Two ways to lose money on the same stock
Mark down too early and you discount units that would have sold at full price. Too late and you are clearing at a deeper cut, or carrying stock into a season where it is worth less again.
Most businesses lean one way consistently, driven by whichever mistake was most painful recently. That is not a strategy, and it is visible in the data if anyone looks.
The question is sell-through against time remaining
The core calculation is straightforward: at the current rate of sale, will this stock clear before the end of its selling window? If not, by how much is it short?
- Estimate the remaining selling weeks for the line.
- Predict the sales rate at the current price, using the line's own recent performance and comparable products.
- Project the closing stock position.
- If it will not clear, estimate the uplift needed - and what discount depth has historically produced that uplift for similar products.
- Compare the margin cost of the discount against the expected residual value of unsold stock.
That last comparison is what makes the decision rather than a rule of thumb. If unsold stock has meaningful residual value, waiting is cheaper than it looks; if it will be written off, acting early is worth more.
Staged reductions usually beat one big cut
A single deep markdown clears stock and gives away margin on units that would have moved at a smaller reduction. Successive smaller cuts capture more of the demand curve.
| Approach | Margin captured | Risk |
|---|---|---|
| One deep cut, late | Lowest | Also the most residual stock beforehand |
| Staged reductions | Highest in most cases | Needs monitoring and discipline |
| Early small cut | Good if demand responds | Gives away margin if it would have sold |
Staging requires the discipline to review on a schedule and act. Businesses that set a markdown calendar and stick to it generally do better than those making ad hoc decisions under pressure at the end of a season.
Sizes and variants complicate everything
Aggregate sell-through hides the real position. A line that is 70% sold may have cleared the middle sizes and be sitting on the extremes, which will not clear at any sensible discount.
Decide at the level the customer buys. Marking down a whole line because the broken size curve looks like slow sell-through discounts units that would still have sold at full price to the customers who wanted them.
Measuring whether it worked
Markdown effectiveness is hard to measure because there is no control - you cannot also not mark down. The practical approach is comparison across similar lines treated differently, and consistency in how decisions are recorded.
Record the decision, the reasoning and the state at the time, for every markdown. After a couple of seasons that gives an evidence base connecting decisions to outcomes, which is what turns this from instinct into a process.
The markdown decision is made every week you do not make it.