Discounting is the easiest lever in the business, which is why it gets pulled first and argued about least. It is worth doing the arithmetic once, properly, because most teams have never seen it written down.
The arithmetic
Take a product that sells for $50, where $30 goes to cost of goods, shipping and payment fees. Your contribution is $20. Now run twenty percent off. The price becomes $40 and none of the costs move, so your contribution is $10.
You did not give away twenty percent. You gave away half.
What that means for volume
To earn the same total profit at that discount, you do not need twenty percent more orders. You need twice as many. That is the number worth putting in front of everyone in the room before a promotion is agreed, because it changes the conversation from whether the discount will lift sales to whether it can plausibly double them.
When it is still the right call
Clearing stock that is heading for a write-off. Acquiring a customer you have good reason to believe will buy again, where the first order was never going to be the profitable one. A genuine launch moment. The test we apply is whether the discount buys something beyond the sale itself. If it only buys the sale, it is usually too expensive.
What we would rather change
The thing in the box, not the number on the price tag. A bundle, a larger size, a subscription price, a gift with purchase. In our experience these move conversion nearly as well, protect the margin, and avoid the worst side effect of a discounting habit, which is teaching your best customers to wait for the next one.
A discount is not free marketing. It is the most expensive marketing available, paid entirely out of profit, and it deserves the same scrutiny as any other line in the budget.