Quick summary: The first pillar of demand planning is explained in this article, which covers refining reference measures, and the different ways this can be approached in Netstock.
The purpose of the cleansing step is to clean historical data for optimal forecasting results, ultimately saving all changes into the final history. Currently, cleansing processes in Netstock happen in the background, such as by removing outliers automatically, and is often outside of the control of the user. In Netstock Demand Planning, control is given back to the users to adjust these items, as required.
💡The Final History (which is presented in the forecast row with the year) is known as the Resultant History in IBP.
To begin the process of cleansing the sales history, select the year for cleansing from the table:
This expands the entry into different measures used to calculate the final history for each year:
Standard Measures
The standard measures are as follows:
Imported history: These are the transactions that were imported from the ERP. These cannot be edited.
Outliers: The current outlier protection (currently included in Netstock from results generated by the forecast engine) are displayed here for transparency.
Lost Sales Calculation: This calculation automatically increases sales history to compensate for sales that may have been lost during periods where there was no available stock. For more information on lost sales compensation, see here.
Promos: Because promotions are not considered real demand, any items that form part of the sales history can be added in here, which will deduct from the final history.
Overrides: In the event that a total correction/override is required for the final history for a particular month, the override can be captured, and the total final history will be updated.
Outliers and Lost Sales are new fields that were not available in Pivot Forecasting.
💡The Imported History (which is presented in the Imported measure) is known as the Actual History in IBP. Promos or Overrides are known as the Adjusted History measure in IBP.
Reference Measures
Reference measures are used as additional reference points to influence forecasts by tracking what happened in the past. A new import type has been included to accommodate these reference measures. Netstock Demand Planning supports up to 20 reference measures.
Any additional reference measures will be added in as rows underneath the standard measures entries, and are indicated by a book icon:
Visualizations of Metrics and Reference Measures
Any of the standard and reference measures data provided in the table can be viewed on the graph by selecting the relevant measures from the panel at the bottom of the graph. Measures that are grayed out (not selected) are not displayed:
Single Entries
Any fields in the table that contain the gray arrow in the cell can be edited. When a cell is selected, the cell will turn purple:
It is important to double-click on the cell so that a cursor is displayed:
Any cells that have had an edit made to them have a purple arrow, which allows users to quickly see cells having unsaved changes:
Any editable cell in the table can be amended in this manner.
Multiple Entries
Netstock also supports the ability to update multiple cells simultaneously. To select multiple cells, select the starting cell and drag the cursor along to the final cell. Netstock will highlight the range for update:
The same can be accomplished by selecting the first cell in your range, holding the control key, and then selecting the last cell in your range.
Alternately, you can select the row name for any editable row and the entire row will be selected:
This opens an adjustment panel at the bottom of the screen, and displays the selected time range in the graph. In the below example, the panel is entitled “Adjusting Overrides” to match which measure is being adjusted in the table:
Adjustment Panel
There are various options to select in the adjustment panel that ease the cleansing process for the user.
Selecting "Add to" with a value in the adjustment field adds this amount to each of the highlighted cells within the table. In the below example, a value of 10 has been entered and the Adjust button selected:
A value of 10 has then been added into each highlighted cell in the table:
Selecting "Subtract from" with a value in the adjustment field subtracts this amount from the highlighted cells within the table. In the below example, a value of 3 has been entered and the Adjust button selected:
A value of 3 has then been subtracted from each highlighted cell in the table:
Set to updates each value in the highlighted cells to the value specified. For example, setting the value to 70 updates the value in each cell to 70:
Spread takes the total specified in the adjustment field and spreads this evenly across each cell. In the below example, a value of 50 is spread across 5 months, resulting in a value of 10 for each month:
The percentage slider increases and decreases the values in the selected cells by the percentage specified. This percentage can be typed into the field, or the slider can be used to adjust. These changes will be displayed real-time in the table and graph.
Any value inserted into the block can be removed from the field by selecting "Clear":
Finally, it may be useful to copy, add or subtract from other measures provided. To do this, select the source values from the drop-down list, as well as the corresponding operation. In the below example, the values from Imported history will be copied to the highlighted cells:
Should any changes made need to be undone, select the Undo button on the top right-hand side of the table. All changes made will revert back to their previous state.
At this point, any changes made to the cleansing step can be saved using the Save button on the top right-hand side of the table. Once data is saved, any cells where changes are processed will change from a purple arrow to a gray arrow:
