Make Smarter Mistakes

III. Follow-up is as important as follow-through.

Plenty of mistakes can be buried before anyone finds out. The bug in Christine Miyachi's printing software, for example, was the result of a simple coding error. Miyachi and her coworkers were able to stop shipments of the flawed program and replace it with an earlier, bug--free version. Because the next release was only a few months away, Iris never shipped the "buggy" release. And customers didn't seem to notice the omission.

Quickly remedied, with few repercussions, these mistakes are the kind that most people fix and forget. They are also the kind most likely to need -- and reward -- thorough investigation. Simple mistakes don't always have simple causes; they can be the result of systemic problems that will recur if not corrected. And simple mistakes often involve more than one person, which means there are issues of group dynamics to confront. "If nobody sees it, does it need to be fixed?" asks Bill Rosenzweig, 38, a partner at the San Francisco-based Venture Strategy Group, which consults with emerging companies on brand marketing and organizational development. "Well, yes, because the problem might be that nobody had been looking."

Dennis Matthies agrees. "A full explanation for a simple mistake can have many levels of analysis," says Matthies, who developed a method of "self-coaching" during his tenure at Stanford University's Center for Teaching and Learning. "A thorough evaluation often reveals something about your habits and those of your colleagues, and about the work processes at your company."

The bad news: Understanding those habits and processes means developing rigorous techniques for investigating what went wrong -- techniques often much more rigorous than those required to fix the problem. It means debriefing the relevant players, analyzing their actions, and examining your own shortcomings. The good news: Little mistakes can yield big insights.

Christine Miyachi learned that lesson. After spending hours combing through old emails and retracing the moves of about a half-dozen staffers, Miyachi discovered that the minor coding slip-up wasn't an isolated mistake. It was just one byproduct of an organization-wide problem. Overworked and pressed for time, staffs in two departments were routinely taking shortcuts around their quality-assurance procedures. "None of us was going through the procedures we had designed," she says. "We all had gotten lackadaisical. We just hadn't got caught before."

As a result of her investigation, Iris developed more stringent quality controls that require each staffer to sign off before the team goes to the next level of testing. If the process uncovers any error -- even "the teeniest one" -- testing has to start over. "It adds a week to our release schedule," Miyachi says. "But our software is more reliable." It's a big change that grew out of a small mistake.

IV. Seize the moment of truth.

A luxury hotel loses track of a valued guest's reservation. An airline loses a passenger's luggage. Whenever such problems surface, there's a brief period during which front-line employees can turn a bad situation into a memorable one. If front-liners act wisely, a disgruntled customer walks away with renewed faith in the company. If they don't, the company may never get a second chance. Customer-service gurus call these encounters "moments of truth."

There are moments of truth inside the company too -- rare times when people can gather to learn from their mistakes. Consultant Hart has simple advice about when to initiate these moments of truth: the sooner, the better. "Once you've captured information for people, you have to build it into their knowledge base," he says. "If you don't do it quickly, it's hard to go back to later." Hart suggests that you sit down with the group, offer a play-by-play analysis of what went wrong, and map out a plan to avert similar errors.

"If you have a process that lets you stop and recognize what you've learned, it brings things to a deeper level of clarity," agrees Bill Rosenzweig. He began to appreciate the value of "captured learning" in 1993, when he was CEO of the Republic of Tea, a specialty beverage company. A Canadian distributor proposed selling the company's products throughout Canada. Rosenzweig got so excited that he signed on before asking tough questions, such as whether the distributor would agree to maximum sales orders.

The unintended result: the Republic of Tea supplied so much tea to this new distributor that it couldn't meet orders from long-standing customers. It was a classic business mistake. "I thought that if the relationship was strong, we could always work something out," says Rosenzweig, whose correspondence with the tea company's cofounders, Mel and Patricia Ziegler of Banana Republic, was turned into the cult classic The Republic of Tea (Doubleday Currency, 1994). "But I was naive, and it wreaked havoc on our operations."