But go to Nucor's headquarters and you won't find an executive vice-president in sight. That's because they have all been banished to Nucor's 16 plants -- where they are busy running them. "Plant managers have to be good business managers," notes Nucor chairman Iverson. "Each plant does its own marketing and sales. Each division is a stand-alone business with its own operating targets."
Larry Roos, with 21 years' experience in the steel business, serves as plant manager at Crawfordsville. This year the mill should produce 1.8 million tons of steel, up from 1.1 million in 1993. Roos spends much of his typical 12-hour day roving through the mill in a flame-retardant jacket and steel-toed shoes. Roos (pronounced Rose) understands the operation so well that he can boil its performance down to a single number.
Nucor's conversion cost -- the cost of turning a ton of scrap into a ton of finished steel -- animates the operation and drives the decision-making process. That number, says Iverson, is currently around $170, and analysts believe the company has a $50 to $75 advantage over competitors, making Nucor a low-cost producer in the industry.
Nucor keeps pounding away at its conversion cost by insisting that the management stay close to its roots. Under Roos at Crawfordsville, which employs 475 people, there are just seven department heads and 24 supervisors. All have come up through the mill, and all still spend much of the day there. In a good year more than half the compensation received by Nucor department heads and supervisors is based on return on assets. That directly rewards -- or penalizes -- them for decisions they make regarding the manufacturing process and capital expenses. Put in the wrong kind of furnace and your paycheck gets singed.
In the past two years Nucor workers have lowered the time it takes to melt the steel from 72 minutes to about 65, which has allowed for the pouring of an additional 25 tons of steel during a typical 12-hour shift. Nucor owes that increased yield not to any one revolutionary change but to a host of small changes effected by workers. Their contributions range from tinkering with the chemistry of the melt to replacing a 4-foot-long exhaust pipe with one measuring 10 feet.
Crawfordsville is in permanent evolution. Workers are doing so much experimentation, Roos says, that "half the time I don't know who's doing what out there." Dave Smith, manager of the melt shop, recalls a recent instance when the grade of sand used to plug the "taphole" in the vessel carrying the molten steel solidified, impeding the pouring of the steel into the caster. Workers started bringing in different grades of sand and gravel. One day Smith looked out his window to see workers shoveling limestone out of the driveway.
In a climate of rampant experimentation, "we do have failures," admits Iverson. He cites the replacement in one instance of an electric arc furnace with an induction furnace -- an $8-million mistake, but one that seems to have left Iverson unfazed. "The real problem much of the time is that people don't take enough risks," he claims.
At Computer Associates, by contrast, the management leaves little to chance. Risk taking comes from the top down. Charles Wang promotes a process dubbed zero-based thinking, which routinely reexamines cost and strategic questions on all projects. "We are always asking, 'Why are we doing this? What do we need to get done?" says Russ Artzt, executive vice-president and chief of research and development.
Philip Montrowe, a manager of development support, says that Wang serves as the master inquisitor in a business in which technology often seduces its creators, obscuring the market's needs in the process. "The road to being a $2-billion company is littered with software companies that built better mousetraps but couldn't persuade people that they needed to catch mice," Montrowe says.
Computer Associates is run by a triumvirate of programmers: Wang, president Sanjay Kumar, and Artzt. Under them are three more seasoned technicians, and together all six track every one of the company's 300 products. "I take products home myself on the weekend" for quality testing, says Wang. "I love technology -- and I know ours is better than other people's."
Wang's assured zeal filters down through the organization and creates a mentality more Parris Island than Long Island. "People feel good when you push them. People who push themselves, we take care of them," says Artzt.
At Computer Associates the bootstrapping ethic comes down mostly from the top, rather than through worker initiative. "In the Computer Associates' Bill of Rights, you have the right to make a mistake," says Sanjay Kumar, echoing a dictum of Wang's. "You don't have the right to cover your ass over that mistake." Yet mention of a bill of rights does not guarantee democracy. "We're really more of a benevolent dictatorship," Kumar concedes.
Wang's "benevolence" keeps the organization on edge. And his omniscience penetrates, it seems, to all corners of the company. Last year Wang heard from customers that the sales force was making excessive promises. Ever the hands-on leader, he herded a handful of salespeople into a conference room. "I told them, 'OK guys, 'fess up," Wang recalls. Soon a tense session turned into a mass confessional. "This is fundamental to how we operate," says Wang.
* * *
The Workforce: A willingness to be 'reinvented'
A bootstrapper of any size, Charles Wang believes, cannot succeed with workers who are just smart and hard charging. In the people who work for him, Wang prizes flexibility above all. In most big companies, that quality is lost as roles become narrowly defined, fiefdoms spring up, and the organization ossifies. "I don't want people whose livelihood depends on fulfilling a certain function," says Wang, verbally hammering away, as usual, in his sixth-floor office.
