Print this article | Return to Article | Return to CFO.com
Companies have learned some useful lessons about demand planning, but will improved forecasting techniques guarantee happy holidays?
David M. Katz, CFO Magazine
July 15, 2010
For the first few months of this year, the U.S. economy was moving steadily toward recovery. The Credit Managers' Index (CMI), a key indicator of corporate economic trends, registered a rise each month from August 2009 until April 2010, and other benchmarks gave cause for optimism.
Then came May, with the BP oil leak, worries about European sovereign debt, disappointing employment data, and a sudden drop in the CMI index. Talk of a double-dip recession resumed, and CFOs were torn: Should they ramp up for growth or once again batten down the hatches?
The dilemma as to whether to reduce inventory if another downturn is coming or gear up for a resumption of growth is particularly acute now, as companies debate how to prepare for the holiday shopping season.
"No one is confident that they should buy up into this fourth quarter in a significant way, but, by the same token, they don't want to miss opportunities," says Bryan Eshelman, a retail and consumer products specialist with business advisory firm AlixPartners. "So it's really investing in the capability to react quickly that will help those that are most successful this fourth quarter."
Fortunately, the Great Recession prompted many companies to reevaluate their methods of demand forecasting and planning. Companies now test the market more frequently, assessing customer intentions via such means as dipping into markets tentatively before plunging into huge inventory orders. Merchandising methods, dubbed "hold-and-flow" and "delayed-distribution," are increasingly helping managers to look before they leap.
At the same time, corporate forecasters have grown more cognizant of the role companies themselves play in sparking or dampening demand. Rather than rely solely on such venerable prediction metrics as same-store sales, they are attempting to gauge how their own pricing, advertising, and sales practices affect customer behavior and how they can refine those practices to boost sales.
For these practices to succeed, some experts say, CFOs will need to begin linking sales predictions to production planning, instead of allowing the two functions to operate separately. At Coach, a maker of luxury handbags and other wardrobe accessories, CFO Mike Devine says that his role is "to make sure the organization maintains discipline in having our production forecasts line up with our sales forecasts." Rather than enforcing preexisting production, inventory, and sales plans, finance chiefs must be able to make quick adjustments as conditions change.
Before the recession, CFOs typically approached forecasting and planning "on a strategic basis," says Charlie Chase, business enablement manager for software company SAS Institute. "Now they need to do it on a tactical basis," he says. Think of it as an early Christmas present the company can give itself.
Going with the Flow
The recession schooled finance chiefs on the inherent risks of placing large inventories in a supply chain. Previously, companies typically made or bought products and distributed them to stores in bulk because that was the cheapest way to do it. Until, that is, demand dried up and companies had to take big write-downs and unload the goods at deep discount, if at all.
Some companies now postpone the manufacturing of goods until they are reasonably sure there is a market for them. More specifically, companies defer completion of a certain portion of a product line. "Previously, we would convert all raw materials purchased into finished goods," says Coach CFO Devine. "Now, we take a deeper position in continuing raw materials [but hold off on manufacturing]. This allows us to respond more nimbly to consumer buying patterns."
Coach might, for example, load up on more leather or silk than it needs to manufacture products for a given season, and use the remaining materials for later seasons. If demand suddenly does spike, however, "we'll have raw materials available to quickly up production," Devine says.
Similarly, some companies are stocking up at their warehouses but delaying the shipment of finished goods to stores. In a May earnings call, Saks announced that it would install such a "hold-and-flow system" in phases beginning this summer. The system "will drive allocation effectiveness by holding back a portion of certain merchandise orders at our distribution center…employing the product through the stores as demand dictates," says Ron Frasch, the luxury clothing chain's president and chief merchandising officer.
Such techniques do add warehousing costs, but also have the potential to boost revenues and gross margins. That's because improved precision in meeting demand makes it more likely that goods will be sold at full price. If they're used successfully, "the benefit of full-price sell-through outweighs any additional expense to warehouse the product," says Matthew Katz, a managing director at AlixPartners. "If you warehouse product and then don't get the full-price sell-through, however, then you've got a problem."
Companies are also adopting a more sophisticated approach to forecasting that includes probing the effects of their own actions on customers. Previously, companies relied solely on past sales-trend data, along with such well-worn gauges as seasonal effects on sales, as a basis for their forecasts, according to SAS's Chase. "Then, whatever is left over is called 'unexplained' or attributed to randomness," he says. "But, in fact, it can be attributed to sales promotion, marketing, price, and [the accuracy of your] economic forecast."
In his recent book, Demand-Driven Forecasting, Chase argues that managers imperil their predictions and their profits if they ignore such factors. "For example, a price change occurring simultaneously with a product sales promotion could erode the profitability of the product or create an unexpected out-of-stock situation on the shelf at the retailer," he writes.
Yet such internal factors must be balanced against broader economic and social factors. Two years ago, OfficeMax, a retailer that does a big business in school supplies, ran a promotion in which crayons, glue, and other specified items were sold for a penny apiece in an effort to lure buyers to shop for all their back-to-school needs at the store.
Unfortunately, this coincided with the start of the recession. Instead of filling their baskets with a mix of loss leaders and full-priced items, OfficeMax customers spread their spending across a variety of competing stores on the basis of price, according to Reuben Slone, executive vice president of supply chains at the company. "People had a lot more time than money," he says, "because unemployment was really beginning to mushroom, and they were literally watching their pennies." Slone says the promotion failed because of the most common kind of error in forecasting: an error in assumptions about human behavior, rather than a numerical miscalculation.
Now, Slone says, OfficeMax layers a bevy of external factors into its sales forecasts, everything from "competitive entry/exit" (what happens to demand when a Staples or Best Buy enters or exits a local market) to weather patterns. The company even employs a "hurricane algorithm," which it uses to gauge how many bottles of water, flashlights, or batteries people will buy in advance of, or after, a storm.
So what do such methods tell Slone about his company's prospects for Christmas, which is a key selling season for office products? "Overall, we're cautiously pessimistic," he says, noting that the company's business is highly correlated to the nation's employment numbers. "We're not planning for a huge resurgence."
In that, he doesn't appear to be alone.
David M. Katz is New York bureau chief at CFO.
A Demand-Planning Action Plan
In these volatile times, it's hard for companies to get a reliable read on what to expect from their customers and how to deal with rapid shifts in demand. Here are four steps that supply-chain experts say are essential to coping with a fast-changing economic landscape:
1. Ditch the long-range forecasts. Rely on rolling forecasts as a corrective to old outlooks that are likely already stale.
2. Avoid gut feeling. Previously, many executives acted on forecasts based on historical sales trends, which were then "adjusted" according to the "judgment" of sales executives. That's a recipe for error. By all means, use the assumptions of experienced managers as a starting point. But be sure to test their hypotheses with rigorous research.
3. Go granular. Don't settle for aggregate forecasts of the fortunes of the entire corporation. Seek data breakdowns by geography, product, and means of distribution (sales via catalog, online, or store, for instance).
4. Launch an S.O.P. "Sales and operating planning" committees are hot in demand-planning circles right now — and rightly so. Composed of executives from both inventory and sales, they can provide collective intelligence to finance executives and add more depth to forecasts of customer behavior. — D.M.K.