This page enables you to carry out forecast accuracy analysis for a specified date range. You can analyze by plan, plan input, and period against actual demand in both unit and revenue terms. This provides valuable insight into the accuracy of each of the Consensus Plan's inputs. It also gives direction based on the selection by period when creating the plan.

Use the context selectors at the top-right to select a location and product combination. Additionally, you can manually enter the required values using the search function.

The Forecast Waterfall table displays forecasted weekly demand.

The Forecast Accuracy Waterfall table displays the weekly forecast accuracy. The context selector at the bottom of the table allows you to specify the forecast accuracy measure.

ParameterDescription
Forecast Accuracy Weighted Average (%)Displays a weighted average percentage value of the forecast accuracy measure.
Forecast Accuracy Simple Average (%)Displays an average percentage value of the forecast accuracy measure.

The Actuals vs Archived Plans graph displays the weekly Base History, and Consensus Plan.

ParameterValue
Archived ForecastSpecify the basis for the data displayed on the page. The options are Demand Plan, Marketing Forecast, Consensus Plan, or Stat Forecast.
Waterfall Start Specify the start of the waterfall analysis period.
Waterfall EndSpecify the end of the waterfall analysis period.
Forecast Error Basis

Specify how the forecast error is calculated. The absolute forecast error is divided by:

  • Actuals: The shipment history is used as the basis for forecast accuracy.
  • Forecast: Divides the forecast error by the forecast. 
  • Minimum Error: Divides the forecast error by either the history or forecast, whichever gives the lowest error. This helps prevent incentivizing behavior that introduces bias into forecasts. 
Forecast Accuracy Measure

Specify the forecast accuracy measure:

  • MAPE: Uses the Mean Absolute Percentage Error (MAPE) value. 
  • Accuracy: Uses the Forecast Accuracy value, defined as 100% minus the Mean Absolute Percentage Error (MAPE) value.