Inventory glossary
What is Standard deviation of demand?
Standard deviation of demand measures how much actual sales in a period swing away from the average, telling you how unpredictable a product's demand is.
A low standard deviation means a product sells about the same amount every week, so you can stock close to the average and sleep fine. A high standard deviation means some weeks are quiet and others spike, so the same average sales figure hides real risk of running out or sitting on excess stock.
Use it to set safety stock. Products with more erratic demand need a bigger buffer than products that sell steadily, even if both have the same average weekly sales. This stops you from treating a predictable candle and an unpredictable seasonal candle the same way.
Formula
Example
A candle shop sells this many units over 5 weeks: 10, 12, 8, 30, 10. Average is 70/5 = 14. Differences from average: -4, -2, -6, 16, -4. Squared: 16, 4, 36, 256, 16, summing to 328. Divide by (5-1) = 82. Square root of 82 is about 9.06 units. That one spike week makes the standard deviation much higher than it would be for steady sales of 10, 12, 11, 13, 10.
Common mistakes
- Using too few weeks of data, which makes one unusual week distort the whole number.
- Ignoring standard deviation and only looking at average sales, which hides risky, spiky SKUs.
- Mixing up standard deviation of demand with standard deviation of lead time, they are different inputs to safety stock.
Standard deviation of demand and SteadyShelf
SteadyShelf factors demand variability like this into its forecasts so order-by dates and suggested quantities account for spiky SKUs, not just their averages.
SteadyShelf is launching on the Shopify App Store. Join early access to get the install link first.
Related terms
Demand variability · Safety stock · Z-score in inventory · Demand forecasting · Service level
FAQ
Why does standard deviation of demand matter more than just the average?
Two SKUs can have the same average weekly sales but very different risk levels. The one with higher standard deviation needs more safety stock because its demand is less predictable, even though the average looks identical.
How many weeks of sales data do I need to calculate this reliably?
More is better. A handful of weeks can be skewed badly by one unusual spike or dip. Aim for at least 8 to 12 weeks of regular sales, more if the product is seasonal, before trusting the number for safety stock decisions.
Let SteadyShelf do this math for every SKU
Order-by dates per supplier, seasonal forecasts and draft POs for your Shopify store. $29/month after a 30-day free trial.
Get early access