CPG Company Takes Customer Fill Rates to the Next Level by Connecting Forecasting and Retailer Data

• 分析与建模 - 预测分析
• 功能应用 - 库存管理系统
• 平台即服务 (PaaS) - 数据管理平台

• 消费品

• 销售与市场营销
• 仓库和库存管理

• 需求计划与预测
• 补货预测
• 供应链可见性(SCV)

• 软件设计与工程服务
• 系统集成

For this organization, providing great customer service means meeting retailers’ orders on-time, in-full. The company uses customer fill rate as the metric to gauge their service performance. High customer fill rates are driven in large part by accurate demand forecasts which allow the company to ensure sufficient inventory availability. A general challenge for the CPG industry is that the innovation and marketing strategies important to drive growth also have an adverse effect on forecast accuracy, making the business harder to predict. Furthermore, the company made a structural shift from direct-store delivery to warehouse distribution. While this provided value gains in supply chain efficiency, the shift makes it even harder to predict customer demand. To mitigate the impact of these strategies on customer fill rate — and even achieve a net gain in service — the business realized the time had come to take demand forecasting performance to the next level.

This global consumer packaged goods (CPG) company was set up in the early 1900s to commercialize an innovative food product. By the beginning of the 21st century, it had experienced decades of expansions and acquisitions and built its portfolio to offer multiple product categories, primarily in packaged food. Headquartered in the United States, the company now operates in over 180 countries across Europe, Asia, Australia, Latin America and North America. It employs 34,000 people and generates over $13 billion of revenue annually. The business manufactures all goods in-house and distributes to consumers through retail partners. In recent years, shifting consumption patterns and changing market preferences have altered demand in numerous product categories. In response, the company’s strategy has been to balance growth through initiatives such as continued product innovation, marketing and advertising, new package formats, increased investment in emerging markets, divesting non-core brands, and harmonizing distribution methods in the United States. Though some of these initiatives focused on efficiency, the company’s underlying priority is to deliver great customer service to retailers, and this requires agility to anticipate and quickly respond to customer demand.

The company selected e2open for this next stage in their forecasting excellence journey because of its trusted relationship, proven performance and quick time-to-value. An underlying driving factor was the integrated nature of e2open’s planning capabilities and business network on a single platform. In particular, the ability to bring together the data and applications required to get a new step-change in accuracy without the technical challenges that normally come from using retailer information for supply chain planning was important. In fact, different teams within the company were already using these capabilities from e2open, but in a siloed manner. Normally, integrating these systems would be costly and time-consuming. However, in this case, since both were already part of e2open’s integrated planning platform, there was no technical barrier in linking the two systems, making it a natural evolution. The first step toward boosting forecasting performance for any company is to use real-time signals to sense demand. As a leader in space, the organization had been sensing demand with e2open for years. The next step in forecasting was for the supply chain team to take advantage of the predictive information contained in external signals such as point-of-sale (POS) scans, store-level inventory and retailer warehouse withdrawals. Likewise, account teams at the company have used e2open retail operational management and store-level execution solutions for years, so they already had access to decision-grade data from many of their channel partners. Now the time had come to connect these systems to use retailer data to drive better forecasts.

• The company reduced its weekly forecast error by more than 40% using e2open Demand Sensing.
• By connecting Demand Sensing to Demand Signal Management and systematically using retailer data as an input for forecasting, the company achieved an additional 12% reduction in forecast error.
• Joining the two systems also reduced inventory levels, unplanned production changes, and the number of expedited shipments required to meet service commitments.

• Increased business coverage from 55% to 70% of its sales.
• Reduced weekly forecast error by more than 40%.
• Achieved an additional 12% reduction in forecast error.