Back to Pricing Dashboard User Documentation - Main Page
The aim of the Forecast Evolution page is to:
- Make historical comparisons: by comparing past forecasts to the current live forecast, users can understand how accurate previous forecasts were and identify any trends or patterns in forecasting errors.
- Evaluate performance: By comparing the frozen forecast with actual results, users can evaluate the performance of their forecasting models and make necessary adjustments to improve accuracy.
Data accessibility is linked to the visibility model, if your profile does not allow to see costs/ICM data, you will not see all the information of this page!
Forward Evolution Primary Rules
Today, DL keeps 2 forecasts per day starting from 8th day of the month.
To limit data volumetry in Pricing Dashboard, the first snapshot of the week is prioritzed, ie, you will have 4 snapshots in 1 month.
Forward Evolution Navigation
Select a Start and End Run Period ie the 2 forecast periods that you wish to compare.
- Version selection: Users shall select both forecast run periods to be compared.
- Unit selection: Users can choose which currency and unit to use.
- Time filters: Users can filter on time dimensions to reduce his/her scope of comparison.
- Dimension selection: Users can change the time dimension used for evolution charts.
- Quick filters: Users can use predefined filters on time dimensions.
- KPI selection: Users can either focus on sales amounts or quantities.
- Forecast Evolution over Time: This chart provides a global comparison for 2 run periods over time, comparing old forecasts to new actuals in the process.
- Forecast Period: This chart provides a high level comparison of multiple run periods, following how the overall amounts were estimated more or less optimistically week after week.
The next section shows the reasons behind the variation:
The Detailed Table shows the granular information based on the dimensions relevant for the GBU
