Inventory glossary
What is Weighted moving average?
Weighted moving average is a demand forecasting method that averages past sales over a set number of periods but gives more weight to recent periods than older ones.
For a small Shopify store, sales from last week usually tell you more about what happens next week than sales from two months ago. A plain average treats every period the same, so it reacts slowly to real changes in demand. A weighted moving average lets recent weeks count more, so your forecast catches a trend sooner without overreacting to one unusual day.
Use it on your better-selling SKUs where you have steady sales history. Pick the weights based on how fast your demand shifts: if sales swing a lot week to week, lean more weight on the most recent period. Recalculate it each time you add a new period of sales data.
Formula
Example
A candle store sold 40 units three weeks ago, 50 units two weeks ago, and 60 units last week. Using weights of 1, 2, and 3 (most recent week weighted highest): (40x1 + 50x2 + 60x3) / (1+2+3) = (40 + 100 + 180) / 6 = 320 / 6 = 53.3 units. A plain average of the same three weeks would be 50 units, so the weighted version reflects the recent upward trend more closely.
Common mistakes
- Picking weights arbitrarily without checking if they actually improve accuracy against past actual sales.
- Using too few periods, which makes the forecast jumpy and overly sensitive to one good or bad week.
- Applying the same weighting scheme to every SKU regardless of how stable or erratic its demand is.
Weighted moving average and SteadyShelf
SteadyShelf uses recent sales patterns like this to forecast demand per variant and turn that into order-by dates and suggested order quantities.
SteadyShelf is launching on the Shopify App Store. Join early access to get the install link first.
Related terms
Moving average forecast · Demand forecasting · Exponential smoothing · Average daily sales · Seasonal forecasting
FAQ
How is a weighted moving average different from a simple moving average?
A simple moving average gives every period in the window equal weight. A weighted moving average assigns higher weight to more recent periods, so it reacts faster to recent changes in demand.
How many periods should a small store use for a weighted moving average?
There is no fixed number. Three to six periods is common for weekly sales data. Fewer periods react faster to change but are noisier, more periods are smoother but slower to catch trends.
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