Dead stock: prevent it, clear it, learn from it
Dead stock is cash frozen on a shelf, and it gets more expensive the longer it sits.
The move is to define 'dead' by your real sell-through cycle, count the full carrying cost, clear it down a ladder from markdown to liquidation, then root-cause the buy so you don't repeat it. AI can flag the dead SKUs and do the carrying-cost math from your export; you decide how hard to clear.
Fastest path: one prompt, end to end
🤖 AI prompt — paste into ChatGPT / Claude
You are an inventory analyst. Use MY data only.
I will paste my current stock: SKU, units on hand, unit cost, last-sold date, normal sell-through cycle.
My monthly storage cost (or cost per unit of space): [amount]
My contribution margin %: [%]
Do this:
1. Define 'dead' relative to each item's cycle, not a flat number: e.g. no sale in 2 to 3x its normal sell-through window. List which SKUs qualify and how much cash they tie up (units times unit cost).
2. Estimate the carrying cost of holding it: storage + the opportunity cost of the frozen cash + the attention it steals. Give a rough monthly bleed so I feel the urgency.
3. Build a clearance ladder: step 1 markdown, step 2 bundle it with a mover, step 3 flash sale, step 4 donate or liquidate. Give a trigger (time or units left) to advance each step.
4. Root-cause the buy: for the worst offenders, ask what went wrong (over-ordered, wrong variant mix, seasonal miss, bad forecast) so I fix the decision, not just the pile.
Do not guess any number; use what I paste and ask if something is missing.
Output: dead-SKU list with cash tied up + carrying cost + clearance ladder + root-cause questions.
Or do it in 4 steps
- Define 'dead' by cycle, not a calendar. An item that normally sells through in 30 days and hasn't moved in 90 is dead; a slow-by-nature item on a 6-month cycle isn't. Flag anything past 2 to 3x its normal window so you catch real problems, not naturally slow sellers.
- Price in the full carrying cost. Dead stock costs more than its shelf: it's frozen cash you could reinvest, plus storage, plus the attention it steals from movers. Put a monthly number on the bleed; seeing 'this pile costs me X a month' turns 'maybe later' into 'clear it now'.
- Clear it down a ladder, don't fire-sale first. Step down deliberately: a modest markdown, then bundle it with a fast mover, then a time-boxed flash sale, and only then donate or liquidate. Each rung recovers less, so start high and drop only when a trigger (weeks passed, units left) says to.
- Root-cause the buy after. Once it's cleared, ask why you bought it: over-ordered, wrong sizes, chased a trend too late, trusted a bad forecast. Fixing the buying decision is what stops the next pile; clearing the current one only treats the symptom.
Worked example (labeled): 200 units at 12 unit cost, normally sells through in 30 days, zero sales in 90 = dead, with 2,400 in cash frozen and roughly 80 a month in storage plus opportunity cost bleeding out.
Ladder: week 1 mark down 20 percent, week 3 bundle one free with a best-seller, week 5 flash sale at 50 percent, week 7 donate the rest for the tax write-off and the shelf space.
Root cause: over-ordered a single color on a trend that had already peaked, so next buy splits across variants and orders smaller first.
Define dead by cycle, feel the carrying cost, step down the ladder, and fix the buy; for the catalog-level 'should this product exist at all' call, use the when-to-kill-a-product note under store & product.