SELLWISE INSIGHTS
Why discounts can hurt e-commerce profitability
A transparent contribution example explains why more orders need not mean more profit—and what to measure before discounting.
Start with the question, not the coupon
A promotion can increase order count while reducing the money left to run the business. The useful question is not simply whether conversion rose. It is whether the offer created enough additional contribution to cover the discount and associated costs. Revenue, gross margin and contribution are different measures; a dashboard should name the one it actually computes.
Shopify’s margin guide distinguishes gross, operating and net margins. Our example below is narrower: contribution after a deliberately simplified set of variable costs. It is not a full profit-and-loss account or a recommended discount level.
A small discount, a larger contribution gap
Imagine a product sold for €100 with €60 of variable costs. The contribution is €40. A 10% discount lowers the price to €90; if those costs stay unchanged, contribution falls to €30. That is a 25% reduction in contribution per order, not a 10% reduction. These are invented inputs for arithmetic, not observed merchant results.
One hundred full-price orders contribute €4,000. At €30 per discounted order, matching that amount requires at least 134 orders: roughly one third more. This comparison assumes unchanged costs and basket mix. Returns, payment fees, fulfilment, acquisition costs and inventory constraints can change the result.
Measure what changed because of the offer
Counting coupon users does not reveal how many would have bought anyway. Compare an eligible group receiving the offer with a properly designed holdout, using a predefined outcome window and contribution measure. Keep refunds and offer costs visible. Do not stop a test just when the chart looks encouraging.
Discounts can be rational for a specific inventory or acquisition objective. Record that objective, the cost assumptions and the uncertainty. Consider fixing missing delivery information or product guidance before paying for an intervention. Doing nothing is a valid comparison. Sellwise’s illustrated offers are concepts, not evidence of available production activation or measured uplift.