Raising prices is the most powerful lever most stores own and the one they use last, because the fear is concrete and the benefit is abstract. The arithmetic reverses that. The goods, the parcel and most of the fee do not move when a price does, so almost every net cent of an increase flows straight through to contribution, which means a ten percent rise typically lifts what a unit keeps by around a quarter. That asymmetry produces a number worth having before the argument starts: how much volume you could lose and still be no worse off.
- A 10% increase on the example product lifts contribution from €15,35 to €19,37, a gain of about 26%.
- The tolerance line follows: total contribution holds even if roughly 20,8% of orders never come back.
- The tax authority and the gateway take a share of the increase, so plan on the net gain rather than the sticker.
- Raise in steps, watch orders against the line, and reverse early if reality disagrees.
Compute the tolerance before the meeting
One minus the ratio of contribution before to contribution after is the share of volume you can afford to lose. It converts a nervous conversation into a threshold: nobody has to predict what customers will do, they only have to argue about whether more than a fifth of them would leave. A price increase impact calculator does the division on your own figures, and the answer is usually wider than the fear.
Two deductions keep the number honest. VAT rises with the sticker and belongs to the tax authority, and the gateway charges its percentage on the larger gross, so the contribution gain is smaller than the sticker increase. Both are small, both are certain, and planning on the sticker figure overpromises every rise.
Where to raise first
Start where comparison is hardest: own-brand products, bundles, anything with strong reviews or no identical listing elsewhere. Commodity items that the next tab sells at a visible price tolerate far less, and raising those first is how a store concludes that price increases do not work.
Check the unit economics before choosing, too. A product already contributing well can carry a rise comfortably; one contributing almost nothing needs the rise most and is usually the one whose customers are most price-sensitive, which is an uncomfortable pairing worth naming rather than discovering.
| Line | Before | After |
|---|---|---|
| Sticker, incl. VAT | €49,00 | €53,90 |
| Net price | €41,18 | €45,29 |
| Payment fee on the gross | €0,93 | €1,02 |
| Contribution per unit | €15,35 | €19,37 |
Stage it, and keep the history
A single large increase is one experiment with no way back; two smaller ones are two experiments with a decision point between them. Stage the rise, hold each step long enough to read four weeks of orders, and compare against the tolerance line rather than against how the week felt.
Whatever happens, record the change with the date it took effect. Effective dating is what lets you compare the weeks either side honestly, because the old orders keep the old price and the new ones carry the new one, rather than the whole history quietly re-pricing itself the moment somebody edits a field.
How to communicate it, or whether to
Most price rises need no announcement. Changing a price is ordinary commerce, customers rarely hold a mental record of what a product cost last quarter, and an announcement invites a conversation about a decision that has already been made. The exceptions are subscriptions and repeat purchases, where the customer sees both prices, and there the notice is not marketing but basic courtesy.
Where an explanation is warranted, one honest sentence about costs beats a paragraph of apology. What does damage is a rise that arrives alongside a reduction in what the customer receives: smaller portions, slower delivery, a component quietly downgraded. Customers forgive a higher price far more readily than they forgive a quieter one that pretends nothing changed.
What to watch afterwards
- Orders per week against the tolerance line, not revenue, which will rise for a while even if volume falls.
- Conversion rate on the affected products, which moves faster than order counts and warns earlier.
- Return rate, which occasionally rises with price as expectations do.
- Contribution per order, which is the number the whole exercise was for.
- Customer messages about price, counted rather than felt: three complaints is noise, thirty is data.
The version of this that fails
Raising prices to fix a problem that is not about price. A store bleeding through untracked returns, a heavy parcel on a light basket or an ad budget below its break-even ratio will not be rescued by a rise, and may lose the volume without curing anything, which is the pattern behind high revenue and no profit. Diagnose which question you actually have, and if the answer is margin, raise with the tolerance line in hand and the history dated behind you.