Forecasting and Forecast Pro tips by MXI
Forecast Pro users in the UK & Ireland who want to improve forecasting accuracy are always asking us which is the best way to use Forecast Pro where the historical time series data is inconsistent, is of poor quality, has too much noise in it or no discernable pattern etc. New users are naturally interested how Forecast Pro can be used to good effect in their organisation.
For forecasting one the oldest tricks in the book here is to consider aggregating the data. Companies that require both item-level and aggregate (e.g brand or product group) forecasts tend to concentrate exclusively on a bottom-up strategy: concentrating on forecasting each item and then add the item level forecasts to obtain a group forecast. Normally, however, the item-level data is difficult to forecast because it isle or lacks depth of history. MXI recommends for forecasting using a top-down strategy.
Generally with forecasting at the aggregated level there is more data to be consolidated, it performs more consistently generating better patterning for better forecasting. Forecast Pro eliminates the noise with greater ease and selects best forecasting methodology. The group forecast is generated directly (because group data is longer and more stable) and the item-level forecasts are then adjusted so that they sum to the group forecast.
With Forecast Pro at the group level simply select the Top/down facility and this pushes the forecasts results down to the sku level. The improvement in forecasting accuracy can be dramatic.
This is our view on forecasting and Forecast Pro, but we would like to hear your views. If you have the time can you please fill out our forecasting and budgeting survey