Follow

Customer Pareto in Advanced Planning

This is an important capability to help planners focus forecasting in Advanced Planning against the significant few customers versus the trivial many otherwise known as the Pareto Principle. This in turn ensures a better forecasting result, particularly for the items you are supplying, as compared to forecasting customers individually when their history is not forecastable.

There are three primary approaches to forecasting and demand planning that influences the definition "customer."

  1. When all customers individually do not have enough demand to forecast reliably. These are often web-based retail companies with 1000's of customers ordering once or only a few times a year. Such companies aggregate all their demand in a single "customer" entity in DemandCaster. We recommend that this type of demand aggregation be handled via the data integration.
  2. When a company has a few big distinct customers and/or sales channels. This too is often handled via the data integration process through some logical mapping of data or via the Customer Groups Creation process.
  3. When a company has many customers (500 or more) of which many are unforecastable due to not having sufficient sales history to forecast. In addition, these customers either can't be logically grouped manually or via the data integration process. The Pareto Process described herein is best suited for these types of companies.

Use Case

To understand the impact of the Pareto Principle in forecasting and demand planning, let's assume we have a company with $53 million in revenue and 1258 individual customers. Of these customers, 500 have sales over the last year which, from a forecasting perspective, is too many when considering that only 65 customers generate 80% of the revenue ($44 million). The balance of $9 million in revenue is the bottom 20% of revenue from 435 customers.

By leveraging the Pareto Principle, the system will focus forecasting on the significant customers that make up the lion share revenue. This contributes to driving efficiency in the planning process by not requiring the added time necessary to generate the statistical forecast for many unforecastable customers and in turn require employees to review the results.

The following describes the steps to run and manage the Pareto process for Advanced Planning.

Set Customer Pareto Percentage

This is applied in Advanced Planning Settings under System Settings. The setting is set to 100 by default for all customers.

To commence using the Pareto process, enter a value in the field to define the level of revenue where individual customers may be planned. By entering 80, upon uploading the customer order table, the system will automatically identify the customer that derive the top 80% of revenue. The remaining customers below the cumulative 80% revenue threshold will be grouped in a pre-defined category named "ParetoGroup." Increasing the Pareto number up or down will either add more individual customers to forecast or reduce the number.

In addition, any pre-grouped customers created via the Customer Groups Creation process will be included in the pareto calculation.

The pareto value may be changed. For the change to apply, a new customer orders upload must be executed.

Result after Customer Order upload

The area banded by the red line shows where the system transitions from discrete customers to grouped customers. The ParetoGroup customers will all be treated as a single "customer" demand entity named pareto group.

Please note that this calculation occurs with each upload when the Pareto percent value in system settings is less than 100. As a result of this, individual customers may move in an out of the pareto grouping.

Removing a Customer from Automatic Grouping

At times, you may choose to remove a customer from the automated pareto grouping. This is applicable with new significant customers that require closer planning attention or if there is a specific customer than requires more attention.

To do so, click the edit icon of the customer remove from the Pareto group.

Then, un-check the "Allow automatic grouping to ParetoGroup" check box and click Save.

DemandCaster - Customer Maintenance - Google Chrome

Now the customer is marked as "No" in Automatic Grouping column and will not be included in any subsequent pareto grouping calculations.

Please note that there is no option to mass update this setting against many customers at one time. This needs to be managed on a customer by customer basis.

DemandCaster - Customer Maintenance - Google Chrome

Forecast Result

Now the pareto group customers are forecasted collectively providing a much better forecast result as a group and more importantly as items within the group.

Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request

Comments