Terra Technology launches 2011 Forecasting Benchmark Study

The report identifies forecasting trends based on activity from one third of the North American Consumer Packaged Goods (CPG) market and includes insights about performance during promotions and new product introductions.

The study, which spans the 2009-2010 period, encompasses more than 7 billion physical cases that represent more than $200 billion in sales from nine leading multinational consumer products companies.

Results show that weekly forecast error for Demand Planning systems remains high in CPG, at almost 50%, with bias remaining positive at 5%.

Results also show that Demand Sensing provides a major improvement in forecast accuracy across all companies, reducing weekly error by 40%.

Improving forecast accuracy is key to an efficient and cost-effective supply chain since demand uncertainty is the number one driver of safety stock.

"We believe this study is the most comprehensive examination of forecasting performance for the CPG industry," said Robert F. Byrne, President and CEO, Terra Technology.

"It is unique because it provides a truly comparable benchmark for manufacturers to evaluate demand planning performance against their peers.

It also shows the potential of Demand Sensing to improve forecast accuracy, supported by real-life performance figures from leading manufacturers whose supply chains are among the most respected in the industry."

Findings and insights from the analysis include:

• Demand Planning weekly forecast error remains almost 50% in the CPG industry, despite the effort and expense invested in implementing Demand Planning systems. The 5% bias may be indicative of the optimistic nature of Marketing.
  
• Promotional volume jumped more than 75% in 2010 as companies sought to drive sales by offering consumers additional value. Contrary to conventional wisdom about promotions, individual items are forecast as accurately when promoted as they are when not promoted. However, bias is more than five times higher during promotional periods.
  
• High expectations and the absence of prior sales history make new product introductions particularly hard to forecast. In 2010, new items accounted for 13% of the volume and experienced weekly Demand Planning error rates of 65% over their first year compared to 46% for existing products. Demand Sensing provides a consistent step change in forecast accuracy across all aspects of the business, including promotions, new products, fast-moving and slow-moving items.

The 40% reduction in overall error provided by Demand Sensing confirms the benefits of an automated mathematical approach which harnesses real-time demand signals throughout the supply chain to improve forecast accuracy.

The Terra Technology Forecasting Benchmark Study differs from other benchmarks because all raw data is taken directly from the participants' Demand Planning systems, then consistent metrics are used.

The 2011 dataset includes 300 000 item/warehouse combinations with 55 000 base codes and 72 000 items across 350 locations.

To download a copy of the Terra Technology Forecasting Benchmark Study, see: http://www.terratechnology.com/benchmark-study-welcome

Source: AFP