01
What holding actually costs
An unsold item feels free. It is already paid for, it is not costing anything today, and it might still sell. That intuition is wrong in a specific and expensive way, because the cost of holding is not an expense you pay — it is a return you do not receive.
- Opportunity cost. The capital in that item cannot buy the fast-turning item you saw last week. If your working stock turns several times a year, every unit tied up in dead stock is several missed trades, not one.
- Space. Shelving is finite, and full shelves make you decline good lots. Sellers routinely stop sourcing because they are "out of room" while holding stock they will never sell.
- Attention. Aged stock generates repeated small decisions — should I relist, should I reprice, should I move it — that consume the judgement you need for buying.
- Depreciation. Most categories drift downward. Electronics, fashion, and anything with a successor model lose value on a schedule that does not care about your cost basis.
- Condition decay. Boxes soften, sunlight fades, batteries corrode, rubber perishes. An item stored for two years is frequently a worse item than the one you bought.
- Self-deception. Inventory valued at hopeful prices makes the business look healthier than it is, which delays exactly the decisions that would fix it.
02
Identifying aged stock before it is dead
Dead stock is rarely a surprise. It announces itself early and is ignored, because the signals are individually easy to explain away. The fix is to make ageing visible as a routine number rather than a feeling.
- Record a listed date on everything The single most useful field in a reseller's records after cost. Without it, age is a guess, and guesses always run young — every seller believes their stock is newer than it is.
- Define an expected sell-through window per category Not one number for the whole business. Fast fashion and rare collectables have legitimately different clocks, and applying a single threshold to both produces either constant false alarms or none at all.
- Review by age band, monthly Group stock into bands — recent, ageing, overdue, long overdue — and look at the shape rather than individual items. A growing overdue band is the earliest reliable warning that your buying has drifted.
- Watch views, not just sales An item with steady views and no sales has a price or description problem and is recoverable. An item with almost no views has a visibility or demand problem, and repricing will not fix it. These need opposite responses, which is why treating all unsold stock the same fails.
- Track the ratio, not the count Aged units as a share of total stock. In a growing business the absolute count of old items rises naturally; the percentage is what tells you whether the problem is getting worse.
This is the point at which spreadsheet-based tracking usually fails, because age bands require a reliable listed date on every unit and a view across every channel at once. If your inventory lives in several places, ageing is effectively invisible — see migrating from spreadsheets.
03
Diagnose before you discount
The reflex response to an unsold item is to cut the price. Sometimes that is right. Frequently it destroys margin on an item whose problem was never the price, and the discount teaches you nothing because you never learned why it was not selling.
| Symptom | Likely cause | Right response |
|---|---|---|
| Almost no views | The listing is not being found: weak title, missing attributes, wrong category. | Rewrite the title and complete the item specifics. Price is irrelevant to an item nobody sees. |
| Views, no watchers | The thumbnail or the first photograph is losing the click. | Reshoot the lead image. This is the highest-return fix in the table. |
| Watchers, no offers or sales | People want it at a different price, or something in the description worries them. | A targeted offer to watchers, or a clearer description of the concern. |
| Views and watchers, competitors selling | You are genuinely priced above the market. | Reprice to the market, and revisit what you paid. |
| No demand anywhere on any channel | The market for this is thin or gone. | Stop optimising. Move to the liquidation ladder below. |
Diagnosis is also worth more than the individual item, because the same defect is usually present across a whole batch listed in the same session. Fixing one title often reveals two hundred listings with the same weakness. This is covered in more depth in when listings do not sell.
04
The liquidation ladder
When an item is genuinely dead, work down a ladder of escalating concessions rather than jumping straight to the bottom. Each rung recovers less capital but is faster and more certain than the one above it, and moving down on a schedule prevents the indefinite drift that is the real cost.
- Relist properly. New photographs, a rewritten title, complete attributes, correct category. Treat it as a new item that happens to already be in the building.
- Move channels. An item invisible in a general marketplace can be obvious to a specialist audience, and vice versa. This is frequently the rung that works. See choosing where to sell.
- Reprice to the market. Not to your cost. What the item is worth is entirely unrelated to what you paid, and holding out for your cost basis is the single most expensive habit in this whole subject.
- Offer it actively. Send offers to watchers, run it in a promotion, or accept a lower best offer you would previously have declined.
- Bundle. Combine slow items into a lot that is worth someone's postage and attention. Bundling is very effective for the long tail of small parts and accessories.
- Auction it with a low start. The market decides. This is uncomfortable and it works, particularly for items where you have lost confidence in your own valuation.
- Sell it to another reseller as a job lot. Recovering a fraction quickly beats recovering nothing slowly, and someone whose specialism it is will pay more for the pile than the market pays you per item.
- Donate or dispose, and record it. Space and attention have value. Writing the loss off deliberately is a decision; leaving it on a shelf is the same loss without the benefit.
05
Making it automatic rather than emotional
Individual liquidation decisions are painful because each one is an admission that a purchase was wrong. Made as a policy in advance, they stop being personal and start being maintenance.
- Set an age threshold per category at which an item is automatically reviewed. The review is mandatory; the outcome is not.
- Set a second, longer threshold at which the item must move down at least one rung of the ladder, whatever you feel about it.
- Run a clearance pass on a fixed schedule — quarterly works for most sellers — and give it a target, such as clearing a set share of the overdue band.
- Cap the total capital allowed in the long-overdue band. When it is exceeded, no new sourcing in that category until it comes back down.
- Record every liquidation with its original cost and the reason it failed. This is the file that improves your buying.
A quarterly clearance also has a pleasant second-order effect: it produces cash and empty space at a predictable time, which is precisely when a good bulk opportunity becomes affordable. Sellers who clear on a schedule find they can act on lots that sellers holding dead stock have to decline.
06
Preventing the next batch
Dead stock is a buying diagnosis, not a selling one. Every dead item was purchased, usually for a reason that seemed sound, and the reasons repeat.
- Review liquidations by category, not individually One failed item is bad luck. Eleven from the same category is a policy error, and it will be invisible unless you look at the group.
- Check whether you bought demand or scarcity Dead stock is disproportionately made of things that were rare rather than wanted. Sell-through rate at the point of purchase is the check that catches this — see valuing inventory before you buy.
- Look for the depth pattern Buying five of something that sells one a year is a different error from buying one bad item, and it is much more expensive. Multiples require evidence of repeat demand, not just a good unit price.
- Be honest about affection Categories you personally love are over-represented in most sellers' dead stock, because expertise and enthusiasm are hard to separate. If a category has never turned a profit, the fact that you enjoy it is a reason to collect it, not to trade it.
- Recompute your real margin including the failures A category's profitability must be measured across everything you bought in it, including the units that never sold. Counting only the winners makes almost every category look good and is the most common way sellers keep repeating an unprofitable line.
That last calculation is the one that changes behaviour, and it requires cost and outcome recorded against every unit including the ones you wrote off. It is the practical reason to keep proper records rather than an accounting formality — see inventory accounting and COGS.
07
Practice
Exercise
Total the capital you have parked
- Sort your active listings by the date you listed them.
- Count everything older than 90 days and total what you paid for it.
- That number is capital you have already spent and not recovered.
- Take the oldest ten and decide today: relist, reprice, or clear. Do not defer any of them.
Check yourself
An item has been listed for eight months. Should you discount it?
Not until you know why it has not sold. No impressions is a category or specifics problem, and discounting an invisible listing changes nothing except your margin. Impressions without conversion is where price becomes a plausible cause. Diagnose first — the discount is the last step, not the first.
Why is aged stock so hard to see?
Because it has no line in your accounts. The capital, the space, and the attention it absorbs are real costs that never appear as a transaction, so nothing prompts you to act. That is why the ageing view has to be built deliberately — a reliable listed date on every unit, visible across every channel at once — and why the clearance schedule should be set in advance rather than at the moment you finally notice.
Progress is saved in this browser only. No account, nothing sent anywhere.
08
Common questions
At what age is stock officially dead?
There is no universal number, and adopting someone else's will mislead you. Set the threshold from your own category data: if most items that sell do so within a certain window, an item well past that window is behaving differently from your normal stock and deserves a decision. The useful discipline is having a threshold at all, not the specific figure.
Should I sell at a loss?
If the alternative is a larger loss, yes. Recovering part of your capital now and redeploying it into stock that turns will usually outperform holding for a full-price sale that may not arrive, once you account for the trades you cannot make while the money is tied up. A recorded loss also reduces taxable profit, so the after-tax gap is smaller than it appears — check the treatment in your jurisdiction with an accountant.
Does relisting an old item actually help?
It helps when the item genuinely changes — new photographs, a rewritten title, complete attributes, a different category. Recycling the same listing with a fresh date does very little on most platforms and burns the time you could have spent diagnosing the real problem. Change something that affects whether buyers find it or trust it.
How do I decide between discounting and bundling?
Discount when the item has demand at a lower price. Bundle when the item has almost no demand at any price but has value as part of something larger — small parts, accessories, incomplete sets, and the long tail of cheap items where postage exceeds the value of any single piece.
Is holding ever the right answer?
Occasionally, for genuinely appreciating categories where you have evidence rather than hope, and where the capital involved is small enough not to constrain your trading. Treat it as an explicit investment decision with a review date, kept separate from your working stock — not as the default outcome for anything that failed to sell.