Take It or Leave It: What Your Cost of Capital Says About Early Payment Discounts
November 17, 2026

November 17, 2026

Most large organizations have a standing answer to early payment discounts. Either the company takes them when offered, on the reasoning that a discount is free money, or it declines them, on the reasoning that cash is better retained. Both answers are applied consistently, neither is revisited often, and both produce the wrong result a substantial share of the time.
An early payment discount is neither an accounts payable matter nor a procurement win. It is a cash deployment decision. Paying an invoice 20 days early in exchange for a discount means committing the company's cash at a specific return, and whether that return is attractive depends on what else the cash could do and on whether the cash is available at all.
This article works through the arithmetic, sets out where the threshold sits and what else belongs in the decision, and covers the mirror error that frequently costs more than the discount question itself.
A discount offer has three components: the discount percentage, the number of days earlier the payment must be made to earn it, and the original term. The calculation converts those into an annualized rate so the offer can be compared against other uses of the cash.
The standard formula divides the discount by the amount actually paid, then annualizes over the days gained. Expressed directly, that is the discount divided by one hundred minus the discount, multiplied by 365 divided by the number of days gained.
For a 2 percent discount in exchange for paying 20 days earlier, that is 2 divided by 98, multiplied by 365 divided by 20, which comes to approximately 37 percent on an annualized basis.
Thirty-seven percent is a high return by most standards, which is why 2 percent for 20 days is generally worth taking. The figure surprises people, and that surprise explains why the decision is so often made poorly. A 2 percent discount sounds small. Thirty-seven percent annualized does not. The same arithmetic also identifies offers that are not worth taking, which is where many organizations err in the opposite direction.
McKinsey published a decision matrix in January 2025 comparing the annualized value of a discount against the number of days paid early, dividing the result into three zones. Its matrix assumes a weighted average cost of capital of 12 percent. That is an illustrative figure rather than a recommended threshold, and each organization should substitute its own.
The three worked examples it provides are instructive because the answers are not obvious.
This gains 50 days in exchange for half a percent, which annualizes to under 4 percent. Against a 12 percent cost of capital it is a clear decline. Many organizations accept this type of offer reflexively because the discount is presented as a saving.
This gains 40 days for 1.5 percent, which annualizes to approximately 14 percent, modestly above a 12 percent threshold. McKinsey treats this as a judgement call depending on how far the company currently needs to prioritize cash. It is the zone where a decision rule does not substitute for a conversation.
This gains 15 days for 3 percent, which annualizes to approximately 75 percent. Against almost any plausible threshold this is worth taking, and an organization declining it as a matter of policy is forgoing a substantial return.
Framing the comparison against weighted average cost of capital is a useful discipline, and treating it as a universal decision rule is a mistake. WACC is a blended, long-run figure describing the cost of the company's overall capital structure. The decision in front of an accounts payable team is short-dated, specific, and constrained by circumstances WACC does not describe. Four considerations sit alongside it.
Available liquidity. A discount that clears the return threshold is still the wrong decision if taking it strains the cash position or draws on a facility the company would rather keep undrawn. The relevant comparison in that case is against the marginal cost of the funding actually being used, not against a blended average.
Short-term borrowing and deposit rates. Where cash is genuinely surplus, the more accurate comparison may be against what that cash currently earns, or against the rate on the revolving facility the payment would otherwise draw on. Both can sit some distance from WACC.
Alternative uses of the cash. Capital committed to early payment is unavailable for inventory, capital expenditure, or debt reduction. Where other claims are competing for the same cash, the threshold is effectively set by the best of those alternatives.
Covenants and seasonality. Period-end balance sheet positions and covenant headroom can make an economically attractive discount unattractive at a particular moment. The arithmetic does not change; the constraint does.
The practical conclusion is that the annualized calculation identifies which offers are worth examining. The threshold against which it is judged should be set by treasury with those constraints in view, and should be capable of changing when conditions do.
The reason to revisit this now is that the threshold itself has shifted while the policies resting on it have not.
A 0.5 percent discount in exchange for paying 30 days earlier is a different discussion in a 5 percent cost of capital environment than in a 10 percent one. Offers that were marginal may now be clearly uneconomic, and offers the company has been declining by habit may have moved across the threshold in either direction.
The same requirement should also be taken into account as governance, recommending that cost of capital be incorporated as a standard consideration both when negotiating terms with vendors and when evaluating and executing existing discounts. The word existing carries weight. Most organizations hold a stock of discount arrangements agreed under conditions that no longer apply, and in many cases nobody owns the question of whether they still make sense.
The discount question attracts attention because it involves a number. The larger leak often involves no number at all.
Many organizations pay invoices before they are due without receiving any discount in return. This happens for ordinary operational reasons: payment runs processed daily rather than batched, or forward-looking runs that settle invoices not yet due. Each instance is small and individually defensible. In aggregate the company is deploying cash at a zero return while funding it at a cost.
There is a supplier-facing argument for doing this deliberately rather than simply paying later. A payment cycle that is communicated and adhered to gives suppliers predictability, which has real value to them, and is a better position than an unpredictable schedule that happens on average to run early.
A structural problem sits underneath both errors. Discount terms and net terms are usually managed by different people against different objectives.
Procurement negotiates discount terms, often as part of a wider commercial package and often without reference to the treasury position. Treasury manages cash and funding cost but is not in the room when the terms are agreed. Accounts payable executes whatever arrives. The result is a portfolio of arrangements that no single function has evaluated as a portfolio.
Assessing them together changes some decisions. A supplier offering an attractive discount and a supplier whose net terms sit well below market represent different opportunities requiring different conversations, and the better answer for a given supplier may be to improve the net terms and decline the discount, or the reverse. Those trades become visible when both sides are assessed against the same funding threshold and against aggregated market benchmarks for comparable suppliers, which is the kind of market context Calculum's platform exists to supply.
The allocation that works in practice separates the rule from the negotiation.
Treasury owns the threshold. It sets and publishes the rate against which discount offers are judged, reviews it as conditions change, and defines the range within which a judgement call is permitted rather than escalated. Procurement negotiates against that published rule, which is considerably easier than negotiating without one. Accounts payable executes to a defined schedule rather than a continuous cycle.
The frontline decision-making is key, arguing for a clear and practical policy framework so that teams can manage the trade between payment terms and discount percentage day to day rather than referring each instance upward. A buyer who knows the threshold can make the trade at the table. A buyer who does not will accept whatever is presented as a saving.
The practical conversion is short. Publish the threshold. Audit the existing stock of discount arrangements against it. Measure how much is being paid ahead of contractual due dates. Batch the payment run and communicate the cycle. Assess discounts and net terms in the same review rather than in separate ones.
Convert the offer to an annualized rate and compare it against the relevant funding threshold. Divide the discount by the amount actually paid, then annualize over the days gained. A 2 percent discount for paying 20 days early is 2 divided by 98, multiplied by 365 divided by 20, or approximately 37 percent annualized. If that exceeds the cost of the cash being deployed, the discount is generally worth taking.
Usually. It offers 2 percent for paying 20 days early, which annualizes to approximately 37 percent, comfortably above most funding thresholds. The offers that merit more scrutiny are small discounts for long acceleration, such as half a percent for paying 50 days early, which annualizes to under 4 percent and is generally worth declining.
It is a useful starting point rather than a universal rule. WACC is a blended, long-run measure of the cost of the overall capital structure, while the decision is short-dated and specific. Available liquidity, short-term borrowing and deposit rates, competing uses for the cash, and covenant or seasonal constraints all belong in the judgement. The annualized calculation identifies which offers are worth examining; treasury should set the threshold they are judged against.
The company deploys cash at a zero return while funding it at a cost. It usually happens for process reasons rather than deliberate ones: daily payment runs instead of batched cycles, or forward-looking runs that settle invoices not yet due.
Treasury should own and publish the threshold and define where a judgement call is permitted. Procurement should negotiate against that published rule. Accounts payable should execute to a defined payment cycle. The important element is that the threshold exists in advance, so the person at the negotiating table can make the trade between terms and discount without escalating it.
Calculum is an AI-powered payment terms intelligence platform that helps enterprise finance, treasury, and procurement teams benchmark payment terms against aggregated and anonymized market data, identify optimization opportunities, and make data-driven working capital decisions. With coverage across millions of companies globally, Calculum lets organizations assess discount terms and net terms against the same market reference rather than managing them as separate decisions.