Key takeaways
- Reorder point = average daily demand × lead time in days + safety stock.
- Lead time demand covers normal sales while you wait for the order; safety stock covers the surprises.
- A simple safety stock is (max daily sales × max lead time) − (average daily sales × average lead time); a statistical one uses a service level and the variability of demand.
- The reorder point says when to order; economic order quantity (EOQ) says how much.
- Recalculate at least every season, and whenever a supplier's lead time or your sales rate changes.
What a reorder point is
Every item you stock takes time to replace. Between placing an order and receiving it, you keep selling. The reorder point is the stock level at which you need to order, so the new stock lands just as the old stock runs low.
Order too late and you stock out: lost sales, rushed shipping, unhappy customers and, on marketplaces, possible ranking or account penalties. Order too early and cash sits on shelves, storage costs climb and stock risks going out of date.
Reorder point = (Average daily demand × Lead time in days) + Safety stock
Free toolReorder point calculator
The two parts: lead time demand and safety stock
Lead time demand is what you expect to sell while waiting for an order. If you sell 40 units a day and the supplier takes 10 days from order to arrival, you need 400 units on hand when you order.
Lead time should be the full time from deciding to order until stock is ready to sell: approving the purchase order, supplier processing, production, transit, customs and receiving. Using only the shipping time is a common reason reorder points come out too low.
Safety stock is a buffer for when demand runs higher or the supplier runs slower than average. Without it, you run short in roughly half of your order cycles: every time demand during the wait comes in above average. How big it should be depends on how variable your sales and lead times are, and how costly a stockout is.
Three ways to set safety stock
Pick the method that matches your data. A small shop with a spreadsheet can start with the first or second; the statistical method needs a few months of daily or weekly sales history.
- Fixed days of cover: hold a set number of days of average sales, such as 5 days × 40 units = 200 units. Easy, but ignores how variable the item is.
- Max minus average: (maximum daily sales × maximum lead time) − (average daily sales × average lead time). Uses your worst recent case, which often makes it generous.
- Statistical: Z × standard deviation of daily demand × √(lead time in days). Ties the buffer to a target service level, the share of order cycles with no stockout.
Max minus average: SS = (Max daily sales × Max lead time) − (Avg daily sales × Avg lead time) Statistical: SS = Z × σ(daily demand) × √(Lead time in days)
Worked example
A store sells an average of 40 units a day of one item. Over the last three months, its busiest day was 55 units and the standard deviation of daily sales was 12 units. The supplier averages 10 days lead time; the longest recent delivery took 14 days.
Lead time demand is 40 × 10 = 400 units.
With the max minus average method, safety stock is (55 × 14) − (40 × 10) = 770 − 400 = 370 units, and the reorder point is 770 units.
With the statistical method at a 95% service level (Z = 1.65), safety stock is 1.65 × 12 × √10 = 1.65 × 12 × 3.162 ≈ 62.6, rounded up to 63 units. The reorder point is 400 + 63 = 463 units.
The gap is large because the max minus average method assumes the busiest days and the slowest delivery happen together. The statistical figure assumes lead time is steady. If lead time also varies, use the combined formula below.
| Method | Safety stock | Reorder point |
|---|---|---|
| No safety stock | 0 | 400 |
| Fixed 5 days of cover | 200 | 600 |
| Max minus average | 370 | 770 |
| Statistical, 95% service level | 63 | 463 |
| Statistical, 95%, lead time varies (σ = 2 days) | 146 | 546 |
Choosing a service level
The Z value comes from the normal distribution. A higher service level means fewer stockouts but more stock, and the cost rises steeply near 100%. Going from 95% to 99% raises safety stock by about 41% in the formula above, while going from 90% to 95% raises it by about 29%.
Many businesses set higher levels for best sellers and items where a stockout loses the customer, and lower levels for slow or easily substituted items.
When lead time varies as well as demand, a common extension is SS = Z × √(L × σd² + d² × σL²), where L is average lead time, σd the standard deviation of daily demand, d average daily demand and σL the standard deviation of lead time. In the example, with σL = 2 days: √(10 × 144 + 1,600 × 4) = √7,840 ≈ 88.5, × 1.65 ≈ 146 units.
| Service level | Z value | Safety stock in the example |
|---|---|---|
| 90% | 1.28 | 49 |
| 95% | 1.65 | 63 |
| 97.5% | 1.96 | 75 |
| 99% | 2.33 | 89 |
How reorder point and EOQ work together
The reorder point answers when to order. Economic order quantity answers how much. EOQ balances the cost of placing orders against the cost of holding stock.
In the example, annual demand is 40 × 365 = 14,600 units. If each order costs $50 to place and receive, and holding one unit for a year costs $2, EOQ = √(2 × 14,600 × $50 ÷ $2) = √730,000 ≈ 854 units. That works out to about 17 orders a year, one roughly every three weeks.
So the rule for this item becomes: when stock falls to 463 units, order about 854. Average inventory is roughly half the order quantity plus safety stock, about 427 + 63 = 490 units.
Round the EOQ to the supplier's case packs or minimum order quantity. The total cost curve is flat near the optimum, so rounding a little either way changes very little.
EOQ = √(2 × Annual demand × Cost per order ÷ Holding cost per unit per year)
Free toolEOQ calculatorKeeping reorder points accurate
A reorder point is only as good as its inputs. Sales rates shift with seasons and promotions, and supplier lead times drift.
- Use recent sales, and exclude days when you were out of stock, which hide real demand.
- Raise reorder points ahead of known peaks such as holidays or a planned promotion.
- Track actual lead time on every purchase order: order date to date received.
- Count stock regularly. A reorder point fails if the system thinks you have stock you do not.
- Review slow movers separately; a reorder point built for a fast seller can overstock a slow one.
- Watch inventory turnover to check the buffer is not tying up too much cash.
