EVERY YEAR AROUND LABOR DAY, JUST BORN maker of those gooey pink and yellow marshmallow chicks called Peeps, begins to get ready for its big Easter season. It turns to its sales forecasts, and it purchases mass quantities of sugar, food coloring and packaging to meet the anticipated demand. And every year it gets stuck with leftover Peeps and unused materials that cost it sacks of money.
Just Born manufactures 600 million Peeps a year, 80 percent of which are sold during the Easter season. The 78-year-old, almost $100 million Bethlehem, Pa.-based company also makes Peeps for Valentine’s Day, Christmas and Halloween, but Easter is make-or-break time.
Just Born depends heavily on monthly sales forecasts based on historical data, much of which is gathered from sales representatives who have to wrangle that information from independent candy brokers, whose interests do not always parallel Just Born’s. The data Just Born does manage to retrieve is therefore, to use Just Born Supply Chain Director Don Petraitis’s word, “problematic.”
“[Our forecasts] always change as it gets closer to the season,” Petraitis says. “Either we’re caught short and can’t respond to demand in time, or we don’t sell enough and are left with overstock.”
Petraitis, a 10-year operations veteran with Just Born who admits he had “no formal training or experience” in dealing with the supply chain operations before he was put in charge last June, has already figured out that forecasting software isn’t the answer to the company’s problems. The IT department had tried using Excel spreadsheets, then a forecasting software package from Demand Solutions, but the more sophisticated application did not lead to just-in-time Peeps. It was the quality of the information that mattered, not the software. Now Just Born hopes to link with its suppliers through the Web in a fully automated supply chain, although Petraitis says that’s down the road a ways.
The good news for Just Born is that at least it’s thinking about supply chain management.
The bad news is that it took them so long to do it.
Ramping Up and Winding Down
LOTS OF BUSINESSES, FROM ful-fillment and delivery to clothing manufacturers, lawn and gardening suppliers, and oil companies have to deal with surges in demand that occur predictably around the winter holidays or the onset of hot weather. Supply chain management is relatively easy when it’s handling steady state demands; the real test comes with handling huge seasonal fluctuations.
“If a big part of your revenue comes from activity that goes on in a seasonal period and you don’t build to support it, you are out of business,” says Herb Kleinberger, leader of the global retail practice at New York City-based PricewaterhouseCoopers. “You’re toast.”
Case in point: Computer City, the chain launched by RadioShack in the mid-1990s. Designed to compete with stores like CompUSA and Circuit City, it failed after four years because it didn’t know how to manage its inventory, says Chapman Kistler, vice president of the supply chain service line at Cap Gemini Ernst & Young, a global management consultancy based in New York City.
“RadioShack’s core competency was in stocking one of everything,” Kistler says. But when Christmas came, having one of everything didn’t help Computer City cope with a run on new ink-jet printers.
Just Born’s Peeps problem is a classic example of poor supply chain management. “Everyone is guessing along the supply chain as to how much to make and ship,” says Kleinberger. “And they’re all guessing in separate ways from each other, so everyone builds in extra safety stock to buffer forecasting errors. You add it up along the chain, and you end up with a huge cork that’s clogged with excess capacity.”
Or, in this case, pink and yellow Peeps.
The cost of sitting on excess product varies by category and industry, but it can be as much as 10 percent to 25 percent of distribution costs, according to Kevin O’Marah, the service director of supply chain strategies at Boston-based AMR Research. Transporting, insuring and storing extra capacity can increase overall costs by 10 percent or 11 percent, particularly during peak times, O’Marah says.
Planning ahead to predict demand, staffing appropriately to handle spikes, knowing how to make the tough purchasing and warehousing decisions, and ensuring the proper collection of data for forecasting are all components of supply chain excellence vital for companies that experience seasonal fluctuations. Even if you don’t do most of your business around various holidays, chances are you have to deal with seasonal fluctuations of one kind or another. “Very few companies have no seasonal peaks,” Kistler says. Or, for that matter, dips.
(Industries that typically don’t experience seasonal surges include pharmaceuticals, fast-food restaurants and B2B distributors in the aerospace sector.)
If you haven’t already started thinking about managing your supply chain operations to deal with spikes, you’re late, but maybe you’re not yet toast. Here are three steps to help you build a flexible supply chain that can handle those scary seasonal ups and downs.
Dynamic Databases or How to Store 360 Million Pizzas
ON A NORMAL SUNDAY, DOMINO’S Pizza sells about 750,000 pies. But on Super Bowl Sunday, it sells 1.1 million, a 46 percent spike, says CIO Tim Monteith.
Domino’s business also spikes on weekends and whenever the weather gets nasty. There seems to be something about rain, sleet and snow that makes Americans simultaneously hungry and too lazy to do anything about it other than picking up the phone. To stay on top of demand, Monteith relies on his data warehouse to keep his business running.
Ann Arbor, Mich.-based Domino’s has 4,850 stores nationwide, and each store sells about 1,000 pizzas per week. In 2000, Domino’s made 270 million pounds of dough and used 146 million pounds of cheese, 13 million pounds of Italian sausage and 9.9 million pounds of mushrooms to make more than 360 million pizzas. The demand for materials goes through the roof in the days prior to major spikes, Monteith says.
“To handle peaks, you must have the right tools,” he says. “You must have a good database to learn what causes spikes and how to predict them.”
Because each store can experience wild fluctuations depending on local conditions, Domino’s database keeps historical data on sales at all 4,850 stores. Computers at each store automatically tabulate the amount and nature of supplies used each day, which allows the store managers to know exactly how much needs to be ordered and what is in danger of going bad.
“The inventory is dynamically adjusted,” Monteith says “If [a customer orders] a pepperoni pizza, the computer will deduct one roll of dough, 5 ounces of cheese, and a unit of sauce and pepperoni from the system.”
Monteith ties the data gathered from the automated ordering system tightly to sales forecasts, and that provides him with a daily history that is useful when planning for the next Super Bowl Sunday or even just the next rainy Sunday. He uses Manugistics Group software (see “Crystal Ball Technology,” Page 104) to manage Domino’s warehouse and distribution system. To further organize the data from each store, Domino’s is rolling out an ordering and forecasting system by Breakaway International Systems that will be up and running in U.S. stores by 2003. The new system includes software by Intellisol International that will help the stores forecast their staffing needs depending on sales levels.
“Data aggregation and trend analysis are where the CIO and IT organization can provide the most value,” says Bob Moncrieff, a director at PRTM, a global management consultancy based in Waltham, Mass. In creating a supple supply chain, “the most important step to take is creating a well-structured data warehouse in which you can store, format, and manipulate supply and demand data,” says Moncrieff. “This will help you see seasonal issues very quickly.”
Integrated Channels or You Can’t Wait Until the Bloom Is Off the Rose
PROFLOWERS.COM, AN ONLINE FLOWER dealer, does 40 percent of its annual business on just two dates: Valentine’s Day and Mother’s Day. It also sees a surge, though less dramatic, in demand around Christmas, Thanksgiving and Easter.
During Mother’s Day week, the San Diego-based company ships about 250,000 flowers and plants, as opposed to 12,000 in a nonpeak week, says Mark Sottosanti, vice president of planning and logistics.
Unlike clothes and computers, or (to a certain extent) pepperoni and Peeps, flowers are very perishable. They must be grown months in advance of when they will be needed, and they begin dying the moment they’re cut. Most flowers can live a maximum of 10 days once cut, Sottosanti says.
“It’s crucial from a product and a business standpoint that we know what our growers have and what sales for our spikes will be,” he says.
In order to work within those 10 days, the company uses a proprietary Web-based system to link its independent growers to its network. Unlike online flower sellers FTD or 1-800-Flowers, which route orders through a network of growers, wholesalers and florists, Proflowers.com ships its flowers directly from the growers. When a customer places an order on the website, it is stored in the system and automatically routed to a grower a day or two before it is scheduled to be filled. So if a customer orders a dozen roses on Jan. 15 to be delivered on Feb. 14, Valentine’s Day, the grower will not see the order until Feb. 12 or 13. The order will be shipped the next day via refrigerated truck to a FedEx ramp and then flown to its destination. Because the growers give daily inventory updates, the company always knows how many of each type of flower it has to work with.
If the worst happens and the growers run out of roses on Valentine’s Day, the website will stop taking orders for roses.
Proflowers.com monitors the inventory data and orders so that “if we see that a flower is on track to be sold out, we’ll bump up the price to suppress demand,” Sottosanti says. And the system will not take orders for products that are not in stock.
If there’s excess inventory sitting around after a spike, the company will offer a free vase or a discounted price to try to move it. One time, Sottosanti says, a misjudgment in forecasting led to a surplus of 20,000 tulips after Easter. The company donated them to a local festival. Only about 1 percent of the flowers ordered annually have to be thrown away, he says.
When Proflowers.com needs to ramp up for seasonal peaks, Sottosanti outsources his extra call center and e-mail needs to third-party contractors for two- to three-week periods. When the peak ends, so does the contract.
Creative Staffing or It Takes a Village to Sell That Much Fruitcake
HARRY & DAVID, THE CATALOG GIANT that sells fancy fruits, candy, preserves and fruitcake, normally gets by with 300 employees in its two call centers. But every year beginning in September, the company begins to ramp up for Christmas by hiring 3,000 seasonal employees and opening a third call center.
Talk about being prepared; Harry & David puts the Boy Scouts to shame.
During the Christmas season, Harry & David, which is owned by Medford, Ore.-based Bear Creek Corp., processes about 75,000 orders and ships more than 250,000 packages a day, as opposed to 500 orders and 18,000 packages a day in the off-season. So the temporary staff has to be comfortable with the job at hand, says Rod McLeod, CIO of Yamanouchi Consumer, the parent company of Bear Creek.
The company has two permanent distribution and call centers?in Medford, Ore., and Hopewell, Ohio. To handle the seasonal surge, it leases space and equipment for a third, temporary center in Eugene, Ore., which typically holds 500 workstations that are staffed during peak hours from the day after Thanksgiving to Jan. 15. The lease is multiyear, though the space goes unused the rest of the year. Temporary staff starts a rigid training program in October, months before the peak season kicks into gear. The program costs a lot, McLeod says, but he believes it’s a necessary expense, and it’s built annually into Harry & David’s budget.
“To handle a peak, you must pay attention to the way you staff up,” McLeod says. Once the peak has passed, you have to staff down. “It’s difficult to hire someone, train them for a few days, have them work for you for six weeks and then say good-bye.”
In order to make sure it gets good people for the temporary work, Harry & David offers a highly competitive paycheck and employee discounts that the temps can use for their own Christmas presents. McLeod says that the temps come back year after year. As a result, customers always get a trained voice on the other end of the line.
Bear Creek has a department dedicated to training seasonal employees. Last year, that practice bore fruit when heavy snow and ice closed the Ohio call center for a time between Thanksgiving and Christmas. Within 48 hours, McLeod had a temporary office in a conference center in Medford set up for a SWAT team of 50 seasonal and permanent employees ready, willing and trained to man the phones. Disaster averted. If the shadow staff hadn’t already been trained, McLeod believes the triage would not have been successful. It’s common practice for companies to contract with third parties if they need extra warehouse space or a temporary building to handle increased seasonal demand, but when it comes to staffing, the question of whether to outsource becomes more complicated. “If the positions you want to outsource are customer-facing, be very careful,” Cap Gemini Ernst & Young’s Kistler says. “You need a flexible workforce, but it’s terrible for a customer if they call during a holiday and they get a first-day trainee on the phone.”
The HoneyBaked Ham Co. of Georgia learned how not to outsource its catalog call centers the hard way. The 42-year-old, $130 million company does almost 60 percent of its business during the Thanksgiving, Christmas and Easter holiday seasons, and customer orders for those holidays begin three to four weeks out. To handle the rush, the Atlanta-based company contracts from October to January with a third-party call center. Several years ago, it experimented with outsourcing the entire customer fulfillment process and got the door slammed on its fingers.
“We outsourced everything?from the order call to inventory management and shipping,” says Chuck Bengochea, COO of the Georgia HoneyBaked. “It didn’t work out. We ended up disappointing far too many customers, and we took it back in-house.”
The company eventually took the service provider to court over the problems it suffered?which, according to Nancy Gibson, senior vice president of marketing at the Georgia HoneyBaked, included more than 20,000 orders that did not ship on time.
“It’s one thing if a toy doesn’t arrive on Christmas,” Gibson says. “It’s a whole different story if your Christmas dinner doesn’t show up.”
Designing your supply chain operations to handle seasonal spikes demands taking a hard look at your data and planning on a continuous basis. You don’t have much choice about it. Either you do it and stay in the game, or you don’t and you end up eating year-old Peeps.
Not a happy prospect.