Blog · 23 September 2026
Automatic ordering and out-of-stocks: how to set the minimum stock and the reorder point?
In shortAutomatic ordering fires when book stock falls below the reorder point, and orders up to the maximum stock. The reorder point must cover sales during the delivery lead time, plus a safety stock: reorder point = daily sales × lead time in days + safety stock. A reference selling 20 units a day, delivered in 2 days, with one day of safety, has a reorder point of 60. Set on the yearly average, that reorder point is wrong on Saturday and on promotion; set against wrong book stock, it never fires.
What automatic ordering really does
Store inventory systems order on their own, according to three parameters per reference:
- the reorder point (or minimum stock): the book stock level below which an order goes out;
- the maximum stock (or target quantity): the level the order tries to reach;
- the lead time: the time between the order and the product reaching the shelf.
Nothing else. The system sees neither the shelf, nor Saturday, nor the promotion, unless they have been described to it. According to Gruen, Corsten and Bharadwaj, in-store ordering and forecasting are the first cause of out-of-stocks, ahead of replenishment: setting these three parameters is therefore a direct lever.
The reorder point formula
Reorder point = average daily sales × lead time (in days) + safety stock.
Safety stock covers variations: one day of sales is a reasonable minimum for a product with regular sales, two days for a product whose sales vary a lot.
| Reference | Sales per day | Lead time | Safety | Reorder point |
|---|---|---|---|---|
| Semi-skimmed milk 1 L | 200 | 1 day | 1 day (200) | 400 |
| Flagship biscuits | 40 | 2 days | 1 day (40) | 120 |
| Common hair colour shade | 2 | 3 days | 2 days (4) | 10 |
The maximum stock is then set on storage capacity (shelf plus back room) and on the desired ordering frequency: ordering every two days implies a maximum of at least two days of sales above the reorder point.
The four mistakes that make automatic ordering useless
- Setting on the yearly average. The reference’s sales vary by weekday and by season. A reorder point calculated on the average is too low on Saturday and before holidays, too high in January. Recent systems accept weekday profiles; otherwise, the twenty best sellers deserve a manual setting before each peak period (Saturday and pre-holiday out-of-stocks).
- Ignoring the promotion. A promotion multiplies sales by three to ten; without raising the reorder point during the operation, the automatic order goes out too late (promotions and out-of-stocks).
- Trusting wrong book stock. The reorder point compares against book stock. If book stock is overstated, the order does not go out, and the out-of-stock lasts until the stocktake. That is phantom stock, covered in phantom stock: what to do when the book stock says the product is there.
- Forgetting the real lead time. The lead time to enter is not the carrier’s but the time between the order and the shelf: receiving, checking and replenishment included. A day forgotten here is a day of out-of-stock per cycle.
How to check your settings
Once a month, pull the fifty best-selling references of each aisle with their reorder point, their real daily sales of the last four weeks and their lead time. Any reference whose reorder point is below sales × lead time is an out-of-stock scheduled every cycle. This review generally reveals reorder points never revised since the reference was created, while its sales have doubled.
What shopper reporting changes
A product reported missing by shoppers while automatic ordering triggered nothing is a wrong parameter or phantom stock. ShelfAlert gives that list every day; it points directly at the references to review. 14-day trial, no payment card: see the plan for store managers.
Sources
- Gruen, Corsten, Bharadwaj, “Retail Out-of-Stocks: A Worldwide Examination”, GMA, 2002 (share of ordering and forecasting causes): full study (PDF).
- DeHoratius, Raman, “Inventory Record Inaccuracy: An Empirical Analysis”, Management Science, 2008: abstract.