.

Transfers between a customer's own accounts, refunds and one-off receipts are filtered out where they can be identified, since they are not earnings.

Variable and seasonal income is usually averaged over a period, which means a good recent month counts for less than a consistent record.

Committed outgoings are separated from discretionary ones

Recurring debits are classified: housing costs, utilities, credit repayments, insurance and subscriptions. These form the committed base that must be met before anything new is added.

Discretionary spending is treated differently, because it can flex, though a pattern of high discretionary spending against modest income is still informative.

Payments to other lenders are cross-checked against the credit file, and a repayment visible on statements but absent from the file prompts questions.

Strain signals carry disproportionate weight

Certain patterns are read as pressure regardless of the balance: returned direct debits, unarranged overdraft use, frequent small transfers at month end, and payments to short-term lenders.

Gambling transactions are commonly identified and assessed, particularly where the volume is large relative to income or clustered around payday.

None of these is disqualifying on its own. What matters is whether the account recovers each month or grinds progressively closer to its limit.

Balance behaviour matters more than the closing figure

Assessments look at the shape of the balance over the period: how long it spends below zero, the lowest point reached, and whether the trend across months is up or down.

An account that dips and recovers reads differently from one that never returns to positive, even where both end the period at the same figure.

This is why moving money in before an application changes little. The history is what is analysed, not the position on the day.

Categorisation is automated and imperfect

Transaction classification relies on merchant descriptors and pattern matching, which misfires on unusual payees, personal transfers and payments made through intermediaries.

Misclassification can turn an ordinary transfer into an apparent commitment, or hide a real one, and the applicant rarely sees the categorised version.

Where data is obtained through an open banking connection, the permissions granted, the retention of the data and the rights attached to it differ by jurisdiction.