Free tool · No sign-up
Safety stock calculator
Work out how much buffer stock to hold for a SKU from your target service level, demand variability, and supplier lead time. Runs in your browser; nothing is sent or stored.
Result
–
units of safety stock
- Z-score (service factor)
- –
- Demand during lead time
- –
- Reorder point
- –
- Safety stock covers
- –
Need this for every SKU, every month? Varox plans safety stock across your catalog for free.
Safety stock formula
The calculator uses the standard formula that accounts for both demand variability and lead time variability:
Safety stock = Z × √( L × σd² + d² × σL² )
- Z is the service factor for your target service level (1.645 for 95%).
- L is the average lead time, expressed in demand periods.
- d and σd are the average and standard deviation of demand per period.
- σL is the standard deviation of lead time, in the same periods.
If lead time never varies (σL = 0), the formula simplifies to the version most textbooks start with:
Safety stock = Z × σd × √L
The max–average method needs no statistics. It assumes your worst demand and your worst lead time arrive together:
Safety stock = (max demand × max lead time) − (average demand × average lead time)
It is easy to explain, but it ignores how unlikely that worst case is, so it usually holds far more stock than the target service level needs.
Service level and Z-score
The service level here is the cycle service level: the probability that you do not run out between placing an order and receiving it. The Z-score is the matching point on the normal distribution.
| Service level | Z-score |
|---|---|
| 80% | 0.84 |
| 90% | 1.28 |
| 95% | 1.64 |
| 97.5% | 1.96 |
| 98% | 2.05 |
| 99% | 2.33 |
| 99.9% | 3.09 |
The curve gets steep at the top. Going from 95% to 99% raises safety stock by about 41% (2.33 ÷ 1.645), which is why few teams set 99% for every SKU.
Worked example
A SKU sells an average of 40 units a day with a standard deviation of 12. The supplier's lead time averages 14 days, give or take 3 days, and you target a 95% service level (Z = 1.645).
- Fixed lead time: 1.645 × 12 × √14 = 74 units.
- With lead time variability: 1.645 × √(14 × 12² + 40² × 3²) = 1.645 × √16,416 = 211 units.
- Max–average with a worst day of 60 units and a worst lead time of 20 days: 60 × 20 − 40 × 14 = 640 units.
Two lessons from the same SKU: three days of supplier unreliability nearly triples the buffer, and the max–average rule holds three times the stock of the statistical method for the same item. The reorder point for the 211-unit case is 40 × 14 + 211 = 771 units; see the reorder point calculator.
Where the inputs come from
- Average demand and its standard deviation: from 6–12 periods of sales history in the same unit. If your history is monthly, use monthly figures and set the demand period to month.
- Leave out stockout periods. A week with zero stock records zero sales, not zero demand. Counting it lowers the average and inflates the variability.
- Lead time and its variability: from the order date and the receipt date of recent purchase orders for the supplier.
- Forecast error: if you plan from a forecast, replace the standard deviation of demand with the standard deviation of forecast error.
Common mistakes
- Mixing periods, such as monthly standard deviation with a lead time in days. The calculator converts lead time for you; spreadsheets often do not.
- Using one service level for the whole catalog instead of, for example, 98% for A items, 95% for B, and 90% for C.
- Ignoring lead time variability, which is often the larger term, as the example shows.
- Setting the buffer once and never revisiting it after demand, suppliers, or seasons change.
Safety stock across a whole catalog
A calculator works for one SKU. In Varox, safety stock is set as months of average demand for each ABCD class, and the Delivery Planner keeps every SKU's projected stock above next month's forecast plus that buffer, month by month, with lead times and pack sizes applied.
Plan safety stock for every SKU
Upload sales and stock from Excel and get a forecast and delivery plan with buffers applied. Free for up to 10,000 active SKUs.
Questions
What is a good safety stock level?
There is no single right number. It depends on how much demand and lead time vary and how often you are willing to run out. A 95% cycle service level is a common default; many teams use higher levels for their best sellers and lower ones for the long tail.
What is the difference between safety stock and reorder point?
Safety stock is the buffer held for variability. The reorder point is the stock level that triggers a new order: expected demand during the lead time plus the safety stock.
Should I use forecast error or demand variability?
If you plan from a forecast, use the standard deviation of forecast error (actual minus forecast) instead of the standard deviation of demand. A good forecast explains part of the variation, so the buffer only needs to cover what the forecast misses.
How often should I recalculate safety stock?
Monthly or quarterly for most catalogs, and whenever a supplier changes lead time or a product enters a new season. Stale inputs are a bigger source of error than the choice of formula.
Why does my safety stock look too high?
Usually because lead time variability is large, because the service level is close to 100% (the Z-score grows fast above 98%), or because demand and lead time are in different units. Check the periods first.