Safety stock and reorder point, with real numbers

The example

A SKU sells 10 a day on average. Day to day it moves around: the standard deviation of daily sales is 4. The supplier takes 5 days. You want to be in stock 95% of the time.

1. Safety stock when only demand varies

Safety stock = Z × standard deviation of daily sales × √lead time. Z comes from the service level: 90% is 1.28, 95% is 1.65, 98% is 2.05, 99% is 2.33.

2. Reorder point

Reorder point = average daily sales × lead time + safety stock. When stock on hand (plus anything already ordered) drops to this number, place the order.

3. When the supplier is unreliable too

If lead time itself moves around, add that variation: Z × √(lead time × demand std² + average demand² × lead time std²). If the 5 days are sometimes 4 and sometimes 6 (std about 1 day):

The shortcut that overshoots

A common rule is (max daily sales × max lead time) − (average daily sales × average lead time). With a best day of 18 and a worst lead time of 7 days that gives 18 × 7 − 50 = 76, five times the formula. It's easy, but it plans for your best sales day and your worst delivery at the same time, which rarely happens. It's fine for a few critical SKUs, expensive across the whole catalogue.

Which service level?

Going from 95% to 99% raises Z from 1.65 to 2.33, about 40% more safety stock for 4 points of availability. Keep 98–99% for bestsellers and items that lose you the customer when missing, and 90–95% for the long tail.

Try it with your numbers

The calculator below does both methods side by side, takes your service level, and gives you a link you can share with the team. Nothing is uploaded.

Next: see which SKUs run out first from one spreadsheet