How deep should your markdowns be?

There's a moment every merchant knows. A product isn't moving, the stock is sitting there, and you decide to discount it. So you pick a number. Twenty percent feels about right. Maybe 30 if you're impatient.

That number was a guess. And it matters more than almost any other pricing decision you make, because a discount is the one lever that changes your revenue and your margin at the same time, in opposite directions.

The good news: you don't need a model to set it well. You need two pieces of arithmetic that most merchants have never been shown.

First: what a discount actually costs you

Here's the thing that surprises people. A discount doesn't come out of your revenue — it comes out of your margin, and your margin is a much smaller number. So the percentage feels small while the damage is large.

Take a product you sell for $100 that costs you $60. Your gross margin is $40, or 40%.

Now take 10% off. The customer pays $90. You still pay $60. Your margin just went from $40 to $30 — you gave away a quarter of your profit on that item with a discount that looks modest on the label.

Which raises the real question: how many extra units do you need to sell to make up for it?

Not more revenue. The same profit. A third more units sold, more picking, more packing, more shipping, more support — to stand exactly where you started.

And it gets steep fast. Here's the same 40%-margin product at increasing depths:

Read that 30% row again. On a 40% margin product, a 30% discount means you need four times the sales volume to earn the same gross profit. Nobody's discount drives a 4× lift. And at 40% off you're selling at cost — infinite volume still earns you nothing.

Which means margin decides your range, before anything else

The same 20% discount is a routine promotion on one product and a catastrophe on another:

This is why apparel and jewellery brands can run the 40%-off sales that would bankrupt an electronics reseller. It isn't nerve. It's margin structure.

Second: how much time you have

Everything above assumes you're discounting to make more profit. Sometimes you're not — you're discounting to get your cash out of a product that isn't going to sell at full price. That's clearance, and it plays by different rules.

In clearance the question isn't "does this discount pay for itself in volume". It's "what's the least I can discount and still be empty by the date I need to be empty". Which needs one number:

A product with 10 weeks of supply and 12 weeks of season doesn't need a markdown at all. One with 30 weeks of supply and 8 weeks of season needs roughly a tripling of its sales rate, and no polite 10% is going to do that.

Notice how different this is from how markdowns usually get set. The depth follows from the size of the gap between the stock you hold and the time you have to move it — not from what feels normal, and not from what a competitor is doing.

Don't take one deep cut. Build a staircase.

Here's where most clearance goes wrong even when the math is right. A merchant works out they need a big discount, applies 40% off in one move, and sells the whole lot in a weekend.

Selling out fast feels like success. It usually means you left money on the table: if it cleared in three days at 40% off, it would probably have cleared in three weeks at 20% off, and you'd have kept the difference on every unit.

The fix is a phased markdown — a staircase instead of a cliff. You start shallow, hold, watch the sell-through rate, and go deeper only if the rate isn't enough to hit your date. Retail has done this for decades with fixed schedules like 20% off after 30 days and 40% after 60; the principle is that phased liquidation preserves more margin than blanket clearance, because each step is a test of whether a deeper cut was ever needed.

Two things the staircase needs to work

• Enough time. Three steps need weeks, not days. This is the argument for starting markdowns earlier than feels comfortable — early markdowns can be shallow, late ones have to be deep.
• Actually watching the rate. Each step is only a test if you check the result before setting the next one. A staircase you don't monitor is just a slow cliff.

Where the floor is

Every product has a price below which selling it is worse than not selling it. Roughly: what it cost you, plus the pick, pack and ship, plus any payment fees — and if you're selling below that, each additional sale actively costs you money.

With one important exception, and it's the exception that justifies genuinely deep clearance. Inventory that never sells isn't worth its cost — it's worth zero, minus the money you spend storing it. Carrying costs for retail inventory are commonly estimated at 20–30% of inventory value per year, so stock sitting for a year quietly consumes a fifth of its own value in space, capital and handling before you write any of it off.

Against a write-off, a painful discount is the better outcome. A 30% markdown that clears is better than a 100% loss in six months. Just make sure you're actually in that situation rather than assuming you are — which is the subject of the companion piece below.

The short version

• Work out the break-even volume before you commit. Discount ÷ (margin − discount) tells you the extra units you need. If the answer sounds unachievable, it is.
• Your margin sets your ceiling. Deep discounts belong to high-margin categories; on thin margins even 20% off is close to the edge.
• For clearance, let the calendar set the depth. Weeks of supply versus weeks remaining tells you the lift you need.
• Phase it. Shallow first, deeper only if the sell-through rate demands it — so the deepest price touches the fewest units.
• Start earlier than feels comfortable. Time is what lets you be shallow. Waiting is what forces you to be deep.

Read next