Supply Chain Manager
Supply chain glossary

Demand forecasting

MAPE (Mean Absolute Percentage Error)

By the Supply Chain Manager team · Updated October 16, 2026

MAPE (mean absolute percentage error) averages each period's forecast error as a percentage of actual demand, ignoring whether it was over or under.

It is popular because percentages are easy to explain. Its weakness is low or zero demand: a forecast of 2 against an actual of 1 is a 100% error, and an actual of 0 makes the calculation impossible. Slow and intermittent SKUs can therefore produce huge, misleading MAPEs.

A common variant, symmetric MAPE, divides by the average of actual and forecast to reduce the problem with small actuals, but it has quirks of its own. For a range with many slow movers, measures that divide total error by total demand, such as WAPE, are usually easier to interpret.

How to calculate MAPE

MAPE = (1 ÷ n) × Σ ( |Actual − Forecast| ÷ Actual ) × 100

MAPE example

Illustrative numbers: actuals of 100, 50 and 10 against forecasts of 90, 60 and 20. Percentage errors are 10%, 20% and 100%. MAPE = 43%, driven almost entirely by the 10-unit week.

Why MAPE matters for distributors

For a distributor with many slow movers, MAPE can make a decent forecast look terrible. WAPE or MASE usually give a fairer picture across a whole range.

Common MAPE mistakes

  • Using MAPE on SKUs with zero-demand periods.
  • Averaging MAPE across SKUs, so tiny items dominate the result.

Examples on this page use made-up round numbers to show the calculation. They are not benchmarks or results from real businesses.