Performance Improvement
H. JAMES HARRINGTON

The modern quality system is based on the concept that it is better to prevent errors than correct them. Even the ISO 9000 standard requires an organization to have both a corrective and preventive action process. Quality gurus advise us to establish systems that will prevent errors from occurring and, when errors do occur, correct them and prevent them from recurring.

I remember an all-hands meeting I held with my IBM quality team back in 1981. At this meeting, I asked the team to work with me to put to bed our quality problems. One of my inspectors in the third row, Shirley, raised her hand and said, "Jim, we don't want to put the problems to bed. What we want to do is bury them. If you put them to bed, they can get up and bite you again."

Shirley was absolutely right. We need to do everything economically justifiable to keep problems from recurring. But most organizations face so many problems that, once they have slain one quality dragon, two more are breathing fire down their necks.

It seems that things are getting worse. Our processes are changing at an ever-increasing rate, downsizing has reduced the number of people who can fight these dragons, and there seem to be more dragons to fight than ever. We would all like to be the knight in shining armor who slays the fierce dragon and saves the organization from disaster.

However, this is the problem. The real job begins when the cheers of the appreciative crowd die down and the knight must turn over rocks to find the dragon's eggs and destroy them. The truth is, most organizations can't find the dragon's eggs because the knight has ridden off to fight another dragon.

Why does management allow this to happen? Obviously it would be better to prevent problems than to correct them. The answer is simple: Top management made it to the top because they're good firefighters. They're take-charge people who make things happen and resolve problems. Their armor is stained with the blood of many conquests. They give the laurel wreath to the knight who slew the last dragon.

At IBM, we had a saying: "If you want to get ahead, create a problem that you know you can solve." We held weekly progress meetings to review the week's accomplishments and our preparations for the following week. At these meetings, manufacturing, engineering, quality and production control reported the status of each project. Production control was asked if they had all the parts to support next week's production. If they answered "Yes," we moved on to the engineering report. If they answered, "We have sufficient quantity for all but two of the 5,379 part numbers," the meeting's focus changed drastically.

Production control was charged with presenting a plan that would secure the necessary parts by the coming Monday. Knights were sent out to save the day. An early Monday morning meeting was scheduled to find out if production control's knights had slain their dragons. When production control reported that the parts were in receiving inspection, everyone cheered with relief and turned away to fight another dragon. The only time the production control manager received recognition was when he created a problem that he could solve. In today's environment, managers who solve problems get promoted, even if they caused the problem.

In quality assurance, we had an engineer whom I will call "Joe V." Joe was lucky because the projects assigned to him flowed through manufacturing and out to the field with very few quality problems. He very seldom objected to any design release and never had to shut down the line. Joe seemed always to be in a position to accept the transfer of a production line from product engineering to manufacturing. He was assigned to good products, and he never had dragons to fight. Other good quality engineers were out there fighting the dragons, shutting down the line, not allowing products to be announced, rejecting equipment because its Cpk was not equal to 1.5. These were the quality engineers who got promoted to management and built a great quality reputation for the quality assurance function.

What a mistake I made. After many years of watching Joe in action, I began to realize that he was the real quality professional. He didn't work with the development management team but with the designer and technician. The first preliminary design already reflected Joe's manufacturing and quality experience. The manufacturing engineers asked Joe what type of equipment they should purchase. Manufacturing management searched out Joe's advice on where to assign their people and how to train them. Joe's projects had few problems because he practiced prevention.

Individuals or organizations that truly practice prevention have little visibility in the total organization. They quietly do their jobs and create no waves. As a result, they often are overlooked. If we truly believe in prevention, we need to develop more effective ways to recognize people who prevent problems from occurring. We need to cheer the knight who has no dragons to fight and be less excited about the knights who are fighting the battles.

About the author

H. James Harrington is a principal at Ernst & Young. E-mail him at jharrington@qualitydigest.com .