Any Which Way but

By Larry Barrett  |  Posted 2003-11-01 Print this article Print

New Balance had nearly faded into irrelevance six years ago. Then, accurately forecasting demand for its shoes became a sport for the entire company and its retailing partners. Now it's making up ground quickly, in its footrace with Nike.

Right"> Any Which Way but Right

Since New Balance's focus is on the everyday athlete, it garners a fairly stable clientele, based on predictable demographics. That makes it easier to forecast demand accurately.

Neither New Balance's sales force nor Holland really had the tools to deliver accurate forecasts until about two years ago.

Back then, Holland supplied a template in a Microsoft Excel spreadsheet that sales representatives would fill in with their monthly forecasts.

But there was no statistical reliablity.

When Holland first got to New Balance in 1999, she was supposed to collect forecasts from about half of the company's 120 sales representatives, compile them and create overall predictions of what shoes and clothes the company's factories should turn out and when.

But she was lucky if she got 20 sheets back each month. "We were just barely into collecting rep's forecasts,'' she recalls.

The problem for the sales representatives was filling out the sheets consumed a lot of time—as much as a day for the forecasts of larger accounts. They had to pore through reams of printouts to plug answers into Holland's rows and columns. And time spent forecasting is not time spent closing. For salespeople paid on commission, that's like taking money out of their wallets.

The problems multiplied for Holland. The format of the electronic spreadsheets was not protected. That meant, first of all, that reps would delete columns, type in the wrong style names, and move information around as they saw fit. Holland became more a proofreader than a planner; it would take at least a day for her to validate the information on each sheet she did receive and put the answers into the correct form for collating and analyzing. "It wasn't a very streamlined process," she says.

Even then, rolling up the data was hard. She tried to combine the forecasts into a single table, where a planner could see combined results but also 'pivot' into other tables that carried the underlying data from different regions or representatives.

"I tried to create this huge pivot table," she says. "But the file was so large that it would crash on my computer. It was basically impossible to do much analysis off of it."

The information from the field was next to useless. "When we sat in forecast meetings we weren't relying on the reps too much," says Holland. "It was just feedback. We really had to kind of decide what the forecast was without their input."

The seat-of-the-pants approach meant sudden spikes in orders to factories for some products and backlogs of others. In some cases, plants overseas would be geared up, on a contract basis. Then there would be deep valleys of production, when inventory that had piled up was sold off. New Balance executives won't reveal actual numbers, but they admit that there was difficulty in getting orders to customers on time.

Senior Writer
Larry, of San Carlos, Calif., was a senior writer and editor at CNet, writing analysis, breaking news and opinion stories. He was technology reporter at the San Jose Business Journal from 1996-1997. He graduated with a B.A. from San Jose State University where he was also executive editor of the daily student newspaper.

Submit a Comment

Loading Comments...
eWeek eWeek

Have the latest technology news and resources emailed to you everyday.