Henry Petroski is a professor of history at Duke University with a dark specialty: engineering failures. His first book, To Engineer is Human, dissects the anatomy of several disasters, from the Tacoma Narrows Bridge to the walkways at the Kansas City Hyatt Regency Hotel. If you’re squeamish about flying or driving over bridges, this book won’t make you feel better. But it’s a captivating window into how engineers think, valuable for anyone in business.
Engineers are natural skeptics. They treat each new engineering project as a hypothesis to be disproven. By imagining a structure under every conceivable situation, engineers are forced to think in the negative. How could this building collapse? How could this bridge fail? What could go wrong? Even a structure as rigid as the Brooklyn Bridge should be treated as an accident waiting to happen. That it has stood for over one hundred years is no guarantee that it will stand tomorrow, despite its structural soundness today.
To think like an engineer is to think critically about success and failure. If you’re seeking business wisdom but you’re sick of the business aisle, Petroski’s books are a good place to look. I’ve curated five insights from To Engineering is Human, Success Through Failure, and his latest, To Forgive Design.
1) Success Does Not Imply Soundness
The I-35W Bridge in Minneapolis stood for more than 30 years before it collapsed in 2007, killing 13 people. An investigation discovered that the steel gusset plates, which secure the trusses, were undersized. To make matters worse, two inches of concrete were added over the years, increasing the load weight by 20 percent. Even though engineers examined the gusset plates during routine inspections, nothing was visibly wrong.
Lesson: Assume weaknesses exist, even if you can’t see them. Find them before they hurt you.
2) Be Aware of “Organizational Drift”
In 1855, engineers completed the Niagara Falls Suspension Bridge, the world’s first working railway suspension bridge. The bridge was a success, but it led to an embarrassing failure nearly 80 years later. Suspension bridges became popular, and every new bridge introduced a small, seemingly innocuous change, that sacrificed safe engineering practices for slender, narrower and more stylish looking decks.
The Tacoma Narrows Bridge, completed in 1940, was especially slim and flexible. At the time, it was the third longest suspension bridge in the world, and it was designed to withstand 100 mph winds. However, it could not hold morning traffic in a forty-two mph crosswind one November morning. All suspension bridges move in the wind, but the Tacoma Narrows Bridge was too light, and it collapsed four months after it was completed.
When a design works well, it’s natural to adopt it elsewhere, again and again, each time with a change. The small incremental changes add up, and all of a sudden they result in disaster.
Lesson:Beware of “Organizational Drift,” the tendency for companies to move away from their original focus too slowly for anyone to notice.
3) Good Engineers Crave Counterexamples
The psychologist Gary Klein talks about the pre-mortem. Before you start a project you should imagine the following scenario: “It’s a year later, we’ve done the project, and it’s been a massive failure.” We’re more likely, as Robert Sutton explained to me in an interview, to imagine a more detailed and accurate future when we’ve considered worst-case scenarios from a future perspective.
Lesson: Before you start a project, conduct a pre-mortem. Imagining failure is vital.
4) The Innovator’s Dilemma
In 2009, a thirteen-story apartment building in Shanghai collapsed, nearly intact. Oddly, the design and construction of the building had nothing to do with its demise. After the building was completed, workers excavated a deep hole on one side to make room for an underground parking lot. They trucked the excavated dirt around the building and piled it on the ground, creating a thirty-five foot heap. Heavy rains saturated the dirt, putting unexpected lateral pressure on the building’s foundation, which began to shift. Eventually, this asymmetrical strain caused the building to fall on its side like a domino. One worker was killed.
After a highly visible engineering failure, our knee-jerk reaction is to blame the design or the designer. The disaster in Shanghai had nothing to do with either. “Had its designers known that the piles would be subject to sideways pressure, they would have made them larger and thus more resistant.” Sound business practices fail for the same reason: unforeseeable, external pressure. In a passage that could come from Clayton Christensen’s The Innovator’s Dilemma, Petroski writes that, “Even if a building is well designed structurally, it can still succumb to failure through no fault of its own.”
Lesson: Expect unforeseeable external events to undermine sound business plans.
5) Failure is Good (In the Long Run) Because it Reveals Latent Errors
In a lecture at Case Western Reserve, Petroski imagines what would have happened if Titanic did not crash into an iceberg. Titanic could have safely crossed the Atlantic thousands of times without sinking. If it had, engineers would have concluded that the design was sound, and built an even bigger ship. The latent errors responsible for killing 1,500 people on Titanic—too few lifeboats and thin bulkheads—would have been ignored, resulting in more deaths, not less.
Smart engineers know not only which methods work but which methods have failed and why. The same is true of successful entrepreneurs. The best business books I’ve read this year—Ben Horowitz’ The Hard Thing about Hard Things comes to mind—are filled with practices that we should avoid, not just business platitudes (skin in the game) and truisms (have a bias for action) that we’re instructed to follow. While we can’t plan for every possible failure—engineering catastrophes will happen as long as we build, just as bankruptcy is an important part of the economy—we can avoid mistakes we’ve already committed.
Lesson: Acknowledge and value failure so that you can learn from your mistakes. Conduct postmortems, just as FAA investigates every aviation accident.
(Pitched this to FC, they weren’t interested. Might pitch to Fortune, eventually to 250Words.)