Colorado Craft-Beer Maker Brews Up High Levels of Customer Service
SupplyChainBrain | 16.08.2010
Demand planning solution helps New Belgium Brewing to maintain high customer satisfaction while keeping inventory at an acceptable level even in the face of rapid growth.
For most of its existence, the New Belgium Brewing Company has lived the classic good news-bad new scenario.
The good news is that the Fort Collins, Colo.-based supplier of craft beers such as Fat Tire Ale has consistently enjoyed strong, steady growth. It sold 3,500 barrels of beer in 1993, and is on track to move nearly 700,000 barrels in 2010.
The bad news is that each time New Belgium reaches another growth milestone it also encounters a new set of supply chain management problems.
This situation reached crisis stage in 2003 as New Belgium began expanding into the potentially lucrative—but also highly competitive—California market. “That was a pretty turbulent time,” recalls Nate Turner, New Belgium’s director of supply chain management. “Our initial sales forecasts were on the high side. That made it apparent that we needed more sophisticated supply chain management tools.”
Until then, New Belgium managed its supply chain with a single forecasting application with limited capacity for manipulating data. Any attempt at data analysis required exporting data to an Excel spreadsheet, which also had its limitations, particularly when it came to determining whether actual sales were in line with system-generated forecasts.
“This system was decent when we were a small company,” Turner says. “But as we continued to grow, we needed a program that could talk to our other systems and allow us to get out of Excel.”
With management’s permission, Turner began searching for a new supply chain management solution, eventually settling on an application called Demand Commander from a vendor known as Demand Foresight.
The developer claims that Demand Commander allows manufacturers and distributors to develop forecasting models that not only pinpoint the current demand patterns for their individual companies, but also have the ability to “learn” how those patterns change over time and automatically adjust forecasts accordingly.
When integrated into sales and demand planning activities, Demand Commander can help reduce a company’s item-level forecast errors at least 25 percent to 50 percent, according to Gene Tanski, CEO of Demand Foresight, who stands behind the claim with a money-back guarantee.
Turner can’t say whether New Belgium’s forecast error rate has dropped at least 25 percent since adopting Demand Commander, but he hasn’t requested a refund.
“Our mean absolute percent error has come down considerably—and has remained consistently low—since we adopted this solution,” is all Turner can say. “Now that we have reliable demand plans, we are not jerking our production people around. With our previous system, the production people hated me because I was constantly sending them major scheduling changes. Now, when we have to make a change, it’s only a subtle change rather than something that causes major disruptions.”
Turner says he chose Demand Commander because it appeared flexible enough to support a growing, constantly changing business like New Belgium. With assistance from Demand Foresight, he says, the program was configured to fully support New Belgium’s demand planning efforts at the time it was adopted—and the system has adapted as the company has grown and changed.
The system really has excelled at helping New Belgium identify and manage its primary supply chain constraints, which change frequently, Turner says.
“Because we’re always in expansion mode our constraints do shift,” Turner explains. “Sometimes it’s the brew house; sometimes it’s packaging. Right now, our primary constraint is the filtering and finishing process.”
That particular constraint exists because New Belgium still produces all of its beers in a single plant located at its Fort Collins headquarters even though it now ships product to 26 states stretching from California to Georgia and Tennessee. “We’re exploring the possibility of establishing regional breweries on both the east coast and west coast,” Turner says, “but we want to be sure those regions consistently have the volume to support those facilities before making that investment.”
For now, he says, New Belgium is successfully managing its supply chain by using Demand Commander to track customer demand and generate production schedules that account for the company’s capacity constraints.
Turner says Demand Commander has proved to be the ideal solution to support New Belgium’s sales and operations planning (S&OP) process.
Turner, as the company’s chief supply chain planner, keeps track of what distributors are shipping to retailers in order to create a top-down forecast. At the same time, the New Belgium sales teams are talking directly to the distributors about the specific promotions they plan to run at individual stores in their territories over the coming weeks. This exercise, known as Market Plan Tracking, results in a bottom- up forecast.
During the third week of each month, a cross-functional team with representatives from the sales, supply chain and production groups meets to discuss both sets of data and develop a single forecast that will drive production for the next month. Before this meeting takes place, both the top-down and bottom-up numbers have been placed in Demand Commander, which develops initial baseline forecasts for both sides.
“On Monday of that third week, the supply chain group meets with sales to smooth out its forecast,” Turner says. “On Tuesday, we work with production to identify any concerns they might have relative to the forecast. On Wednesday, subgroups from both organizations meet with a management team to iron out any issues and create a final forecast and a monthly production schedule.”
Once a production schedule is in place, Turner continues to monitor supply chain activity via point of-sale data and EDI notices on upcoming orders that flow directly into Demand Commander. Keeping tabs on these data allows Turner to determine if the current production schedule is in line with actual demand, or if he needs to alert the production department to make the “subtle changes” necessary to keep production in sync with the demand.
“This system has helped us maintain consistently high customer service levels—while keeping inventories within a level we can live with—even as our business has become more complex in terms of the number of products we sell and the number of customers we serve,” Turner concludes. “We also are pleased with the role Demand Foresight has played as a business partner. They understand that our business is constantly evolving, and they continue to personalize the system to meet our needs.” – Sidney Hill