How Should You Actually Determine a Credit Limit?

Every credit limit a distributor sets is a bet in one of two directions. Set it too conservatively, and you lose a good customer's order to a competitor willing to say yes faster. Set it too loosely, and you find out six months later — usually on an aging report, usually at the worst possible time — that you extended more trust than the account could support.

Most credit managers know this tension intuitively. What's harder is having a repeatable process that actually manages it, instead of relying on instinct built up over years of "just knowing" a good account when you see one. Instinct isn't wrong, exactly — it's incomplete. Here's what a more complete process looks like, drawing on the methodology NACM (the National Association of Credit Management) has taught credit professionals for decades, along with the financial-benchmarking and academic research that sits underneath it.

Judgment still matters — it's just not the whole job

It's worth saying up front: no credit scoring model or piece of software replaces a credit manager's judgment entirely, and the people who've spent decades in this field are the first to say so. As one NACM affiliate president put it, character is something you have to experience — no model can assess a business owner's actual willingness to pay the way a credit professional who's worked with them can. That's real, and it matters.

But judgment works best sitting on top of a structured process, not substituting for one. The rest of this article is about that structure.

The foundation: the Five Cs of Credit

The starting framework taught across the credit industry — banking and trade credit alike — is the Five Cs: Character, Capacity, Capital, Collateral, and Conditions. No single "C" should be evaluated in isolation. A complete credit picture requires weighing all five together, and strength in one area can reasonably offset a weakness in another.

For a wholesale distributor evaluating a new or existing account, the Five Cs translate roughly like this:

Character — Has this business paid its obligations reliably in the past? What does its ownership's track record look like?

Capacity — Can the business actually generate enough cash flow to pay you on time? This is where a customer's own DSO and cash conversion cycle matter — a business that's slow to collect from its customers will be slow to pay you, structurally, regardless of intent.

Capital — What's the business actually worth, net of liabilities? This is the net worth or equity cushion behind the credit you're extending.

Collateral — Is there anything securing the exposure — lien rights, personal guarantees, or other recourse if the account defaults?

Conditions — What's happening in this customer's specific end-market? A distributor selling into construction faces different seasonal and cyclical exposure than one selling into industrial manufacturing or healthcare.

The Five Cs tell you what to look at. They don't, by themselves, tell you what number to actually put on the account. For that, you need a method.

From framework to formula: how credit limits actually get calculated

NACM's own credit training material outlines several distinct methods firms use to translate the Five Cs into an actual dollar figure. None of them is meant to be used alone — NACM's guidance is explicit that these are typically combined and averaged, with one method producing a preliminary estimate that's then refined by others. The methods most relevant to wholesale distribution:

By Formula. This method uses key financial data — net worth, inventory, current assets, or net working capital — divided by the customer's estimated number of creditors, to arrive at a reasonable exposure per supplier. The practical challenge is obvious: you rarely know exactly how many other creditors a customer has. That's why this method is typically used to produce a preliminary estimate rather than a final number, refined against other inputs.

Payment Performance. A deliberately conservative approach: start new or unproven accounts at a modest limit, and grow it as the customer builds a track record of paying on time. This is especially useful when little or no payment history exists yet — which describes a large share of new wholesale distribution accounts.

Expectation of Use. Sometimes called the requirements method — you estimate the customer's expected annual dollar volume of purchases, divide by the expected number of orders across the year, and use that to set a working limit. Like the formula method, this is typically a starting estimate, later refined by actual performance.

Payment Record. Once a customer has purchase history with you, limits can flex based on demonstrated ability to pay on terms — a method that makes ongoing credit decisions more routine, while triggering a more detailed review the moment an account misses a payment.

Agency Rating. Using third-party credit report data to set or validate a limit — simple to implement and easy to communicate across departments, but entirely dependent on how much data a rating agency actually has on the account.

In practice, a reasonable wholesale distribution process looks like this: use the Formula or Expectation of Use method to set a preliminary number for a new account, apply Payment Performance discipline for the first several purchase cycles, and let Payment Record data take over once there's enough history to trust it.

Layering in third-party data — without over-relying on it

Once a customer has enough of a file to be scored, third-party data adds real value. Dun & Bradstreet's scoring layers, commonly referenced in NACM's own credit training materials, include a Predictive Score ranging from 450 (high risk) to 850 (low risk) built on twelve months of payment history, alongside a separate Viability Score that estimates the probability a company will cease operating within the next twelve months. NACM members also have access to the National Trade Credit Report, which aggregates tradeline payment data submitted by NACM affiliates across 40 locations nationwide, showing a real-time score alongside a two-year trend of percent-past-due history.

These tools are genuinely useful. But they share a structural limitation worth naming directly: they only work as well as the underlying file is deep. Which brings up the problem that matters most in wholesale distribution specifically.

The thin-file problem in wholesale distribution

A large share of the customers a distributor extends credit to are exactly the accounts third-party bureau data is weakest on: smaller operators, newer businesses, or companies that simply haven't accumulated enough reported trade history to generate a reliable score. This isn't a fringe case — it's closer to the norm in B2B trade, where reporting to commercial bureaus is far less universal than consumer credit reporting.

For these accounts, the highest-signal input available isn't a bureau file — it's how the business is actually paying you, specifically. Days-to-pay trending later each cycle, a partial payment pattern starting, rising utilization of an existing credit line — these are signals that exist for every customer regardless of bureau file depth, and they often show up well before any external score would move. Treating internal payment-behavior data as a primary input, not a fallback, is arguably the single highest-leverage adjustment a distributor's credit process can make.

Sanity-checking against real benchmarks

Two additional data sources are worth building into a credit process as a reality check, rather than a primary decision input.

Wholesale and distribution DSO typically clusters in the 30-to-45-day range, since invoices tied to shipped goods with documented delivery tend to compress dispute windows compared to service industries. The Credit Research Foundation's Q4 2025 domestic trade receivables summary put the broader cross-industry median DSO at 40.5 days. If your portfolio's actual DSO is running well above your stated terms and above these benchmarks, that's a signal your credit limits — or your collections process — may be out of calibration, independent of any single account's score.

For customers willing to share financial statements, RMA's Annual Statement Studies remain the industry-standard source for benchmarking a company's ratios — current ratio, quick ratio, and dozens of others — against peers in the same industry and size class. The data has an unusually long pedigree: material first appeared in the March 1919 issue of the Federal Reserve Bulletin, and RMA's modern studies are now built from well over 100,000 financial statements across hundreds of industries. It's a useful way to tell whether a specific customer's Capacity and Capital genuinely stack up against others in their space, rather than against a generic rule of thumb.

It's also worth remembering, per long-standing research on trade credit from the Federal Reserve, that trade credit isn't just a risk to manage — it's a competitive tool sellers use deliberately, and buyers actively seek out, as part of how B2B commerce functions. The goal of a good credit process isn't to minimize risk to zero; it's to extend credit as confidently and accurately as possible, because credit itself is part of what wins and grows the relationship.

Putting it together: a repeatable process

A reasonable end-to-end process for a wholesale distributor looks like this:

Gather the inputs — financial statements if available, bank and trade references, and whatever third-party data exists for the account. Run two or three methods and reconcile them, rather than leaning on just one. Set an initial limit, erring conservative if the data is thin — this isn't a permanent ceiling, it's a starting point that should move as real data accumulates. And build in a review cadence: a limit set once at onboarding and never revisited is the single most common way credit processes quietly drift out of calibration.

What it actually costs to get this wrong

Both failure modes have a real, quantifiable cost — they just show up in different places.

A limit set too conservatively shows up as lost revenue that never gets tracked as a loss, because it just looks like a sale that went to a competitor instead. A limit set too loosely shows up later, compounding: an overextended account drags out your portfolio's DSO, ties up working capital that could otherwise fund growth, and — if it eventually defaults — becomes bad debt that's far more expensive to absorb than the margin on the original sale ever justified. Given that wholesale distribution already runs on some of the longer DSO cycles in B2B commerce, a credit process that leans too loose compounds a structural cash-flow challenge the industry already faces.

Neither failure is inevitable. Both are usually the result of the same root cause: a credit decision made once, from incomplete information, and never revisited.

The fix isn't more manual review — it's a process that keeps up automatically

Everything above is achievable by hand for a handful of accounts. It stops being achievable the moment your book grows past a few dozen customers, which is exactly when most credit teams quietly fall back on gut feel — not because the framework stopped mattering, but because nobody has time to run it consistently.

That's the specific gap Thor's Credit Applications are built to close: a proprietary risk score that doesn't require a bureau file, applied consistently to every applicant, with a recommended limit your credit manager can see — and see the reasoning behind — in minutes instead of days.

If you want to see what that looks like against your own accounts, book a free demo and we'll walk through it live.