Demand forecasting
Coefficient of Variation (CV)
By the Supply Chain Manager team · Updated October 16, 2026
The coefficient of variation (CV) is the standard deviation of demand divided by its mean, a scale-free measure of how volatile a SKU's demand is.
Standard deviation alone cannot compare a SKU selling 1,000 a week with one selling 10. CV can, because it expresses variability relative to average demand. It is the basis of XYZ classification.
The time bucket matters a great deal. A SKU that sells a few units on random days looks extremely erratic on daily data and quite steady on monthly data. When comparing SKUs or setting XYZ thresholds, use the same bucket for all of them, ideally one close to how often you review and reorder.
How to calculate coefficient of variation
CV = Standard Deviation of Demand ÷ Mean Demand
Coefficient of variation example
Illustrative numbers: SKU A averages 200 a week with a standard deviation of 40 (CV 0.2). SKU B averages 10 with a standard deviation of 15 (CV 1.5). SKU A is far more predictable even though its standard deviation is larger.
Why coefficient of variation matters for distributors
CV tells you where to trust automation and where to hold extra buffer. High-CV SKUs need more safety stock for the same service level, or a different policy altogether.
Common coefficient of variation mistakes
- Comparing CVs calculated on different time buckets (daily versus weekly).
- Including one-off bulk orders that inflate variability.
Examples on this page use made-up round numbers to show the calculation. They are not benchmarks or results from real businesses.