Every discount you grant is a cost decision wearing the clothes of a pricing decision. It is signed off in a sales conversation, recorded as a price adjustment, and then it disappears. Nobody books it as an expense, nobody reviews it next quarter, and nobody asks what it bought. Yet a discount is the only lever in the business that reduces what you receive without reducing anything you do.
How do you calculate the true cost of a discount?
Take the discount off the price, leave every cost exactly where it is, and read what is left of the margin. Because a discount removes revenue without removing any work, the whole amount lands on margin. The true cost is that amount, plus the off-invoice terms attached to it and any extra cost to serve the deal created, measured against the volume the discount actually bought.
A discount removes revenue and nothing else
Start with the arithmetic that gets skipped. A five point discount does not cost you five percent. It costs you five percent of price against a margin that is usually much smaller than the price. If a product sells at 100 with a variable cost of 70, the contribution is 30. Grant five points and the price becomes 95, the cost stays at 70, and the contribution becomes 25. You gave away five and lost a sixth of the margin on that unit.
The reason is structural rather than accounting policy. Discounting changes the invoice, not the factory, not the warehouse, not the support queue. The same picking, the same delivery, the same order handling and the same invoice all still happen and still cost the same. So the entire discount falls through to margin, at full value, on every unit it touches.
Start from the pocket price, not the list price
Most companies measure discounting against the list price, which is the one number in the chain that nobody actually receives. Between the list price and the money you keep sits a series of deductions, and the headline discount is usually the smallest of them.
Rebuild the chain deduction by deduction. List price, then the negotiated invoice discount, then volume rebates, then early payment terms, then promotional support, then freight allowances, then returns and credit notes. What remains is the pocket price, and it is the only figure worth comparing against cost. Companies that run this exercise for the first time often find that the pocket price sits several points below what the commercial team believed it was charging, because each deduction was approved by a different person, at a different moment, for a different reason.
How much volume does a discount have to earn back?
The usual justification for a discount is volume. That claim is testable, and the test is a single line of arithmetic. To hold total contribution flat, the required volume increase equals the discount divided by the contribution margin that remains after the discount.
Take the earlier example. Contribution margin of 30 percent, discount of 5 points, remaining margin of 25 points. Five divided by twenty five is 20 percent. The customer has to buy a fifth more volume just to leave you exactly where you were before you discounted. If the discount is 10 points against the same 30 percent margin, the required increase is 50 percent. This is why deep discounts almost never pay for themselves: the volume they need is larger than the market can supply at that account.
Run this number before the negotiation, not after it. It converts an argument about generosity into a question with an answer: is this account capable of the increase, and did it deliver the increase last time?
The discounts that never appear on the invoice
Off-invoice terms are where the real erosion lives, because they are granted in a different conversation from the price and are almost never added back to it. Extended payment terms are a discount paid in financing cost. Free delivery is a discount paid in logistics. A dedicated account manager, custom packaging, a faster service level or an accepted pattern of small emergency orders are all discounts paid in cost to serve.
None of these show up when you audit the price list, which is why the price list keeps looking healthy while the margin does not. The fix is to convert every concession into money and attach it to the account that received it. Once cost to serve sits next to the pocket price, you can see the accounts where a modest headline discount is hiding a large total one.
A discount is not a price adjustment. It is an unfunded transfer from your margin to your customer, and it repeats every month until somebody stops it.
How to make the cost visible before you grant it
Discounting rarely fails because of a single bad deal. It fails because there is no moment in the process where the cost of the concession is visible to the person granting it. The sales manager sees a price and a target. The cost sits in a different system, in a different format, owned by a different function.
Three changes close that gap. Give every deal a pocket price rather than a list price and a discount percentage. Attach the cost to serve of the account to the deal record, so the margin shown is the margin after the work the account actually generates. And put the required volume increase on the approval screen, so the trade being made is stated as a number rather than assumed as a benefit. None of this restricts what a salesperson may agree to. It only makes them agree to it with the price of the concession in front of them.
The last step is review. Pull the twelve months of discounts already granted, group them by account, and compare the volume promised with the volume delivered. Discounts are recurring commitments that nobody re-approves, and that is exactly why the review finds money every time it is run.
ProfitAudit 360 rebuilds the pocket price and the cost to serve behind every customer, so the next discount conversation starts from evidence rather than from the list price.
Related reading: the pocket price waterfall, the margin cascade and how to price a new product using cost data.