Demand forecasting is predicting what will sell, by style, size, and account, in order to size a production run or a seasonal buy.
Oct 5, 2026
A brand forecasting production and a retailer forecasting a buy are solving different problems with the same word. The brand needs a total to cut against a minimum; the retailer needs a figure that fits an open-to-buy and a floor plan. A forecast built for one rarely serves the other.
Either way the useful output is not a single number. It is a range with the risk named: how wrong this can be before the production run stops being economic, or before the buy eats the budget meant for in-season reordering.
The strongest input is sell-through by style and size on comparable product, not total revenue history. A style that cleared at full price tells you something a style that cleared after two markdowns does not, and last season’s size curve usually predicts better than any model of the next one. Sell-through per style and per account is one of the cleanest inputs; see the sell-through rate calculator.
Models help most where the data is dense and the pattern is dull: reordering carryover basics, sizing a repeat colourway, flagging an account whose orders are drifting. They help least with the thing brands most want forecast, which is whether a genuinely new product will work.
Pre-orders, stock orders and reorders in one platform.
Free to start, no credit card required.
We use cookies to measure and improve your experience. For more information, please read our cookie policy.