Dissension in the ranks -- or at least unfettered free enterprise at work -- is the key to Spectrum Associates' remarkable growth. Should your company look like this?

It's part of psyching out a competitor that sometimes you talk a little trash. In sports, you train your disdain on anything from your opponent's moves to his mother. In the marketplace, you dis the product. It's where the enemy lives.

So it's predictable that two rivals introducing competitive software products would indulge in reproaches and threats. "They're in deep left field on this product," says one, referring to the other's upcoming release. "It's a $50,000 Cadillac, when what the market wants is a $10,000 bicycle," adds his partner. "It's priced all wrong." Desperate for market share, these challengers are conspiring to field a low-end alternative and wage war over price. "We'll sell it for thousands less. Absolutely. We'll give it away if we have to."

"They're nave. They don't understand what the market wants, and they'll never pull it off," retorts John Nugent, a cofounder of Spectrum Associates. His small Woburn, Mass., software company has spent nearly $5 million and two years developing the disparaged product, the first of a series of manufacturing-software products called Point.Man, in a bet-the-business gambit to be taken seriously as a software company.

Nugent and his partner, Tony Baudanza, are prepared to defend their product, slated for release this spring, against all comers -- the Goliaths who have outspent them as well as the gadflies who would undercut them. "The future of this company is riding on it," says Nugent. As for Jack Saint and Finbar Crean, the pair who have pledged to capture the low end of the market and bedevil Point.Man in the process, they're not causing Nugent and Baudanza any sleepless nights. The Spectrum principals know what those two are up to. They employ them.

Longtime managers at Spectrum Associates, Crean and Saint profess no plans to start their own company. They figure they can launch an assault from their bower on Spectrum's payroll. "We're just going to go ahead and do it right from here," says Crean, a sly smile curling beneath his mustache. "The next thing they know, the product is going to be sold, and there's nothing they can do about it."

"Don't you see," asks the boyish Saint, "why we love it?"

Some people just beg for a pink slip. Crean and Saint, engaged in a willful act of sedition, punishable under ordinary circumstances by termination, appear to be asking for the door. "Nah," says Nugent. "It's just part of the game."

At Spectrum, a scrappy little company that has long fostered a fighting spirit within its ranks, this latest conspiracy is business as usual. "Oh, I'll whisper in their ear and tell them I think they're wrong," says Nugent. "But we're willing to let them go on because one day they may be right. And I'd rather find that out while they're here than have them prove me wrong from the outside."

Hire thine enemies: is it a new commandment? For Baudanza and Nugent, it's the golden rule of growth. And if the insolence of Crean and Saint's scheme should offend some, it heartens the founders of Spectrum. They have sown it. It is the inevitable consequence of a strategy and a culture the founders uphold as an homage to free enterprise and the secret to the company's considerable success.

"We've kept the company growing," says Nugent, a compact man who smokes two packs of Kools a day and appears older than his 35 years, "by making sure nobody gets comfortable." Not even the owners. "The fear of losing has been our greatest inspiration."

In just five years closely held and self-financed Spectrum Associates has grown more than 6,000%, to an estimated $25 million in revenues last year. Operating profit margins in recent years have exceeded 24%, Nugent maintains. And healthy cash flow has rendered debt or even a line of credit unnecessary. All of it, the founders aver, is the harvest of healthy competition within as well as outside the doors of their modest offices.

"Look, it's the reality of capitalism that you have competitors," reasons Nugent. "Our people learn to face that reality inside the company first. The basic principle of a free market means you let the best win. We happen to believe that."

Like a scrimmage before the game, such competition prepares the company for the battles it undertakes outside. "It creates a micromarket right here," Nugent says. The sparring lets "us apply the pricing pressure, the delivery pressure, and the growth pressure that we encounter in the marketplace." The result: a work force in a perpetual state of readiness. "It keeps us healthy," he says. "A little insecurity can be very healthy."

Infighting is a time-honored pastime in American business. In the glory days of mainframe innovation, IBM was renowned for internal jousts among competing groups. Fifteen years ago at Data General a product-development duel wrought a celebrated minicomputer and a Pulitzer Prize-winning book that chronicled that computer's development -- The Soul of a New Machine. Overtly sanctioned at many large companies, such internecine skirmishes remain rare and almost never blessed at small companies, which can scarcely afford the redundancy that in-house horse races entail.

Yet seven-year-old Spectrum is a veritable Colosseum to a host of internal contests. Events have ranged from a Christmastime sprint to steal a broken lawn reindeer to a yearlong match to wrest control of a pivotal project. Not all have been sporting. But the employees as well as the founders have profited by subscribing to the notion that a business should be run like a tournament.

From the beginning the founders have made it clear they're not in business to coddle employees. "Don't look to your company," Nugent informed those who signed on. "The company is not your parent. It's just a facility where you can come and lower your risks significantly because you have benefits, you have a base salary. But you're on your own."

Baudanza told new hires, "The risk is on you." The sober truth, he cautioned, was that "your business here could fail." Nugent and Baudanza attribute failures -- and they claim there are few -- to cultural learning disabilities. Baudanza speculates, "Maybe those guys didn't understand the environment they were in."

The managers they hired understood it very well. Or they learned to, in order to survive. The founders sought only self-starters who enjoyed a fair fight and a fat paycheck. Indeed, the first managers -- the five who run service groups and one who heads product development (there are no titles at the company) -- have been as driven as the founders have. Their compensation assures it. Early recruits wooed with promises of limitless income potential and autonomy were dispatched to "grow their own businesses" within Spectrum.

And it has worked. In 1988 Spectrum reported gross revenues of $404,000. In 1989, with nine employees on the payroll, that number more than doubled to $953,000. Revenues nearly doubled again to $1.8 million in 1990 and again to $3.6 million in 1991.

It was dramatic growth for what some in the industry consider a scavenger: a service bureau selling customized software and consulting to small manufacturing operations. "It's not glamorous work. But we knew customers needed the service and support," Nugent says. "And we didn't have the capital to pour into a product. So we decided to sell programming by the hour, not by the box." Only after the founders' appetite for growth compelled them, and the cash flow from services enabled them, did they make their first foray into the product business, with the acquisition of a product line that pushed revenues to $13.9 million in 1992.

Up to that point, they'd sold only their time. And everything they'd done was billable. New hires, for the most part programmers, were expected to cover their costs and hit a revenue quota. In return, there was the promise of cash bounties, which are still offered today. When individuals attain their quotas, they pocket 10% of their billings as a bonus. Programmers make an average of $10,000 above industry norms -- in some cases more than $100,000 a year after bonuses. Service managers earn a 5% override on the revenues rung up by those working beneath them. "If sweeping floors in McDonald's would net us $150 an hour, we would have had people out doing that," says Saint.

"If we see a piece of business, we just chase it," says employee-cum-competitor Crean, the founding father of the company's largest group and self-proclaimed top dog. "We don't wait to see if it's OK or look for permission from John and Tony. We just do it." For his take-no-prisoners approach to pursuing new business, Crean, a self-propelled Spectrum commando, has risen through the ranks to build a small army of his own and pull down a six-figure income.

Indeed, virtually every department shoulders responsibility for its own economic survival. Sales pays marketing for leads. Customer support collects commissions on services sold and then pays programmers to fix bugs or add features. Even development, a classical expense center in most software companies, shares in the spoils typically reserved for the sales jocks. By persuading customers to pay in advance for services they've planned to standardize into products, they've managed to finance their own efforts and collect a kickback. "We've all had to make it rain," recalls Donna Vaillancourt, who heads product development for the company.

The notion that managers run their own businesses took root. Says manager Saint of the group he and Crean now jointly run, "We practically kill ourselves to grow our business units. It's like the rivalry between Ford and GM, or Coke and Pepsi," he explains. "Whoever comes up with the best proposal and the best quote wins. So you're forced to be more responsive to customers. To work until midnight to make sure you don't lose the account or your incentive to somebody across the hall."

Some employees have resented the license to poach. Vaillancourt, for one, found it a rude awakening when she joined Spectrum, in 1990. "This whole competitive thing was culture shock for me. Yeah, I realized that if I wanted to build products -- that was my whole background -- I was going to have to go out and fund it with service dollars up front." She hadn't reckoned that her own colleagues would block her shots. "It took me a while to realize that meant convincing the salespeople to flip work to me instead of to someone else in the organization -- that it meant bidding more aggressively to get a job. I had to get in the game."

"It's not like a free-for-all," Baudanza counters. "Yeah, people compete, but they do it in groups. One individual is not out there trying to do in another person. It's groups vying to be the best." Still, the founders have sanctioned overlap to sustain the game. "One day it dawned on us, 'Hey, this is kind of neat," Nugent says. "We get to shop around."

Reluctant to regulate, the owners referee squabbles -- such as a bidding war over a prospective employee -- as they arise, but they refuse to formalize boundaries or set rules. "Behind it all there's a very healthy thing going on, which is a struggle to do what's right for the customer," argues Nugent. "When I talk to customers, I can say, 'Look, what do you want? The best quality, the best price, and the best delivery. It so happens that we're organized in such a way that we can guarantee all that."

Nor have the managers been oppressed by mission statements or other corporate manifestos. "We decide for ourselves where we think the market is and what will satisfy customers," says Crean. "Then we pretty much just go out and do it."

While the groups exercise the freedom to pursue markets, customers, and revenues as they please, they hold no decision-making power over allocation of funds. Spending has remained under Nugent's vigilant scrutiny. There are no budgets. There is no chief financial officer. Every invoice, every expense crosses Nugent's desk. And his attention to them borders on the fanatical.

"The company is very, very tightly controlled. That's one of the reasons we're successful," he says. "As long as we keep the finances centralized and everything else decentralized, it's a very easy business to manage."

"It's great for John and Tony," Saint agrees. "The business is running itself."

What owner doesn't dream this dream? To devise a strategy that lets you run the business at 120 miles per hour, hit cruise control, and check the cash occasionally to ensure you're taking enough out. Nugent and Baudanza come close to living that fantasy. Autonomous, ambitious managers keep the engines revved, and the founders, having risked only a few hundred dollars, retain more than 90% of the equity. Now courted by venture capitalists, they flirt with the notion of taking the company public. They draw salaries too sizable to discuss. And Nugent enjoys the luxury of a three-day workweek: he spends Monday and Tuesday at home with his kids. And if the competition exacts a toll, the owners haven't been paying it. Their dollar-driven managers have. "If you put four, five, six type-A personalities with an entrepreneurial bent into the same tank, they can end up killing each other," notes group manager Mike Guay. Saint states, though, "We're not out to kill each other. There's a huge market out there. There's enough for all of us."

Yet Saint and Crean are the first to admit that their conduct is not always sportsmanlike. Those two struck an alliance only when their rivalry threatened to put them both out of business. "We were going after the same customers with the same product," says Crean, who'd worked with Saint at another company before they joined Spectrum. Before their truce and the merger that followed, the two couldn't help harboring animosity toward one another. "We didn't like each other anymore. Absolutely. I mean, can you like someone who's taking business away from you? It's money out of your pocket."

Vaillancourt, on this rare point, agrees with Crean. "When I compete, I have to instinctively not like who I'm competing against," she says. "That's the hard part. The people I compete against are also my colleagues, my friends. I kind of feel bad sometimes."

Having lost more than a few conflagrations, Vaillancourt has grown more aggressive, she confesses. And more philosophical. "You're going to win some. You're going to lose some," she muses. "Some you don't talk about."

By mid-1991, after a good four years in the service business, a highly profitable Spectrum Associates had accumulated enough retained earnings and working capital to commit to developing a product. As service consultants, the company had built a library of solutions to common problems. The owners wagered that with some solid coding and a spiffed-up user interface, Spectrum could convert those programs into a series of products.

The decision to enter the costly and often brutal game of selling a software product was neither sudden nor ill-considered. Nugent and Baudanza knew a low ceiling would hang over the company's growth as long as it lingered in the service business. "We were selling one solution to one customer at a time, making money the hard way," says Nugent. If Spectrum came up with a product and managed to control the prohibitive costs of selling and distributing it, then at last the founders would escape the tyranny of selling time.

But for Spectrum, which had passed its young life as a software-services company, entering the product game as an unseeded player was fraught with risk. A passel of hot growth companies had the company well beat in the race to market. And larger players with hundreds of man-years and millions of dollars to invest were already coding complex software products for the very same emerging market Spectrum was eyeing: manufacturers ready for the next generation of software to run their operations.

Spectrum's strategy was straightforward: develop a series of products under the brand name Point.Man, aim for the middle of the market, and begin with an established base of customers. To broaden that base and quickly create a stable of products, it made acquisitions -- one in early 1992 and the second in the summer of 1993. Sales of the two acquired products (one ancient, the other in midlife) would establish a captive audience for Point.Man. If Spectrum then wooed customers who were already running the legacy products with steep discounts on Point.Man, the founders reasoned, they'd secure a toehold in the market, and the race to dominate it could commence in earnest.

But developing a product was more than the founders had bargained for. "Early on, we had horrendous results," recalls Nugent, who'd coded elbow to elbow with development manager Vaillancourt at another software manufacturer before starting Spectrum. In an uncharacteristically conventional move, Nugent and Baudanza had dispensed with the inside auction and simply assigned the Point.Man project to Vaillancourt, who'd come to the company with both the mandate and the track record to build a product. The job was Vaillancourt's, and more important (at least to her), the spoils would be hers as well. Nugent had guaranteed she'd get royalties.

The service managers would settle for the crumbs that fell from the table. They could slug it out in the aftermarket, customizing and integrating Point.Man at customer sites.

But according to Nugent, that orderly new approach bombed. "We never made a deadline and couldn't make delivery dates," he says. Schedules slipped. Costs ran higher than anticipated. Vaillancourt admits that. "Anyone who's ever developed a product knows that it always costs more and takes longer than you hope," she argues. At Spectrum there had been no precedent for the investment Point.Man would require. "Around here, you're either expense or you're revenue," Vaillancourt says. "It's that simple."

Unwilling to bear the overruns, Nugent reverted to his tried-and-true method of stoking a little competition. "We made a decision organizationally to split the project."

As Vaillancourt tells it, there was no discussion. "The decision was presented to me as a fait accompli," she says. "I was kind of blindsided by it."

Nugent told both Vaillancourt, from whom he had ordered the $50,000-Cadillac version of Point.Man, and Crean, her arch rival, to regard their employer as a customer. It was open season. Crean pledged he could develop the next product in the Point.Man series faster and cheaper than Vaillancourt's group could. "If Point.Man sold, I knew we'd be a product company. And the amount of revenue we'd get from services would be insignificant by comparison." In the Spectrum tradition, he bid aggressively for the work. "I wanted to share in more of the company's growth." The royalties Nugent had promised would make the winner's purse quite lucrative: 10% on every dollar of net revenues.

Vaillancourt's suspicions as well as her experience from previous losses told her that Crean's penchant for bargain pricing had swayed the frugal Nugent. "I knew this didn't happen without Finbar's maneuvering." She speculates, "He went behind my back to cut a deal with John. He probably lowballed me to get his people trained and ready to grab some share" of the aftermarket, where nearly everyone is betting on a bonanza of service business.

"I felt betrayed," she says. "But it's part of the Spectrum culture to promote competition. So I accepted it." Why didn't she leave? "I'm extremely well-compensated here," she says. And the royalties at stake, even if she claims only part, instead of all, exert a magnetic power.

Despite the danger of inducing profound schizophrenia into the company's development efforts and the discord wrought by the decision, Nugent contends "it was in the best interest of the company." Finally, Point.Man product development is running on schedule. "It's not uncollegial. It's good for the company," insists Crean.

But good for which company? Spectrum the service company? Or Spectrum the product company? Or the founders' hybrid of both?

The rounds to come, which promise to be even more hotly contested, may settle the question. If nothing else, they will determine who wins the grand prize (and who comes up empty) in the Point.Man sweepstakes.

The sharks are circling for the Point.Man kill," says Guay, who, like Vaillancourt, Crean, and Saint, hopes for a mother lode of service business once the product is released. The aftermarket will inevitably be the venue of round two, since most of the managers -- those who developed Point.Man, those who didn't, and even the two who would undermine it with a competing product of their own -- plan to be in the ring selling services.

Given the in-house competition, Guay cannily hedges his bets. "I think there'll be Point.Man business, but I'm not betting the farm on it." He plans to ally himself with outside competitors marketing rival products as well. "Hey, why not? I'm here to grow my own business. If I can't get a nickel off Point.Man, then I've got to find other markets."

It is not corporate treason, Guay asserts. He and his peers are bound by their paychecks to defend and uphold their own business units. "That's what we're paid to worry about," he notes. While stock-option plans have been discussed, equity stays in the founders' hands. And there is little to reward any broader fealty to Spectrum as a whole. So like feudal lords, the managers confine their attentions to furthering their individual fortunes. "I've got undying loyalty, all right," says Guay. "To my checkbook.

"If it comes to survival of the fittest," he warns, "it's going to get down and dirty. The people at our level didn't get here because we're meek and mild. It'll get dirty. I'm not going to sit here and watch my business evaporate.

"Either it gets really cutthroat or we make a truce," he predicts.

But Nugent and Baudanza have already rejected one proposal to recast the business as a partnership. Last fall, uneasy about the prospect of a civil war over Point.Man and apprehensive that growth would invite expansion teams to their league, the managers proposed a cease-fire. "They wanted to make this place into a law firm," protested Nugent.

"That's when I said to Tony, 'This is communism. We're taking capitalism and we're turning it into communism. That's insane." They dissuaded the partnership faction by planting seeds of doubt. "You're going to rely on these other people for your business?" Baudanza challenged. "Why would you put your earning potential in their hands?" The managers were moved, and their motion was tabled. "I feel we live on the edge right now," says Crean. "Maybe we ought to stay there."

John Nugent and Tony Baudanza have made their business a shrine to that "edge." Bootstrapped in a condemned building, Spectrum has managed not only to survive but to thrive by embracing internal contention. But even the most cunning strategy has a limited warranty. Conditions change. What once enabled you to cope may later cause you to crash.

When Nugent and Baudanza founded Spectrum Associates, in 1986, they set out simply to serve the "plumbing needs" of customers. They engaged in a game of survival, and the urgency of their struggle left no time for rules or penalties. As long as satisfied customers kept paying, "we just figured it would cut the risk to try and run a bunch of different businesses inside the company. All at once."

So the bickering in-house will be tolerated, Nugent says, "as long as we don't compete with ourselves in front of the customer. It's like fighting in front of the children." Yet already one key customer has been treated to ringside seats in a bout that ensued when an eager product salesperson nearly scuttled a large service contract.

Tension between the product and services divisions of the company is natural. "It's a cultural battle," says Nugent. But the product and even the company could become casualties. While it is an industry maxim that companies cannibalize themselves, most wait until after a product has been released. Unlike the products it will compete against, Point.Man faces double jeopardy.

Can the internal rivalry be sustained without imperiling the product? Even Crean, ordinarily an enthusiastic gamesman, is wary. "This is not going to be good. Because every Point.Man services job that comes in, we'll be quoting lower and lower to beat each other." But he'll continue to play both ends, positioning himself to profit by Point.Man's success even as he pits his rival product against it.

In the meantime, Nugent and Baudanza, aware of the game's odds, remain adamant in their distrust of more traditional business models. "It worries me," says Nugent, "but if you follow the conventional wisdom, you end up in the same place as everybody else."


($ in thousands) 1987 1988 1989 1990 1991 1992 1993

Sales $71 $404 $953 $1,837 $3,645 $13,904 $25,200**

Margin* 18% 18% 19% 27% 27% 26% 24%

Employees 2 5 9 16 26 65 161

Assets $20 $87 $72 $132 $150 $4,500 $9,000

*Operating margin before executive compensation, research and development investments, and taxes **Estimate