It seemed like a no-brainer back in the fall of 1999. Nike Inc., the No. 1 athletic shoemaker, was ready to spend US$400 million to streamline and further automate the way it produces, ships and sells shoes. Instead of taking a month to plan and start producing a new line of sneakers, a new software system would narrow that window to a week. By being able to better match supply with demand, the company thought it could avoid getting stuck with warehouses full of shoes that had gone out of style, while boosting sales of the cool pairs that everyone wanted.

It hasn't worked out that way, and Nike's predicament is an expensive example of an increasingly common problem with technology when it is adopted too quickly and too widely.

Last week Nike announced that it expected its earnings for the latest quarter to fall by one-third to about $0.35 a share - dramatically lower than the earlier projections of $0.53. The culprit? The new supply-chain software, which executives at the Beaverton, Oregon, company say never worked right, caused factories to crank out too many unpopular models like the Air Garnett III and not enough of the trendier ones, like Air Force Ones.

"This is not a particularly fun hour for me," Nike Chief Executive Officer (CEO) Philip Knight said last week during a conference call with analysts, several of whom downgraded the company's stock the next day. "I guess my immediate reaction is, 'This is what we get for $400 million?'" Knight stopped short of slamming i2 Technologies (ITWO) - the infrastructure firm that provided the new system - and insisted that Nike would fix the problem and realize big savings. Though unspoken, Knight's message was that i2 had struck out.

Yet supply-chain experts say Nike should have seen the disaster coming. "Blaming the software vendor is a very old practice," says Pierre Mitchell, an analyst at AMR Research, explaining that fiascos like this occur all the time but are usually kept quiet unless they seriously hurt the bottom line. Candy giant Hershey, for instance, was forced to spotlight its botched rollout of a new warehousing and fulfillment system when it fell short of profit expectations by as much as 26 percent in 1999.

Mitchell suggests that the Nike chief was disingenuous. "Phil Knight makes it sound like it's a surprise to him," he says. "If he doesn't have checkpoints for these kinds of projects, if he doesn't know where $400 million of his company's money is going, then he doesn't have control of his company."

But the question of who's more to blame is beside the point. Nike and i2 share in making the fundamental mistake of swinging for the fences the first time out. There's too much at risk in a project as big as Nike's to roll out a huge, complicated system all at once. "Whenever you put software in, you don't go big-bang and you don't go into production right away," says AMR Research analyst Larry Lapide. "Usually you get these bugs worked out ... before it goes live across the whole business."

That's how i2 says it wanted to proceed. "Typically we take smaller steps in bringing up components of the system," says i2's Chris Houck, director of solution marketing. Nike declined to talk about it.

The i2 system bore the additional burden of having to adapt to an unfamiliar business. While the software has been effective in some industries - printed circuit boards, for example - it wasn't designed specifically for the shoe and apparel business. "That increases the risk that something is going to go terribly wrong," says AMR's Mitchell. Still, he and others stress that such a problem is not unique to i2 but plagues all supply-chain software makers as they try to branch out to shoes, clothing and other industries for the first time.

So what exactly went wrong? Using the new computer system, Nike mistakenly sent double orders to its factories, resulting in an oversupply of many slow-selling shoes, a company spokeswoman says. Meanwhile, production of its hot-selling models did not keep pace with demand.

The result: To offset shipping delays for hot shoes, Nike has had to send them by plane - at $4 to $8 a pair, compared with about $0.75 by boat, according to John Shanley, an analyst with Wells Fargo Van Kasper. Nike insists it will not have to cut shoe prices.

That doesn't mean retailers won't. The Web site for Foot Locker, the nation's biggest retailer of Nike shoes, last week was promoting deep discounts on overstocks: the men's Air Terra Humara 2, a shoe for off-road running, is now selling for $49.99, half its list price. And the men's Air Garnett III, for basketball, has been slashed 36 percent to $89.99. Meanwhile, some stores have reported difficulties getting Air Force Ones, the popular basketball shoes.

Executives at Nike say it could take up to nine months to get over the supply glitch. In the meantime, the company and i2 continue to promise that the results will be worth the wait. Other footwear firms have taken market share from the swoosh in the last year.

Nike certainly isn't the only corporation to view technology as a panacea. No one less than Federal Reserve Chairman Alan Greenspan last week sang its promise in testimony to Congress. "New technologies for supply-chain management can perceive imbalances in inventories at a very early stage - virtually in real time - and can cut production promptly in response to the developing signs of unintended inventory building," he said.

Indeed, the $7.8 billion supply-chain software market is expected to more than double by 2004, according to AMR Research, and i2, the dominant player, is starting to see some tough challengers. Ariba (ARBA) , for example, has taken steps away from its sales partnership with i2, acquiring and signing deals with rival infrastructure firms. After Nike's announcement, Ariba President and Chief Operating Officer Larry Mueller couldn't resist the opportunity to take a jab at i2 last week: "Not only is the stock market down, but now I find out I'm not going to get my Air Jordans delivered on time."

Jennifer Couzin contributed to this report.