How we love our information systems, our management theories, our best-laid plans. And how often they fail us. It reminds me of what the playwright and novelist Samuel Beckett once said about his work. "Each time one thinks one starts fresh, new," he mused. "Yet each time one reaches the same impasse. There are many ways to begin, many roads to it, but always the same impasse at the end."
Now I'm pretty sure that Beckett wasn't thinking about supply chains when he voiced that existential lament. But it made me think about the reason why so many big efforts in that critical area fail to meet expectations. It always seems to come down to the same impasse: people.
Major projects like supply-chain overhauls or new IT implementations would be so easy to carry out without the human element mucking them up. Yet we're constantly running up against this eternal truth. "People by nature," said Linda Peel, senior director of value chain applications with Oracle Corp., "are reluctant to change."
Peel spoke from deep experience. Addressing the SCOPE Spring 2012 conference in Chicago, she related some of the frustrations that she's experienced in her years of advising companies on how to fine-tune their supply chains. There was the dock loader who couldn't read a simple manifest. The company that had consolidated multiple plants, but failed to install enough storage tanks to handle the product at the centralized location. The customer that routinely ordered many times the quantity that it needed, because it didn't trust the supplier's forecast.
One recurring root problem, to be sure - but no one solution. As Peel stressed, "There is no best practice. There's only profit. It differs from company to company."
Even so, Peel was preaching the gospel of an approach that seeks to address a wide variety of shortcomings within supply chains. It's called integrated business planning.
She defined the concept as "the intersection of financial planning and sales and operations planning." It's about aligning sales, manufacturing and finance so that a company can generate a more reliable forecast, then synch its supply chain with the needs of customers.
Again, a noble cause with great potential for failure, if the individuals within a company can't transcend their organizational silos and work toward a common goal. Peel told of the eager salesperson who secures a $50m contract for a product that requires a 12-month production lead time. To execute on that deal would mean disrupting the entire organization. In a heartbeat, the sales genius goes from elation over nailing the big contract to being fired. So much for personal initiative that ignores the need for coordinated effort within the supply chain.
The number-one reason why products fail, said Peel, is that they don't get into the distribution channel in time. That's why customers over-order, as a hedge against failing to obtain items when they're really needed. And many companies do a miserable job of forecasting true demand. Beware the supplier who claims to be 98-percent accurate in the aggregate - often that means it's falling short in key product areas.
Barring the rare industry with steady, unchanging demand (can you name one?), no company ever achieves 100-percent accuracy in forecasting. But nearly every one can improve on its record. Peel cited Land O'Lakes, Inc., the nation's second-largest agricultural cooperative. When she spoke in April, the company was already making butter for Thanksgiving. It needed to know, within a narrow margin of error, how much product it was going to sell seven months in the future.
Even a relatively modest improvement can yield big benefits. With a paper-driven system, Land O'Lakes had been achieving forecast accuracy of around 44 percent. It wanted to bump that up to 50.6 percent, with weekly assessments by item and location. In the process, said Peel, it could lower inventory levels by 7.8 percent. At the same time, the company was driving for a case-fill rate of 98.7 percent, working against a benchmark established by the Grocery Manufacturers Association.
To achieve those goals, Land O'Lakes had to broaden its data inputs. In the past, it had used only historical volumes to develop a statistical forecast. Now it sought to include "demonstrated customer activity" in the demand-planning process. In the event, it could smoke out those customers who were ordering too much product. And it could add the simple input of open orders, to tweak the forecast in line with the real world.
By determining how much a customer needs, the supplier can arrive at a consensus forecast that identifies, to a reasonable degree, what's being consumed and when. At the same time, it can account for individual customer behavior.
"Customers do not collaborate with each other before placing orders," Peel said. So it's vital to employ demand sensing, with minimal latency, "to understand what is being sold, who is buying the product and the impact of the demand-shaping process."
One of Peel's more interesting tasks is the development of individualized business cases for her customers. In actuality, they're not that hard to do. According to Peel, 20 cents of every dollar that goes out the front door of a consumer-goods supplier comes back. And the culprit? "You didn't get the right stuff to the right people."
With the addition of new sales channels via the internet, it's become even tougher for suppliers and retailers to hit their demand-forecasting targets. Profit margins, too, are under constant attack. So maybe the time is right to motivate those siloed individuals to adopt an enlightened attitude toward collaboration, both within and without the company's walls. Even the most stubborn person can change in a crisis. So why not declare one?
Next: Where will the right people come from?
- Robert J. Bowman, SupplyChainBrain
Comment on This Article