The Innovator's Solution


Coming Now: The Age of Advanced Demand Analytics

Posted by Jeff Bodenstab on Aug 11, 2015 3:04:00 PM
  1. Supply chains are getting more complex, while

  2. Customers are insisting on ever faster response times.

In her recent report, Making the Case for Increased Adoption of Advanced Supply Chain Analytics, Gartner analyst Noha Tohamy says that it’s no longer realistic to expect planners alone to analyze trends and underlying drivers, predict future scenarios and devise action plans.

“Further increases in data volumes, supply chain complexity and customer expectations for faster response times will only increase the need for smarter analytics that require no or low levels of manual intervention,” she says. As a result this “expected decrease in human intervention requires supply chain organizations to invest in automated advanced analytics.”

One of the best opportunities Tohamy sees for advanced analytics is demand planning. She sees organizations increasingly relying on statistical analysis and machine learning capabilities to predict future demand. She recommends investing in “automated supply chain processes that will decreasingly require manual intervention.” She says planners should instead focus on proficiency in demand analytics and the ability to clearly communicate insights to business users.

The payback is clear, she says. “Results show a strong correlation between analytics adoption and supply chain benefits, with 96% of organizations having achieved ROI in analytics.”


Danone is a case in point. Number 58 on the Forbes list of most powerful brands, Danone offers a wide range of healthy food products like yogurt and Evian water whose demand is strongly impacted by trade promotions, media events and advertising. More than 30% of these products are sold on promotion —accounting for nearly 70% of forecast error.

Danone wanted more reliable demand forecasting to consistently predict the actual impact, or “lift,” to baseline demand from trade promotions and media events—and to help its departments reconcile their sometimes conflicting goals and targets. This required analyzing a huge number of variables with complex interactions—all buried inside “big data” that exhibited a high degree of noise.


Danone added analytics with machine learning capabilities to analyze their demand planning. Their solution provided reliable, detailed modeling of trade promotion uplift for sales and marketing. It enabled the supply chain team to satisfy promotions and media uplifts with timely production and balanced inventory deployment —achieving the target service levels for channel and store-level supply chain execution.

The resulting gains included:

  • Rapid increase in forecast accuracy to 92%

  • Increase in service level to 98.6%, with a 30% reduction in lost sales

  • 30% reduction in product obsolescence

  • 6% increase in net ROI the first year—improved to 8% the next

  • 36% improvement in net uplift from promotions the first year; improved 55% the next

  • Demand planner workload cut in half, with re-focus on higher value-added activity

  • Exceeded service level target for 37 consecutive months—and counting

The implementation also established a robust execution foundation for Danone and a platform for faster response to changes in business objectives. It drives benefits across marketing, account planning, finance and production; allowing them to better coordinate and communicate—leading to more consistent results and fewer emergencies. Or as Tohamy recommends, “Invest in analytics solutions that support not only one functional area, but can improve performance of cross-functional processes.”

Click below for a copy of the Danone case study:

 ToolsGroup Danone Case Study


Topics: Forecasting Demand and Analytics, Machine Learning

Share this article or your comments below: