Guide
How to stop overselling — and stop running out at the same time
Updated 3 min read
The short answer
Overselling and stockouts both come from one cause: the number you advertise does not match what you can actually ship. Fix it with a single stock ledger, stock reserved at order rather than at dispatch, a small buffer on fast movers, and reorder points calculated from real lead time and demand rather than instinct.
- Oversell = advertised more than you hold. Stockout = held less than you could sell
- Reorder point ≈ (average daily sales × lead time in days) + safety stock
- Cancellations damage marketplace performance metrics, not just that one order
- Reserve at order confirmation; deduct at dispatch
Two symptoms, one disease
Sellers usually treat overselling and stockouts as opposites and swing between them — oversell once, panic, hold huge buffers, then sit on dead stock. They are the same failure: the advertised number and the shippable number disagree. Fix the number and both symptoms fall together.
The structural fixes, in priority order
- 1
One ledger, no exceptions
Every channel reads from the same stock ledger. If any seller panel is edited directly, that channel becomes a second source of truth and the problem returns immediately.
- 2
Reserve at order, deduct at dispatch
The moment an order is confirmed, its units stop being available to anyone else — even though they have not physically left. Systems that only decrement at dispatch will oversell during the window in between, which is exactly when a sale event floods you.
- 3
Buffer the fast movers only
Hold back a couple of units on SKUs that sell several times an hour. Applying a blanket buffer to your whole catalogue hides sellable stock on the slow items where lag never mattered.
- 4
Model bundles as component consumption
A combo pack draws down its component SKUs. Treating a bundle as independent stock is one of the most common silent oversell causes.
- 5
Set reorder points from data
Reorder point ≈ (average daily sales × supplier lead time in days) + safety stock. Recalculate monthly, and separately for your festive season, when both sides of that equation change.
- 6
Cycle count continuously
Count a small slice of SKUs weekly rather than everything annually. Small discrepancies are traceable; a year of accumulated drift is not.
What each failure actually costs
| Failure | Immediate cost | The cost people miss |
|---|---|---|
| Oversell then cancel | The order value | Marketplace performance metrics, search ranking and buy-box eligibility |
| Stockout | The lost sale | Listing momentum, and the customer who buys from a competitor and stays there |
| Over-buffering | Nothing visible | Sales lost quietly on every channel, on every SKU, forever |
| Dead stock | Cash tied up | Storage fees, and eventual write-down at a discount |
Frequently asked questions
What causes overselling on marketplaces?
Almost always more than one source of truth for stock — each channel holding its own quantity — or stock being deducted only at dispatch, so orders confirmed but not yet shipped still appear available. Bundles that are not modelled as component consumption are the third common cause.
How do I calculate a reorder point?
Multiply average daily sales by supplier lead time in days, then add safety stock. Recalculate monthly, and separately for festive periods when demand rises and supplier lead times usually lengthen at the same time.
Does cancelling an oversold order hurt my seller account?
Yes. On most Indian marketplaces, seller-initiated cancellations feed a performance metric that affects listing visibility and buy-box eligibility, so the damage typically outlasts and outweighs the single lost order.