Inventory glossary

What is Mean absolute deviation?

Mean absolute deviation is the average size of the errors between your sales forecast and what actually sold, ignoring whether each error was too high or too low.

Mean absolute deviation tells you how far off your forecasts usually are, in plain units. If your candle shop's MAD is 5 units per week, you know to expect forecasts to be wrong by about 5 units, which helps you set buffer stock instead of guessing.

Use it to compare forecasting methods or to size safety stock. A low MAD means you can run leaner. A high MAD means you need more cushion or shorter order cycles to avoid stockouts and overstock.

Formula

MAD = sum of |actual - forecast| / number of periods

Example

A candle store forecasts 50 units a week for four weeks. Actual sales were 55, 45, 60, and 40. The errors are 5, 5, 10, and 10. The average of those is (5+5+10+10)/4 = 7.5. The MAD is 7.5 units.

Common mistakes

  • Confusing MAD with a percentage error like MAPE, which is scaled differently.
  • Treating a single period's error as MAD instead of averaging over several periods.
  • Ignoring MAD when setting safety stock, which leads to buffer levels that do not match real forecast accuracy.

Mean absolute deviation and SteadyShelf

SteadyShelf's demand forecasts improve as it sees more sales history, which keeps the gap between forecast and actual sales, and therefore MAD, as small as possible.

SteadyShelf is launching on the Shopify App Store. Join early access to get the install link first.

Related terms

Forecast accuracy · Forecast error · Standard deviation of demand · Demand variability

Full inventory glossary

FAQ

Is a lower mean absolute deviation always better?

Generally yes, since it means your forecasts are closer to actual sales. But compare MAD against your average sales volume, since 5 units of error matters more for a slow SKU than a fast one.

How is mean absolute deviation different from standard deviation of demand?

MAD averages the absolute size of forecast errors. Standard deviation of demand measures how much actual sales vary around their own average, regardless of any forecast. They answer different questions but both describe variability.

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