BACK IN THE LATE 1960s, SHANNON & Luchs Co. was just one of a dozen or so small real estate brokerage businesses in Washington, D.C. Its managers were all paid in accordance with the norms of the industry, and they received the standard merit raises and bonuses at the end of each year. Then, around 1970, the company overhauled its executive compensation system. In addition to their regular salaries, division heads were given the opportunity to earn a percentage (10% to 25%) of the net profits of their respective divisions, adjusted for overhead and other expenses. The result: sales and profits took off. Today, Shannon & Luchs is one of the largest and most profitable real estate companies in the United States. Company president Foster Shannon gives full credit to the compensation system.

Such tales may sound too good to be true, but they are becoming increasingly common as more and more companies turn to incentive pay as a means of achieving strategic objectives. The trend is easily the hottest one ot hit the compensation field since the cost-of-living raise. It involves a whole different approach to compensation, one that is geared toward achieving future objectives, rather than rewarding past performance. To date, thousands of businesses have adopted such systems, and those that try it swear by it. Most practitioners will tell you that -- in addition to fostering phenomenal results -- incentive compensation allows them to recognize the movers and shakers in their organizations, the people who make things happen, and to inject a new sense of vitality and purpose into the company as a whole.

Testimonials aside, the trend reflects important changes in the business environment. As inflation has declined, companies have found it harder to justify the big raises that were common in the 1970s and early '80s, and so they have begun searching for new ways to keep employees motivated. Even more important has been the pressure of increased competition, forcing companies to become ever more efficient and profitable.

Among the first to move in the direction of incentive compensation were the Fortune 500 companies. A study by Hewitt Associates, in Lincolnshire, Ill., shows that more than 90% of the nation's largest companies had short-term incentive plans as early as 1980. These plans made it possible for participating managers to earn bonuses totaling 16% to 55% of their base salaries, given the achievement of certain operating or financial targets. Since then, thousands of smaller businesses have set up incentive plans of their own.

On the surface, at least, creating an incentive-pay program doesn't appear to be difficult at all -- provided you understand where your company is, and where you want it to be. You have to know, or instance, what you're shooting for, whether it's more profitability, higher sales, better service. As a wise man said, if you don't know where you're going, the odds are you'll wind up somewhere else.

Once you are clear about your objectives, however, the rest falls into place. First, you have to decide who to include in the plan. If you want to increase profitability, for example, and if your business is composed of relatively autonomous operating units or product areas, you may well decide to focus on a handful of key managers -- the ones with the leverage to make sure their respective units make money. On the other hand, you may have a company like Riley Gear Inc., in North Tonawanda, N.Y., a $6-million manufacturer of precision gear systems, whose success depends on its ability to deliver quality products on time at competitive prices. Since every employee plays a role in achieving the company's productivity goals, all 90 of them receive a quarterly bonus check when targets are met.

Of course, you also have to choose the performance criteria by which you'll hold people accountable. Here, your decision is almost entirely a function of your goals. Indeed, two identical companies might deliberately choose different performance criteria. One, for example, might decide to reward nothing but sales growth as a way to spur aggressive selling, while the other might target profits or quality control. The latter business would, in effect, be telling people to say no to some business opportunities. But each company, in its own way, would be encouraging the kind of behavior it wanted.

Then again, some companies might want their employees to pay attention to more than one goal at a time. For several years, Nicolet Instrument Corp., a Madison, Wis., manufacturer of medical and chemical testing equipment, calculated its management bonuses using a formula that took into account both sales growth and return on assets. With two important goals to balance, says chief executive officer and president John Krauss, there were no rewards for leaning too far in one direction. Other companies accomplish the same thing by establishing separate incentive pools tied to the achievement of different objectives.

Whatever measures you choose, they must be readily comprehensible to the employees they affect. If employees don't know what kind of performance you are trying to encourage -- or why it's important -- they aren't likely to respond as you intend. You either have to explain what you are trying to accomplish, or choose other measures. By the same token, the performance criteria must involve aspects of the business that the affected employees can control and monitor. That means providing them with the data -- monthly sales figures, production reports, profit statements -- that will show them how they are doing.

Within these parameters, however, you have a lot of flexibility, not to mention influence. You can, indeed, attach incentives to almost anything -- and thereby determine how managers and employees spend their time and where they focus their attention.

Now, none of this seems particularly difficult. You choose your goals, your people, your performance criteria. What could be simpler? Well, not so fast. The problem is that, at every stage, you have decisions and judgments to make, and any one of them can undermine your plan.

Consider, for example, the choice of performance criteria. Should you establish custom-made targets for individuals, or is it better to tie their incentives to the performance of the company as a whole? Dynamark Security Centers Inc., a $5.5-million franchisor of home security centers and distributor of security equipment, gives each of its nine key managers and supervisors a different set of performance targets. The marketing and training manager, for instance, gets a small slice (about 1%) of sales up to a given level, and a smaller percentage after that. The head of franchise development, on the other hand, gets a cut of the franchise fees and the inventory ordered by new franchisees. "The structure of the deals is basically the same," says chairman Ed Cusick, "but everyone gets his own report card."

Some businesses go even further, linking an individual's reward to the accomplishment of specific tasks. Thus, at Morris Decision Systems Inc., a computer dealership and maintenance company in New York City, the vice-president of finance has to get a new general-ledger system up and running in order to earn a part of his bonus for 1986.

But there can be problems with this approach. First, it takes time and effort to select the right goals. Then there is the administrative burden of monitoring the performance of many individuals. But perhaps most worrisome is the possibility that what's good for a particular individual, or group of individuals, may be awful for the business as a whole.

In the early '70s, for instance, Nypro Inc., now a $65-million plastic injection molding company in Clinton, Mass., began to reward employees for their own individual output. Some enterprising workers found ways to speed up production equipment during their shifts. They refused to share their secrets with their colleagues, however, and the high-speed work undermined quality. So Nypro was forced to switched from individual to group incentives.

Fearing similar problems, many companies require a certain level of overall results before individual bonuses are paid. "You can say, if we earn so many dollars, or if we get into the World Series, you'll get a reward," notes Peter T. Chingos, who heads the compensation consulting practice at Peat, Marwick, Mitchell & Co. But finding the right balance is not easy.

Nor is it easy to establish performance standards for every job. True, you can set quotas for salespeople and determine efficiency ratios for plant managers. You can even measure performance in such areas as quality control: at Soft-Switch Inc., a King of Prussia, Pa., software company, the quality-control manager is rewarded in part on the basis of results from customer-satisfaction surveys. But what do you do with a human-resources manager? Should you measure employee turnover? In many cases, turnover is totally beyond a manager's control. What's more, if you do target turnover, you run the risk of winding up with unambitious employees whose main virtue is that they don't like to change jobs.

To avoid these sorts of decisions, many CEOs prefer to maintain a certain amount of discretion over bonuses. In rewarding vice-presidents and project managers, Joseph Viar takes into account the "degree of difficulty" of the projects they manage. He could pay strictly on the basis of volume of business under management, "but different jobs rely on different mixes of inside people, consultants, and subcontractors," says Viar, president of Viar & Co., an Alexandria, Va.-based consulting company in the data-processing area. Thus they require different amounts of management, and he compensates accordingly.

Then again, you can't use too much discretion in awarding bonuses without undermining your incentive program. If the principal basis for compensation is the boss's whim, the only real incentive is to stay on his good side.

At this point, you still have to decide how much money you should dish out in the form of incentives. It can't be so much as to imperil the business -- by getting in the way of meeting debt service payments, for example -- yet it has to be enough to attract employees' attention. As a rule of thumb, most compensation experts advise that you make available incentive bonuses of at least 10% to 15% over base salaries. Employees will tend to regard smaller bonuses as "tips," which may motivate them to work a little harder and "smarter," but not enough to justify the effort and expense of establishing an elaborate incentive system.

Then there's the related issue of selecting the right performance levels -- a critical part of the process. If the targets are too high, people may give up. If they're too low, you may encourage people to take it easy. What happens, for example, if you surpass the target midway through the year?

And what if you set target levels that inadvertently wind up penalizing your best employees? That's more or less what happened at The Myers Group Inc., a freight forwarder with 65 offices around the country. For several years, the company paid out bonuses according to a formula that rewarded people annually for profit improvements at their individual branches. The formula was designed to motivate those who worked at the least efficient locations, and that it did. But it provided little incentive for employees assigned to the most profitable branches. Moreover, the system became less and less effective over time. The better an office did one year, the harder it was to receive a bonus the next. People grumbled, and so the company, based in Rouses Point, N.Y., eventually scrapped the formula. Now incentives are tied to the overall profitability of each office and of the company.

Once you have settled on performance levels and criteria, you still have to decide how often people will be rewarded -- an aspect of incentive compensation that is often overlooked. After all, the real test of any incentive program is its ability to keep people focused on company objectives. Annual bonuses are traditional, and relatively easy to administer, but can employees stay focused on targets for a whole year? Gordon Lankton of Nypro, the plastic molding company, doesn't think so. His company pays its productivity bonuses on a quarterly basis because "a year can feel like a long time," he says. To make sure that everyone notices, Nypro even uses special profit-sharing checks with a picture of George Washington in the center and "profit-sharing" printed across the top.

On the other hand, quarterly bonuses can be extremely impractical from a company's perspective. Not only does it take administrative effort, but it demands an ability to forecast with precision and to anticipate cash-flow needs. Recently, an air-freight company paid out substantial incentive bonuses at the end of one quarter, only to hit a dry period the next. It hastily revamped its quarterly incentive program. Now nonmanagers get bonus checks after each profitable quarter, but managers don't receive theirs until annual results are in.

So, if you look hard enough, there are solutions to all these potential problems. The bad news is that, once you've come up with a viable short-term incentive plan, you still have to confront the issue of long-term incentives -- the kinds of rewards that ensure employees stay focused on a company's objectives over the long haul. Those kinds of incentives can be just as important as the quarterly and annual ones, maybe more so, and the issues involved are no less thorny. Should you give people real stock, or stock options, or some sort of substitute, such as "phantom equity"? In a private company, how much information should you reveal? How should the value be determined? Who should you include in the plan? How often should you make awards, and at what level? Should you pay dividends? How can people cash out? The list goes on and on. In effect, you have to start all over again, deciding what kind of behavior you want to encourage, and why.

And, as they say on late-night television, THAT'S NOT ALL! You also have to be prepared to change your plan (or plans) at least every couple of years. Why? Because companies change, markets change, people change, objectives change. Even the best plans aren't good forever. Some need to be rejiggered every year -- adjusting the performance criteria, including other people, and so on. From time to time, moreover, you may have to scrap the whole system and start again.

Consider Nicolet Instruments, which recently has been forced to restructure its program in response to a slowdown in its market. The original system rewarded managers according to the performance of individual product groups. It worked fine, says CEO Krauss, when the company was smaller, and growing at 25% to 30% a year. But now the growth has leveled off, and the old rules don't apply.

Incentive compensation takes an enormous amount of time and effort. It also requires that you think strategically about your business, that you provide significant rewards for performance, and that you be willing to share a lot of information with your employees. The systems that work best are the ones with clear objectives that people can understand and clear incentives that they can follow. If you can't provide those things, or don't want to, you might as well save yourself the trouble. Incentive compensation is not for you.

There's only one problem with that attitude. The evidence is overwhelming that a well-designed incentive system can have a major impact on a company's performance, giving it a new competitive edge. So if you don't set one up, you run the risk that your competitors will.

In fact, it could be that the company passing you on the right already has one.