Like Gold, Adams is a man in constant demand, his cell phone incessantly warbling Beethoven's "Für Elise." Every district manager at the company carries a Nextel phone, allowing spontaneous walkie-talkie communication between any of them at any time—no matter whether they're inside the company's gritty City of Commerce headquarters or driving between grand openings in Texas. And while Adams knows how to delegate details to one of his 18 IT team members, he also knows how to take the reins and write code for a new system himself. "I hate recurring costs," he grumbles. "I hate having to pay someone else every time I want to put a new cash register out in the field." When he decides it's time for a new system, Adams balances the cost of buying a pre-packaged solution and then customizing it, against the effort involved in developing a suitable system in-house. With roughly 40 percent of its merchandise consisting of one-time-only close-out inventory that can't be restocked, 99 Cents Only requires systems that can be updated painlessly, and on the fly, with new, previously unrecognized items. If in-house development wins out, Adams usually can be found at the project's center. "I'm a developer by nature," he says. "Most [off-the-shelf] systems out there don't work for us without a lot of customization. Why pay so much money and take so much time customizing something when you can write something yourself faster?"

When 99 Cents Only Stores decided to expand into Texas, the company set a mid-2004 deadline for finding and opening a distribution center there. A few weeks later, David Gold found an "opportunistic buy," as he puts it. Albertson's Inc. was pulling its grocery operations out of Texas, leaving behind a white-elephant headquarters and warehouse complex a few miles outside of Houston. The Golds estimated that Albertson's had poured about $80 million into the facility since building it in 1995. For $23 million, 99 Cents Only got everything: the building, featuring scores of truck bays, 220,000 square feet of refrigerated storage, and nearly half a million square feet of "dry side" storage, as well as dozens of forklifts, computer cabling and hundreds of office chairs and desks. Adams was given less than four months to get the whole operation up and running.

It was tempting, Adams says, to try to retool the warehouse inventory-control system he'd written three years earlier for the City of Commerce distribution center. But, even as he was coding that system, he knew it would eventually need to be replaced by something more powerful—at the time, the company was planning to triple sales within five years. So Adams decided to use a different system for the Texas distribution center. The first few warehouse-management packages he looked at were maddeningly difficult to customize. If a new category needed to be created, or if a previously unknown item came through receiving, it took a computer guru to make the necessary changes. Adams ultimately purchased a system from HighJump Software Inc.—he won't say what he paid, only that it was, of course, a steal—that has proven so efficient in Texas that Adams hopes to begin installing it in the City of Commerce distribution center in early 2004. Picking accuracy—the work done by Avis Walton and his colleagues—is up to nearly 99 percent in Texas, and picking speeds there are some 15 to 20 percent faster than in Los Angeles, where the accuracy rate is only 90 percent. But putting the new system into the fully functioning, 24-hour, three-shift distribution center in California won't be nearly as easy as it was to start from scratch in Texas. So Adams plans to bring a select group of the Los Angeles-based pickers and receivers to Texas to train them on the new system before he installs it at their home facility.

Still, getting the Texas facility up and running so quickly cost 99 Cents Only Stores more money than expected, dragging down third-quarter earnings to below Wall Street expectations and hitting the stock price hard. The company maintains, however, that the new warehouse will be profitable once the store count in Texas reaches about 30, sometime in mid-2004.

This article was originally published on 01-01-2004
eWeek eWeek

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