Log in with GoogleStandard deviation
Needs a recent run of sales to measure the spread, and a service level you pick. It sets the stockout risk explicitly. Use it for products with a steady sales history and a fairly consistent supplier.
Free calculator
Safety stock calculator
Work out how much extra stock to hold so a sales spike or a late delivery does not cause a stockout. Use the standard deviation formula, which sets the risk you accept through a service level, or the simpler max minus average method.
Your results
Hold 40 units of safety stock on top of the demand you expect during the lead time.
- Safety stock
- 40
- z × standard deviation of daily demand × √lead time
- z-score
- 1.645
- From the service level (table below)
- Days of cover
- 2
- Safety stock ÷ average daily demand
- Demand during lead time
- 320
- Reorder point with this safety stock
- 360
Safety stock is rounded up to whole units. The standard deviation method assumes demand is roughly normally distributed and the lead time is fixed. Reorder point and days of cover need an average daily demand.
The start values are illustrative data.
How it works
Both methods, worked through for the start values.
Illustrative data. An example product sells 20 units a day on average, with a standard deviation of 6 units. Deliveries take 16 days on average and 20 days at their slowest. The best day sold 30 units. The service level is 95%.
Standard deviation method
| Step | Calculation | Result |
|---|---|---|
| z-score | 95% service level | 1.645 |
| Square root of lead time | √16 days | 4 |
| Safety stock | 1.645 × 6 × 4 | 39.5 units |
| Rounded up | 39.5 → whole units | 40 units |
| Days of cover | 40 ÷ 20 units a day | 2 |
| Reorder point | 20 × 16 + 40 | 360 units |
At a 95% service level, about 19 in 20 replenishment cycles end without a stockout. Hold 40 units of safety stock: 2 days of sales.
Max minus average method
| Step | Calculation | Result |
|---|---|---|
| Worst case | 30 units a day × 20 days | 600 units |
| Normal case | 20 units a day × 16 days | 320 units |
| Safety stock | 600 − 320 | 280 units |
| Days of cover | 280 ÷ 20 units a day | 14 |
| Reorder point | 20 × 16 + 280 | 600 units |
This method holds 280 units against 40 units. It covers your busiest day meeting your slowest delivery, which is rare, so it holds more than a 95% service level needs. Read it as an upper bound, or use it when you have too little history to measure a standard deviation.
Service level and z-score
A higher service level needs a higher z, and more safety stock.
The service level is the chance of getting through a replenishment cycle without a stockout. z is how many standard deviations of lead-time demand that chance needs.
| Service level | z-score | Safety stock |
|---|---|---|
| 90% | 1.282 | 31 units |
| 95% | 1.645 | 40 units |
| 97.5% | 1.960 | 48 units |
| 99% | 2.326 | 56 units |
| 99.9% | 3.090 | 75 units |
Safety stock is shown for the start values (illustrative data). Going from 95% to 99% adds 16 units.
Which method
Choose by the data you have, and the risk you accept.
Max minus average
Needs only your highest sales day and your longest delivery. It has no service level: it assumes the worst demand meets the worst delivery, so it holds more stock. Use it when history is short or a stockout is very costly.
Seasonal and new products
A quiet month understates peak demand. Measure the spread over a comparable period, and check it again when sales change.
Varying lead times
The standard deviation formula treats lead time as fixed. If deliveries vary a lot, enter a slow but realistic lead time, or use the max minus average method.
This calculator works on one product at a time. Inventory Insights projects demand per product and works back from supplier lead time and a safety buffer to a reorder date, and warns you when that product’s reorder window opens.
Safety stock questions, answered.
What is the safety stock formula?
The common formula is z × σ × √lead time. z comes from your service level, σ is the standard deviation of daily demand, and lead time is in days. With a service level of 95% (z = 1.645), a σ of 6 units and a 16-day lead time: 1.645 × 6 × 4 = 39.5, rounded up to 40 units.
How do you calculate safety stock without statistics?
Use max minus average: multiply your highest daily sales by your longest lead time, then subtract average daily sales times average lead time. 30 × 20 − 20 × 16 = 280 units. It needs no standard deviation, but it holds more stock.
What service level should I choose?
There is no standard answer: it is a trade-off. Each step up costs more stock than the one before. At the start values, 95% needs 40 units and 99% needs 56 units. Compare what extra units cost to hold with the margin a stockout loses on that product.
How do I find the standard deviation of demand?
Take units sold per day for a recent period that matches today’s demand and use the STDEV.S function in a spreadsheet. Keep the units consistent: a daily standard deviation goes with a lead time in days.
Is safety stock the same as the reorder point?
No. The reorder point is the demand you expect during the lead time plus safety stock. At the start values that is 20 × 16 + 40 = 360 units.
Embed this calculator
Paste this script tag into your own page to show the calculator there, with the same maths and no sign-up. It fits the width of your page, resizes with the calculator and follows the visitor’s light or dark setting.
The script adds the calculator and a link to the full version under it. Keep the attribution link: it credits the calculator.
Get a reorder date for every product, not one safety stock for your shop.
Inventory Insights projects demand per product and works back from supplier lead time and a safety buffer to a reorder date, and warns you when that product’s reorder window opens. Stock cover also feeds the labels your campaigns use, so products you cannot deliver lose priority.
Analyse your products for freeLog in with Google, no credit card
See how Inventory Insights works