How to Set a Customer Credit Limit: 4 Methods

A credit limit is the maximum exposure your company accepts with one customer at a given moment. Setting that number by gut feel is the most common mistake in trade credit: too high concentrates risk, too low blocks good sales and pushes the account to a competitor.

Four methods are in common use. They are not mutually exclusive — the sound practice is to calculate with more than one and take the lowest result.

Method 1: percentage of customer revenue

The most direct approach. Apply a percentage to the customer's reported or estimated monthly revenue.

Limit = monthly revenue × exposure percentage

The percentage depends on risk and on how central your product is to their operation: 1% to 3% for a non-essential supplier, up to 10% when you are a core input. A retailer doing $400,000 a month at 3% lands on a $12,000 limit.

The weakness is obvious: self-reported revenue without verification is not reliable information. Use it as a ceiling, never as the only reference.

Method 2: ability to pay

This starts from cash generation rather than company size. It requires financial statements, so it fits larger exposures.

Limit = monthly free cash flow × average terms granted × safety factor

If the customer generates $60,000 of free cash per month, you sell on net 30, and you use a 0.4 factor, the limit lands at $24,000. The safety factor exists because you are not the only supplier competing for that cash.

Method 3: portfolio concentration cap

This method looks inward rather than at the customer. It caps exposure per account as a share of your own equity, revenue or margin.

• No single customer above 5% of total receivables
• No customer whose write-off would consume more than 20% of monthly profit
• Corporate groups treated as one exposure

This is the guardrail against concentration. Plenty of companies discover too late that 40% of receivables sat with three accounts.

Method 4: score-based tiers

Used in high-volume operations where individual analysis is not practical. The score places the customer in a band, and each band carries a preset limit.

The table has to be recalibrated against your own portfolio history — a generic bureau score does not know how that customer behaves with your product. See how credit scoring models work.

Combining the methods

The sequence that works in most operations:

1. Calculate the revenue-based limit (commercial ceiling)
2. Calculate ability to pay, where statements exist
3. Apply the portfolio concentration cap
4. Take the lowest of the three
5. Adjust for score and internal payment history

Starting limit versus relationship limit

A new customer does not get the full number. Healthy practice is to grant 30% to 50% of the calculated limit for the first few months and expand based on payment behavior — the most predictive information there is, better than any external score.

Write the expansion trigger down: three orders paid on time, no delinquency beyond five days, unlocks the next tier.

When to reduce a limit

A credit limit is not a vested right. Reduce it when you see:

• Recurring lateness, even when the account always pays eventually
• A new derogatory filing, judgment or lien
• A sharp drop in purchases with no commercial explanation
• A change in ownership
• Requests for extended terms outside the norm

The reduction should be communicated by sales, in advance, with a reason — that is less damaging than a declined order at the counter. Tracking those signals continuously is the subject of credit portfolio monitoring.

Common mistakes

• Set once, never revisited. Risk changes; the number has to change with it.
• Limits by entity inside a corporate group. Hidden aggregate exposure.
• Confusing limit with terms. Two separate decisions: how much, and over how many days.
• Releasing limits under month-end sales pressure. If it happens every month, the problem is the policy, not the order.

What to take from this

Calculate the limit more than one way and keep the lowest. Start small with new accounts, expand on proven behavior, and define the triggers that reduce the number up front — ideally before you need them.