softening the bullwhip's sting

Remember the old parlor game where a secret's whispered from person to person? The variance between what the first person said and what the last person heard would invariably produce a laugh.

In the supply chain, however, it's no laughing matter. Students of the supply chain know that variance as "the bullwhip effect," the concept that the farther up the supply chain you go and the further removed from actual customer demand you become, the less predictable and more variable order quantities will be.

Every distribution center manager knows the frustration of trying to match inventory to customer demand. The gap between real demand and the forecasts that businesses act on is painfully familiar. Sherman's Law of Forecast Accuracy states: "Forecast accuracy improves in direct correlation to its distance from usefulness." At an aggregate level, forecasts may approach 90 percent accuracy and beyond. But at the shipping dock, where the actual orders must be filled, the forecasts for exactly what items have to be shipped to precisely what locations often prove reliable less than 60 percent of the time.

To address that, companies can turn to collaborative planning technologies and strategies designed to provide earlier notice of forecast variations to the DC without increasing overall lead times. By starting at the shipping dock and sharing information—such as the scheduling of promotional and pricing events or gaps between, say, what the supplier plans to ship and what customers intend to order—companies can increase the time the operations people have to react when the forecast proves wrong.

Most events that contribute to forecast error occur well before the order needs to be shipped. Supply chains rarely run completely out of stock. Given sufficient response time and visibility into inventory, companies can quickly redeploy goods where they're needed. Today's collaboration software can alert managers to demand variances in real time, thus increasing response time. Giving a logistics manager even a day more to react can significantly reduce expediting expenses, premium transportation costs, and other "firefighting" costs while improving overall customer satisfaction.

Though the advantages of collaboration are fairly generally understood, companies all too often focus on the wrong measures when evaluating their collaborative initiatives' effectiveness. To avoid that trap, we offer the following suggestions:

• Don't look at what the forecast variance is, look at why there was a variance. Historical information tends to produce hysterical results.Most forecast variance comes from new events, usually created by sales and marketing or a customer's merchandising strategies. Do you have visibility into these events before you have to fill the order?
• Consider your suppliers and customers as partners not adversaries. They probably didn't intend to ruin your day; they just knew something you didn't. Get to know your counterparts on the other side of the dock. Share your plans with them and ask for validation or deviations from their plans. You'll be surprised by how much information they'll share that will improve your response to unplanned events.
• Don't set objectives to reduce the cost of overtime, premium transportation and other reactionary expediting costs. Do set objectives to identify and eliminate the causes of these expenses.

The cost of forecast error isn't realized in the planning department or headquarters; it's realized at the shipping and receiving docks. Improving the quality of information and processes to increase response time starting at the DC floor will provide a faster, larger return on investment than initiatives that begin with a better forecast. If you can't ship it, you can't bill it. Start at the point where cost and revenue meet.