Executive Briefings

Attaining The Next Maturity Level In Demand Management

Traditional demand management approaches fail miserably in providing companies with an accurate depiction of predicted demand.

Traditional supply chain planning solutions' approach to demand management is invalid. For example, assuming that most items follow a normal distribution is inaccurate for many of the companies that deal with intermittent, sporadic products. Relying heavily on historic shipments and orders to forecast future demand and ignoring the wealth of insight that can be gleaned from recent channel data contribute to low forecast accuracy. In the face of a critical need to better understand future demand at manufacturers and retailers, here are the top five recommendations for improving demand management.

• Focus on demand sensing, not just demand forecasting-Demand sensing is improving visibility to channel consumption data. The shorter the lag time in seeing consumption, the more options companies have to shape demand and create a profitable demand response. Demand sensing can be achieved through many data streams. For example, VMI is an effective way to sense channel demand, by synchronizing demand pulls based on VMI signals for major customer accounts.

• Increase the granularity and frequency of modeling demand-With higher frequencies of demand modeling on actual channel demand, the random nature of many demand streams can be captured and made visible. As a result, many companies are increasing the frequency of demand planning to a weekly or daily process from a monthly one and the granularity from product family to single SKU-ship to location.

• Leverage attribute-based forecasting to drive demand shaping programs-Attribute modeling, the inclusion of channel attributes as the foundation for demand forecasting and sensing, allows companies to move from unit forecasting to attribute-based sensing. By forecasting the attributes, the company can sense demand and connect this to demographics, helping with demand shaping efforts for new product introductions and products with short lifecycles.

• Improve internal and external collaborative planning - Internal and external stakeholders in the demand forecasting process have a wealth of insights that can attribute to a more accurate future demand picture. Leveraging internal collaborative processes, like sales and operations planning and external collaborative relationships with suppliers and channel partners, and incorporating this information in the demand forecasting process can limit the need to rely on latent historical information.

• Build a profitable demand response in your supply network-Remember that accurate demand forecasting is only the means to an ultimate goal: profitable customer fulfillment. The objective is to ensure that your supply network-including inventory, manufacturing, logistics and sourcing functions-can profitably respond to the predicted demand. Building a responsive supply network requires strategic initiatives like network design and supplier selection, tactical efforts like defining a robust S&OP process as well as operational initiatives in areas like order fulfillment and available to promise.

• And finally... instill a demand forecasting excellence culture at your organization-This includes measuring individual forecast bias (whether a user is typically over or under forecast) and rewarding accuracy. It also includes auditing the reasons behind and impact of users' overrides.  And finally it includes building an attractive career path for forecasting experts, to ensure retention and continued engagement.

The Outlook

With continued economic pressures on manufacturers and retailers, the quest for the next-generation demand management processes and technologies will accelerate in 2009. Expect supply chain management vendors to continue to enhance their solutions' functionality in areas like demand sensing and attribute-based planning. Leading companies will continue to offer examples of better demand management processes that rely on short latency in leveraging channel data, heavy internal and external collaboration and tight integration with upstream supply network processes.

Traditional supply chain planning solutions' approach to demand management is invalid. For example, assuming that most items follow a normal distribution is inaccurate for many of the companies that deal with intermittent, sporadic products. Relying heavily on historic shipments and orders to forecast future demand and ignoring the wealth of insight that can be gleaned from recent channel data contribute to low forecast accuracy. In the face of a critical need to better understand future demand at manufacturers and retailers, here are the top five recommendations for improving demand management.

• Focus on demand sensing, not just demand forecasting-Demand sensing is improving visibility to channel consumption data. The shorter the lag time in seeing consumption, the more options companies have to shape demand and create a profitable demand response. Demand sensing can be achieved through many data streams. For example, VMI is an effective way to sense channel demand, by synchronizing demand pulls based on VMI signals for major customer accounts.

• Increase the granularity and frequency of modeling demand-With higher frequencies of demand modeling on actual channel demand, the random nature of many demand streams can be captured and made visible. As a result, many companies are increasing the frequency of demand planning to a weekly or daily process from a monthly one and the granularity from product family to single SKU-ship to location.

• Leverage attribute-based forecasting to drive demand shaping programs-Attribute modeling, the inclusion of channel attributes as the foundation for demand forecasting and sensing, allows companies to move from unit forecasting to attribute-based sensing. By forecasting the attributes, the company can sense demand and connect this to demographics, helping with demand shaping efforts for new product introductions and products with short lifecycles.

• Improve internal and external collaborative planning - Internal and external stakeholders in the demand forecasting process have a wealth of insights that can attribute to a more accurate future demand picture. Leveraging internal collaborative processes, like sales and operations planning and external collaborative relationships with suppliers and channel partners, and incorporating this information in the demand forecasting process can limit the need to rely on latent historical information.

• Build a profitable demand response in your supply network-Remember that accurate demand forecasting is only the means to an ultimate goal: profitable customer fulfillment. The objective is to ensure that your supply network-including inventory, manufacturing, logistics and sourcing functions-can profitably respond to the predicted demand. Building a responsive supply network requires strategic initiatives like network design and supplier selection, tactical efforts like defining a robust S&OP process as well as operational initiatives in areas like order fulfillment and available to promise.

• And finally... instill a demand forecasting excellence culture at your organization-This includes measuring individual forecast bias (whether a user is typically over or under forecast) and rewarding accuracy. It also includes auditing the reasons behind and impact of users' overrides.  And finally it includes building an attractive career path for forecasting experts, to ensure retention and continued engagement.

The Outlook

With continued economic pressures on manufacturers and retailers, the quest for the next-generation demand management processes and technologies will accelerate in 2009. Expect supply chain management vendors to continue to enhance their solutions' functionality in areas like demand sensing and attribute-based planning. Leading companies will continue to offer examples of better demand management processes that rely on short latency in leveraging channel data, heavy internal and external collaboration and tight integration with upstream supply network processes.