Want to improve the quality of your company's product or service? Try these measures. Stop inspecting every part of monitoring every transaction. If you have a quality-control department, get rid of it. Forget production and sales goals -- in fact, forget management by objective altogether. And oh yes: abolish employees' annual performance reviews and merit raises.

There's more, of course, but this isn't the place for the complete lesson. For that you might want to consult W. Edwards Deming -- who, it turns out, has a good deal to say not only about quality control but about how American managers run their companies. To him there isn't much difference between the two subjects.

Dr. Deming, as his fans invariably call him, has long been an icon in Japan, which for the past 36 years has given the Deming award to companies with outstanding quality records. Few American had heard of him until 1980, when an NBC producer named Clare Crawford-Mason made him the star of a prime-time special called "If Japan Can . . . Why Can't We?" Since then, dozens of U.S. companies have hired him as a consultant, have sent representatives to his legendary four-day seminars, and have ponied up $6,800 for a 15-cassette "videocourse" in which Deming explains his philosophy and techniques. At age 86, he's booked up well into 1988.

For all his newfound popularity, though, Deming's impact has been curiously limited, partly because of his own communication problems. His manner can be inscrutable, even off-putting -- "insulting more than enlightening," one would-be client described it. He has a propensity for overblown imagery, referring, for example, to "the seven deadly diseases" of management. And he speaks in the broadest of generalities. "When we were through watching the tapes, we thought, 'Great philosophy . . . but what do we do?" a small-company chief executive officer who sampled the videocourse told INC. a couple of years ago ("Quality Begins at Home" August 1985).

Finally, Deming has acquired what every great-but-obscure thinker needs: popularizers (see box below). And if there's one theme to emerge from the recent spate of books and videotapes about the man, it's that thinking of Deming as a quality expert is like thinking of T. Boone Pickens as an oilman. It's correct, but it misses the big picture.

To Deming, high quality can't be inspected in, added on, coerced, demanded, or achieved through exhortation ("Quality is everyone's responsibility!"). Rather, high-quality goods and services are the natural products of an organization that's working right. To oversimplify only a little, "working right" means identifying problems, then encouraging everyone involved -- workers, managers, suppliers, whomever -- to help solve them.

If that sounds trite, think for a moment of all the annoyances, glitches, inefficiencies, and foul-ups you put up with because things have always been that way. Better yet, ask your employees what they put up with. At Janbridge Inc., a Philadelphia printed-circuit-board manufacturer, one group of workers never knew why some of their boards' holes kept turning up defective; they just kept sending them back to rework. Another group was expected to build boards to military specifications. But since they didn't understand the specs, they usually ignored them.

Eventually, Janbridge made a Deming-inspired commitment to attack these problems. The solution to the defective holes turned out to be classic Deming: an employee team used statistical measurements to identify the precise trouble spot, then figured out how to eliminate it. As for the military manual, Janbridge assembled a group to rewrite the specs in understandable language. Simple enough -- but nobody had ever done it before.

Seeing management through the Deming lens leads to some surprising conclusions. You'd think, for example, that a quality expert would favor 100% inspection. In fact, Deming would argue, it often makes no more sense than a shopper's opening every package in the supermarket before buying. If the production system is good enough, mass inspection is likely to be a waste of time. Similarly, to rely on a quality-control department is to tell workers, indirectly, that high quality isn't their job. Quality has to be built in, not inspected in.

As for production quotas, sales quotas, annual performance reviews, and merit raises, they're all based on the notion that variation among employees is meaningful. Most of the time, says Deming, it's not; the system determines the output, and employees' efforts typically fall within a normal distribution curve. Yet if your evaluations are fair, you will determine that half your workers (by definition) are below average, and you will tell them so. Result: half the work force is instantly discouraged and demoralized, and any sense of common purpose is undermined. Exceptional performers should be rewarded, admittedly. But "exceptional" means just that. It doesn't mean "above average."

To judge from the popularizers, Deming offers not so much a pat formula for improvement as a new way of approaching a company's problems. That, of course, is all he claims to have offered the Japanese when he began counseling their leading industrialists back in 1950. Maybe we should begin paying attention the way they did.