Everyone knows the story. On March 11, 2011, a magnitude-9.0 earthquake off Japan's east coast triggered a massive tsunami and led to a cascading series of problems. It wiped away whole towns, killing over 15,000 people and leaving almost 3,000 more missing and presumed dead. Nuclear reactors melted down, disrupting the national grid, reversing energy policies as far away as Germany, and contaminating a part of Japan for generations to come.
The disaster caused factories to close, not only in Japan, but in the United States and Europe as well. Some automobile factories remained shut or at reduced capacity for months; you can't sell a car or a computer unless it has all its parts, and the disaster had washed away the source of necessary parts. Before companies could fully recover, massive flooding in Thailand caused further disruptions in Asian supply chains. Total losses from those two disasters were very likely greater than the savings companies had won, over the preceding years, by squeezing out costs and making their supply chains so efficient.
You have to expect the unexpected, yet in retrospect, too many companies and leaders have inadvertently made themselves fragile in their pursuit of ever-greater efficiency. Then when a shock inevitably comes, they are unable to absorb it.
Too much efficiency can be just as deadly as too little, if it leaves an organization unable to cope with change — either because it's too fragile to survive a crisis or too rigid to adapt to industry changes. So the goal should not be greater efficiency, but rather efficiency where it makes sense.
This was not a new lesson, at least to the armed forces. As one Army general colorfully put it, "'Just in time' means damn near too late." In combat, the fact that someone is actively trying to disrupt your operation changes the calculation. It's rarely a good idea to pass everything through a single chokepoint or to store all the ammunition in one place.
And it's not just about being prepared for 100-year events. An over-focus on efficiency can also make it more difficult to cope with gradual change.Eliminating errors by finding an optimal way of solving a problem, and then ruthlessly adhering to it, is at the root of quality control and improved productivity.
However, in complex systems, where most decisions are made locally instead of centrally, and where outside forces affect things in often-unexpected ways, both problems and solutions have a disconcerting tendency to evolve. An organization where everyone does the same things in the same way — an efficient organization, in other words — is often ill-equipped to adapt to a changing world. So, as Pixar cofounder and president Ed Catmull advocates, leaders interested in creative outcomes should focus on fixing problems rather than on eliminating errors.
Eliminating all errors makes it hard to compete in the trial-and-error process that's required for a company to adjust, because there are no trials without errors. Clayton Christensen has nicely described how the very things that made a company successful, including its efficiency, can also cause it to become obsolete when it's unable to adapt to disruptive change.
Here the Army hasn't done a good job. Only after more than two years of combat and death did units begin adjusting their tactics in Iraq and Afghanistan, but adjust they did. Yet after more than a decade at war, there have been only minor changes in the systems that govern training and human resources, which still focus on efficiency at the expense of experimentation and innovation. As Peter Drucker once said, "There is surely nothing quite so useless as doing with great efficiency that which should not be done at all."
A lot of businesses have done better, and not just obviously creative businesses like Pixar, Amazon, IDEO, and Google. A number of leading firms, including Cisco, General Electric, GAP, General Mills, AT&T, and Procter and Gamble have adopted approaches that are "inefficient" in important ways. They create deliberate experiments and focus on learning and iterating rapidly. The idea is to get it good enough to begin, and then improve rapidly, rather than trying to get everything right from the beginning. This approach has helped them adapt and succeed in turbulent times.
Companies have begun to change their thinking about resilience, accepting the need for a slight reduction in efficiency as a necessary form of insurance that can help them weather a crisis. Now they need to change how they think about adaptability in this rapidly changing world: identifying how much error is acceptable within each area, not treating them all the same. The quest for perfection that makes sense on a factory floor or in a distribution schedule makes no sense in developing executives or exploring new markets, and it's a mistake to think of talent and supply situations in the same way.
A healthier approach also means managing acceptable losses. That's different — and, ironically, much safer — than trying to prevent failures altogether, by over-relying on past data to try "managing" risks that are inherently unknowable. We need to see those losses not as failures but as investments in the future. To succeed, we have to make it cool to fail in the right places, as long as it's recoverable, and as long as we learn from the failures and adjust.