Special Report: Metrics + Collaboration = Perfect Order

6/1/2005

Metrics + Collaboration = Perfect Orders
How a new committee hopes to redefine perfect order criteria to reflect the transforming food industry supply chain
-by Lisa Terry

A better supply chain requires first agreeing on what to measure, then sharing that data to collaborate on improvements. That's the underlying idea behind the
Perfect Order Metrics devised by a joint industry team from the Food Marketing Institute and Grocery Manufacturers of America. First developed seven years ago, the metrics were recently updated to reflect the transforming food industry supply chain.

"The way the supply chain was viewed in the past, when the delivery was made to the warehouse, the supply chain ended," says Dee Biggs, director, demand fulfillment services for Welch's, a member of GMA's Perfect Order Committee. "Now we're encouraged to think more broadly: it ends when the consumer takes the product off the shelf."

Changing the Standards
That meant a change was needed in the standards, which the committee hopes will be evaluated and adopted by trading partners in order to improve collaboration by January 2006. While the metrics address the entire supply chain, their use is generally driven by retailers, who select and weight the metrics according to their own way of doing business, provide feedback to suppliers on their performance to those metrics and work with them to improve that performance.

"Some retailers are doing a terrific job, but many others have not done [benchmarking]," says Biggs. "We're encouraging folks to start benchmarking internally on a more formal basis."

One of Welch's more sophisticated retail customers, for example, provides monthly feedback to the company on its refrigerated, frozen and dry products, comparing it to competitors. "We see an average of those numbers -- our performance for the month, quarter, year to date, the whole group average and where we rate," Biggs says. "We have a conference call every month with a variety of people interested in that, including transportation, warehouse, inventory management, field sales, team leaders. We go through the report card, figure out what we did right and wrong and single out issues and problems."

The New Metrics
The initial perfect order metrics included orders shipped complete, on-time delivery, accurate and timely invoice and no damage. The new recommended metrics, presented at the March FMI Distribution Conference and April GMA IS/LD Conference by the committee of two retailers and three grocery manufacturers, are:

Cases shipped vs. ordered. The older metric, orders shipped complete, was more akin to a pass-fail grade. Cases shipped shows how close the order came to perfect, and allows the retailer to set a target percentage for the supplier. It's also based on when the product is scheduled to ship, not the actual ship date.
On-time delivery. This metric changed from the delivery time plus or minus a half hour to eliminate the allowance for being late. Now it allows the order to be up to an hour early, but not late, to be considered on time. The manufacturer is not penalized if the retailer isn't ready to receive. The metric applies to customer pick-up as well.
Data synchronization. Advocated by committee member Marianne Timmons of Wegmans, this metric measures the percentage of items fully synchronized between the trading partners. Data synch is an essential foundation for all manner of e-commerce in the industry. A 2003 GMA random item sampling of supposedly cleansed item data found just 35 percent was accurate, pointing to internal data management issues.
Damage (unsaleables). Measures unsaleables dollars to sales dollars. Manufacturers are encouraged to break out damage that occurs at the manufacturer, carrier, customer warehouse and store level to understand where issues occur.
Days of supply. Measures days of forward coverage at the retailer warehouse and the store as well, if possible, to help identify non-optimum levels of inventory at various points in the supply chain.
Order cycle time. A key metric, measuring length of time between a manufacturer receiving an order and the actual delivery of the order to the customer's warehouse, measured in hours, days or weeks. Reducing the cycle time in turn reduces inventory investment.
Shelf-level service. Considered by the committee to be the most important metric on the list, but also one of the hardest to measure. The goal is a fill-rate for each item. For example, if the demand for an item is 20 units per day and 20 units were sold, that's 100 percent service. Right now, many retailers use voids as a metric, gathered by sending a worker around the store to look for shelf holes and enter those SKUs into a hand-held. "It's not terribly accurate," says Biggs. "A lot of times products are moved on the shelf. It's not obvious." Pattern recognition software can help retailers analyze product movement to detect true out-of-stock levels, he says.

Now that the metrics are in place, evangelism is the next task, Biggs says. "By next year we want to be able to show two trading partners that embraced the idea, figured out what to measure and develop a case study to show where to start, where to go, and what the implications are."