Backorder: what it means, and what it costs to rely on one
A backorder is an order a customer has already placed for a SKU you do not have in stock, which ships only when your next supplier delivery lands. It is not a plan, it is a gap: the sale happened, the stock did not, and you now owe fulfilment on a date your supplier controls.
What a backorder is
A backorder is an order a customer has placed for a SKU you do not currently hold, which you fulfil later, once replenishment arrives. The money is often already captured. The shelf is empty. The ship date belongs to your supplier, not to you.
Backorders arrive in two ways, and they are not the same thing.
- On purpose. You let the store keep selling past zero because you know a delivery is close, and you tell the buyer when it ships.
- By accident. Nobody was watching the SKU, it ran down, and orders kept landing. This is the common case in a small store, and the one that hurts.
Why they happen
A backorder is a lead-time problem wearing a customer-service costume. If your supplier takes three weeks door to door and you only send the purchase order the day stock hits zero, three weeks of demand has nowhere to go. Either you owe units to people who have already paid, or you turn that demand away at the product page.
The arithmetic that moves the order date earlier is not on this page. Two siblings in this pillar carry it: the reorder point formula worked through on one SKU, and the walkthrough that turns a velocity figure and a lead time into an actual order date.
What leaning on backorders actually costs
There is no single number here, and any store that quotes you one is guessing. What there is, reliably, is a set of costs that show up in different places on your books.
Cash timing goes backwards. You hold money for goods you have not received. If enough buyers cancel, you are refunding against a supplier invoice you have already committed to.
Support volume rises per order. Every backordered order is a thread: where is it, when does it ship, can I cancel. One SKU out of stock for three weeks can generate more inbox work than the margin on the order.
Some buyers do not wait. A customer who wanted the candle this week may buy it elsewhere this week. You cannot measure what that costs you precisely, which is exactly why it is dangerous to treat backorders as neutral.
Your own numbers get distorted. If you close the SKU at zero instead of backordering, sales for those days read as zero demand. A velocity window that includes a dead fortnight reports a lower units-per-day than the SKU really sells, which pulls the reorder point down and sets you up to run out again. Suppressed demand looks identical to weak demand in a sales report.
Backorder, stockout, pre-order and oversell
| Term | What it means | What the buyer experiences |
|---|---|---|
| Backorder | An order taken for stock you do not hold, fulfilled when replenishment lands | Paid, waiting, ship date uncertain |
| Stockout | The SKU is at zero and unavailable, so the sale does not happen at all | Sees out of stock, buys elsewhere or waits |
| Pre-order | An order taken before first stock exists, usually with a stated release date | Expects to wait, told so up front |
| Oversell | More units sold than you actually have, usually a counting or sync error | Thinks the item is in stock, then gets an apology |
Related terms
A backorder is a symptom. The terms that describe the cause sit one step earlier in the maths. Each term below has exactly one entry in this glossary, so follow the link rather than hunting the index for a second version of the same page.
- Reorder point. The stock level at which you place the order so the delivery beats zero. Get this right and backorders mostly disappear.
- Lead time. How many days your supplier actually takes, door to door, including the days the purchase order sits unopened. Only you can supply this number.
- Sales velocity. Units per day over a chosen window, the other half of the reorder point. The window you pick changes the answer.
- Days of inventory on hand. Current stock divided by velocity, the runway number that tells you how long you have before the shelf is empty. One page, one formula, one view of what a healthy number looks like.
- Safety stock. The buffer you hold for the weeks demand spikes or the supplier slips.
- Stockout. What a backorder is trying to paper over, and what it costs a small store.
The worked order-date calculation sits with the when-to-reorder walkthrough, not here.
Questions
Is a backorder the same as a stockout?
No. In a stockout the sale does not happen: the SKU shows unavailable and the buyer leaves. In a backorder the sale happens and the fulfilment is deferred. Both start at the same place, zero stock, and both trace back to ordering too late.
What is the difference between a backorder and a pre-order?
A pre-order is taken before the item has ever been in stock, usually with a release date you have set and communicated. A backorder is taken on an item you normally stock and have run out of, with a ship date that depends on your supplier. The buyer's tolerance for waiting is very different in each case.
Should I let my store sell past zero?
It can make sense when the delivery date is known and close, and when you say so clearly on the product page and in the order confirmation. It makes much less sense as a default, because a SKU nobody is watching can accumulate weeks of owed units before anyone notices.
How long is too long for a backorder?
There is no universal figure. A useful test is your supplier lead time: if the wait a customer faces is longer than the lead time you could have planned around, the backorder is covering for a late order rather than managing a known gap.
How do I stop relying on backorders?
Order while stock still covers the supplier wait, rather than when stock hits zero. The reorder point and lead time pages in this pillar walk that arithmetic through on a single SKU.
Do backorders distort my sales history?
They can distort it either way. Selling past zero keeps recording demand, which keeps velocity honest. Closing the SKU at zero records no sales for those days, so a 30 day velocity window reads a lower units-per-day than the product really sells, and the reorder point it feeds comes out too low.