VANTAGE POINT: All Products Are Not Created Equal
How to Manage Forecasting for New Product Introductions
Consumer goods companies know that all products are not created equal when it comes to forecasting. Forecasting models fall into three unique categories: qualitative, quantitative and hybrid. Each category has inherent strengths and weaknesses. Qualitative models rely on subjective input from knowledgeable user. Quantitative models rely on sophisticated mathematical techniques to analyze historical demand patterns and project those into future demands. The third category, hybrid, combines the strengths of both qualitative and quantitative models to more accurately predict consumer buying habits for new product introductions (NPIs), seasonal products and product phase outs.
NPIs have proven to be a challenge for most consumer goods companies that only utilize qualitative or quantitative forecasting techniques. It is critical for consumer goods companies to leverage technology and utilize the hybrid or 'attribute' based forecasting models to ensure that you are achieving the highest possible levels of efficiency and forecast accuracy. By tuning your forecasting process to serve the unique dynamics of your business, you can cost-effectively meet your ever changing customer demands.
An accurate forecast and comprehensive demand plan combined with a clear understanding of the trade-off between desired service levels and required inventory investments ensures that you can meet or exceed customer service level objectives by cost-effectively synchronizing supply with demand. Given the complexity of today's global supply chains, technology plays a critical role by providing visibility into specific market demand along with the capability to leverage time-phased inventory policies with a variety of service level objectives to drive your supply chain's performance and profitability.
However, to be successful, your demand management solution needs to provide a selection of various forecasting models that will support the specific requirements of your unique product portfolio. Consumer goods companies who are frequently challenged with forecasting for NPIs should look for a technology solution that offers an attribute-based forecasting model. Attribute-based forecasting has proven to increase forecast accuracy of NPIs, short life cycle products and product phase outs which helps eliminate the uncertainty of meeting customer demand.
The attribute-based forecasting model utilizes historical sell-in or sell-through demand signals from multiple products based on user-defined product attributes. These attributes can include color, fabric, material type, region of the country, etc., which is then combined to create selling patterns that can be assigned to newly introduced products. Mid-season or launch correction capability helps to ensure that the forecast is reacting to initial sales or related demand signals.
Attribute-based modeling consists of four unique processes. Each process is important as they control the forecast of the product at different phases of the product's lifecycle. They are:
1. Creation of demand profiles
2. Assignment of demand profiles to new, seasonal and end-of-life products
3. Automatic revision of the forecast based on demand signals
4. Assess accuracy of demand profile based on demand signals
While there are other forecasting methods that can be used to generate forecasts for new products, they all have one major shortcoming. Other methods are limited to creating the forecast for a single product, or at most two or three products and the demand planner must possess a good understanding of existing products as well as the new products that may be replacing them.
With attribute-based modeling, the demand planner can select products based on multiple attributes. For example, you can create demand profiles for all the red scarves made out of cotton and their selling patterns in the northwest United States. This allows the planner to fine tune their selection, which will ultimately lead to a more accurate forecast.
Another important advantage of attribute-based modeling is the ability to efficiently realign history for events like Easter, which does not occur during the same period each year. In 2007, Easter occurred on April 8 and, in 2008, Easter occurred on March 23. When creating an 'Easter' profile it is critical that the Easter sales of previous year(s) are aligned with the future scheduling of Easter.
While the assignment of demand profiles to new, seasonal, and end-of-life products can be done manually, companies with a large volume of new products may find this to be very tedious. By using an inventory management system that enables 'user-defined attribute' matching capabilities, the planner can set the criteria where the attributes of the new product need to match the attributes selected to create the demand profile.
Forecast accuracy needs to be monitored continually when revising the forecast based on demand signals and profiles. One way of monitoring the accuracy of the forecast is through the use of point-of-sale (POS) data. In recent years, both the availability and accuracy of POS has increased dramatically. It is also the demand signal that most accurately reflects the buying patters of your customers and helps validate the accuracy of the forecast. And through a correctness-of-fit evaluation, your inventory management system should automatically adjust the forecast and/or demand profile assigned to the product further ensuring that you have the most accurate forecast to guarantee that the proper inventory is available to service your customers.
Brown Shoe Company, a leading consumer-driven footwear company and wholesale supplier of women's fashion shoes to U.S. department stores under brand names like Naturalizer, Via Spiga, Franco Sarto and LifeStride, has seen the benefits of using attribute-based modeling and POS data for demand planning. Brown Shoe sources about 90 million pairs of shoes globally and had no way to update their forecast based on consumer activity. By utilizing Logility Voyager Solutions for attribute-based forecasting, which allows Brown Shoe to group items by similar traits, Brown Shoe can create predictable seasonal profiles which are then used to forecast individual items and sizes. This gives Brown Shoe a more accurate weekly forecast to predict consumer sales and improve inventory flow to retailers.
The success of a consumer goods company depends on the ability to meet the demand of customers. Managing your inventory with a technology solution can enhance visibility and optimize inventory management at every level of the business. Adding in an attribute-based forecasting model further helps provide a more efficient supply chain for NPIs and seasonal products, ensuring that your customers can always find your product on the shelf and in the store at the point of demand.
___________________________________________________________________________
Karin L. Bursa, vice president of marketing at Logility, oversees the company's market positioning and strategy development as well as the creation and execution of marketing programs for Logility Voyager Solutions, the company's full suite of collaborative commerce solutions to optimize the supply chain. For more information, contact [email protected].
By Karin Bursa, Vice President, Marketing, Logility Inc.
Consumer goods companies know that all products are not created equal when it comes to forecasting. Forecasting models fall into three unique categories: qualitative, quantitative and hybrid. Each category has inherent strengths and weaknesses. Qualitative models rely on subjective input from knowledgeable user. Quantitative models rely on sophisticated mathematical techniques to analyze historical demand patterns and project those into future demands. The third category, hybrid, combines the strengths of both qualitative and quantitative models to more accurately predict consumer buying habits for new product introductions (NPIs), seasonal products and product phase outs.
NPIs have proven to be a challenge for most consumer goods companies that only utilize qualitative or quantitative forecasting techniques. It is critical for consumer goods companies to leverage technology and utilize the hybrid or 'attribute' based forecasting models to ensure that you are achieving the highest possible levels of efficiency and forecast accuracy. By tuning your forecasting process to serve the unique dynamics of your business, you can cost-effectively meet your ever changing customer demands.
An accurate forecast and comprehensive demand plan combined with a clear understanding of the trade-off between desired service levels and required inventory investments ensures that you can meet or exceed customer service level objectives by cost-effectively synchronizing supply with demand. Given the complexity of today's global supply chains, technology plays a critical role by providing visibility into specific market demand along with the capability to leverage time-phased inventory policies with a variety of service level objectives to drive your supply chain's performance and profitability.
However, to be successful, your demand management solution needs to provide a selection of various forecasting models that will support the specific requirements of your unique product portfolio. Consumer goods companies who are frequently challenged with forecasting for NPIs should look for a technology solution that offers an attribute-based forecasting model. Attribute-based forecasting has proven to increase forecast accuracy of NPIs, short life cycle products and product phase outs which helps eliminate the uncertainty of meeting customer demand.
The attribute-based forecasting model utilizes historical sell-in or sell-through demand signals from multiple products based on user-defined product attributes. These attributes can include color, fabric, material type, region of the country, etc., which is then combined to create selling patterns that can be assigned to newly introduced products. Mid-season or launch correction capability helps to ensure that the forecast is reacting to initial sales or related demand signals.
Attribute-based modeling consists of four unique processes. Each process is important as they control the forecast of the product at different phases of the product's lifecycle. They are:
1. Creation of demand profiles
2. Assignment of demand profiles to new, seasonal and end-of-life products
3. Automatic revision of the forecast based on demand signals
4. Assess accuracy of demand profile based on demand signals
While there are other forecasting methods that can be used to generate forecasts for new products, they all have one major shortcoming. Other methods are limited to creating the forecast for a single product, or at most two or three products and the demand planner must possess a good understanding of existing products as well as the new products that may be replacing them.
With attribute-based modeling, the demand planner can select products based on multiple attributes. For example, you can create demand profiles for all the red scarves made out of cotton and their selling patterns in the northwest United States. This allows the planner to fine tune their selection, which will ultimately lead to a more accurate forecast.
Another important advantage of attribute-based modeling is the ability to efficiently realign history for events like Easter, which does not occur during the same period each year. In 2007, Easter occurred on April 8 and, in 2008, Easter occurred on March 23. When creating an 'Easter' profile it is critical that the Easter sales of previous year(s) are aligned with the future scheduling of Easter.
While the assignment of demand profiles to new, seasonal, and end-of-life products can be done manually, companies with a large volume of new products may find this to be very tedious. By using an inventory management system that enables 'user-defined attribute' matching capabilities, the planner can set the criteria where the attributes of the new product need to match the attributes selected to create the demand profile.
Forecast accuracy needs to be monitored continually when revising the forecast based on demand signals and profiles. One way of monitoring the accuracy of the forecast is through the use of point-of-sale (POS) data. In recent years, both the availability and accuracy of POS has increased dramatically. It is also the demand signal that most accurately reflects the buying patters of your customers and helps validate the accuracy of the forecast. And through a correctness-of-fit evaluation, your inventory management system should automatically adjust the forecast and/or demand profile assigned to the product further ensuring that you have the most accurate forecast to guarantee that the proper inventory is available to service your customers.
Brown Shoe Company, a leading consumer-driven footwear company and wholesale supplier of women's fashion shoes to U.S. department stores under brand names like Naturalizer, Via Spiga, Franco Sarto and LifeStride, has seen the benefits of using attribute-based modeling and POS data for demand planning. Brown Shoe sources about 90 million pairs of shoes globally and had no way to update their forecast based on consumer activity. By utilizing Logility Voyager Solutions for attribute-based forecasting, which allows Brown Shoe to group items by similar traits, Brown Shoe can create predictable seasonal profiles which are then used to forecast individual items and sizes. This gives Brown Shoe a more accurate weekly forecast to predict consumer sales and improve inventory flow to retailers.
The success of a consumer goods company depends on the ability to meet the demand of customers. Managing your inventory with a technology solution can enhance visibility and optimize inventory management at every level of the business. Adding in an attribute-based forecasting model further helps provide a more efficient supply chain for NPIs and seasonal products, ensuring that your customers can always find your product on the shelf and in the store at the point of demand.
___________________________________________________________________________
Karin L. Bursa, vice president of marketing at Logility, oversees the company's market positioning and strategy development as well as the creation and execution of marketing programs for Logility Voyager Solutions, the company's full suite of collaborative commerce solutions to optimize the supply chain. For more information, contact [email protected].