Glossary - Products and inventory
Dead stock
Dead stock is inventory that has stopped selling, or never sold, and has no realistic path to selling at a viable price. It is identified by zero or near-zero sales velocity over a defined window, and valued at unit cost. Ecommerce operators track it because it ties up working capital and keeps charging carrying cost.
| Variable | Definition |
|---|---|
| Window | The no-sales period that defines dead for your category, set per category and applied consistently - for example 90, 180 or 365 days. |
| Unit landed cost | Cost plus freight-in plus duty, the same landed basis used in the step-down. Retail price overstates what is actually trapped. |
Worked example
That $8,800 is cash already spent, sitting on shelves instead of in stock that sells. And it keeps charging: storage, insurance and the return the capital could have earned elsewhere accrue for every month it stays.
What is a good dead stock level?
There is no honest benchmark percentage, because the number depends on how you define the window, how seasonal the category is, and your margin structure - a range built on breadth accepts some obsolescence as the cost of offering choice, while a tight evergreen catalog should run close to none. What good looks like in practice is a small, known, actively-cleared amount rather than an unmeasured pile: the value is computed at landed cost on a fixed cadence, each SKU has a clearance route and a deadline, and the trend is flat or falling. The upstream disciplines that keep it that way are honest sell-through tracking and a working open-to-buy plan.
Dead stock vs related metrics
| Metric | What it tells you | How it differs |
|---|---|---|
| Stockout | Demand arriving with no stock to meet it | The opposite failure: margin forfeited rather than capital trapped |
| Sell-through rate | Share of received units sold in a window | The early warning: a weak sell-through this season is dead stock next season |
| Inventory turnover | How many times stock sells through in a period | An aggregate speed; a healthy average can hide a dead tail behind fast movers |
| ABC analysis | Which SKUs carry the value of the range | The classification that shows where the slow tail is forming |
Common mistakes
- Defining dead by age alone. An old SKU still selling steadily is not dead, and a young SKU at zero velocity may already be. Velocity over the window is the test, not the purchase date.
- Ignoring seasonality. A summer SKU flagged in winter is dormant, not dead. Judge seasonal items against their selling season, or the list fills with false positives.
- Anchoring on what you paid. The cost is sunk. The only question is which route recovers the most cash from here - holding out for full price is usually the most expensive choice.
- Clearing without checking the SKU's role. Some slow items win first orders or attach to winners in baskets. Check that before culling, then cull with confidence.
- Counting the write-down but not the carrying cost. The visible loss is the markdown; the quiet one is months of storage, insurance and trapped capital before the decision finally gets made.
FAQ
There is no universal cut-off; it depends on the category's natural purchase cycle. A common approach is a defined no-sales window, for example 90, 180 or 365 days, set per category and applied consistently.
In rough order of recovery: markdowns and clearance, bundling with sellers, gift-with-purchase, selling to liquidators, donating, or disposal. The cost is sunk, so judge each route purely by the cash it recovers from here.
The capital trapped at landed cost, plus ongoing carrying costs - storage, insurance, and the return that cash could earn elsewhere - plus the eventual write-down. The longer it sits, the more the carrying side dominates the total.
Updated July 2026