The limit you set in 2024 is still shipping in 2026

Customer credit limits should be reviewed when material risk, payment behaviour or exposure changes — not only on an annual calendar. A limit set two years ago reflects the financial position and trading conditions visible at the time. Continuing to ship against it without reassessment turns an old decision into new exposure.

Every credit limit is a decision with a date on it. Most systems just don’t display the date. The limit field looks current because the account is active and the orders keep flowing. The decision behind the number may be two years old, made about a customer who no longer exists in the form you assessed.

How often should customer credit limits be reviewed?

The honest answer: as often as the customer materially changes — and customers do not change on your review calendar. An annual review is a floor, not a rhythm. It exists because filing cycles are annual, not because risk is. In practice most books reprice only a fraction of their limits in any year; the rest ships on history.

Why credit limits become outdated

A limit encodes everything you knew on the day you set it: the accounts filed then, the payment behaviour seen then, the trading conditions assumed then. None of that is fixed. The customer’s position moves; the limit doesn’t.

The past week showed exactly when this bites. Construction sentiment rose to a four-month high and lender confidence improved sharply — while construction starts ran 29% below last year and surveys of SME directors showed cashflow still at cycle lows (covered in our 2–8 August digest). An improving mood is precisely when old limits start to look safe again. But the sentiment is a forecast. The ledger is a fact. A limit re-trusted on mood is the same old decision, extended on weaker evidence.

Which changes should trigger a credit-limit review?

The triggers are specific and mostly observable: payment behaviour drifting later, or your invoices moving down the customer’s queue; new court data — judgments, petitions; filing behaviour changing — late accounts, an auditor change; directors leaving or arriving; ownership or group structure shifting; the customer’s own customer base concentrating. Any one of these is a reason to re-ask the limit question. Several together are the answer.

The cost of not asking arrived in this month’s insolvency filings. Ceiling manufacturer Zentia entered administration in June; in August, administrators’ filings showed £5.6m owed to subcontractors and suppliers, with unsecured creditors unlikely to recover anything. That exposure accumulated on limits set when the accounts looked healthy. The write-off carries August’s date. The shipping that built it happened earlier — invoice by invoice, on decisions nobody had revisited.

How to review limits by event rather than calendar

Three questions turn a static limit field into a managed decision:

1. When was this limit set?

2. What has changed since then?

3. Is today’s exposure still supported by today’s customer?

If the answer to the second question is “we don’t know”, that is the finding. Monitoring turns the second question from an annual project into a running answer — the change arrives as a signal, and the signal triggers the review. We have argued before that the check was never the decision: approval is the moment you know the most about the customer and the least about what happens next. The limit is that moment’s longest-lived artefact.

A credit limit is not a property of the customer. It is a memory of a decision — and memories age. The limit remembers 2024. The customer is living in 2026.

See how Grand helps at heygrand.com.