Measuring and Rewarding at the Magic Factory

The CMO thought and thought, then he said, “First, I want our tester wizards to find more defects before the potions and spells leave the factory.”

The ball said, “OK. Find more defects. I'm sure that's possible. What else?”

“Second, I want our developer wizards to fix all of those defects before they leave the factory.”

“Before the wizards leave the factory or the defects the factory?”

“Both.”

The ball clouded over, the elevator music played and played and played, and eventually the ball said, “Measure the behaviors you want. Reward them.”

So, that's what the CMO did. When a tester wizard found a defect, the CMO paid that tester twenty coins. And, when a developer wizard fixed a defect, the CMO likewise paid that developer twenty coins. The tester wizards promptly started finding more defects, and the developer wizards promptly started fixing more defects. This was their focus. The CMO was very happy. The tester and developer wizards were very happy. Some of them bought new cars and iPods. Quality went up a bit. Customer complaints went down a bit. The CMO was delighted.

Many months passed, and one morning the CMO stood in front of his bathroom mirror, devastated. Quality was up, but production had dropped. So had sales, and therefore revenue. The tester wizards and developer wizards spent so much of their time finding and fixing defects that production was bound to drop. That was understandable, he thought. Surely, it was the cost of quality?

He put on his wizard's hat, hopped on his broom, opened the window, and took off. As he flew, he worried. He alone knew that Magic Factory, Inc. was going broke. How could this be? Quality—or, rather, the number of defects found and fixed—was up. They were the same thing, weren't they? You measure one and you get the other, right? He threw his arms up in the air, in despair, and then caught them on the way back down.