How to Transform Failure into Triumph

Man errs as long as he strives.

—GOETHE

IN THE EARLY stages of development, rockets tend to blow up, drift off course, and otherwise explode. The rockets that were launched as precursors to the Moon landing were no exception. Problems occurred in virtually every mission.

In December 1957, two months after the Soviet satellite Sputnik had become the first in Earth orbit, Americans attempted to even the score.1 The rocket, called Vanguard, launched about four feet above the pad, hesitated, and then sank back down, exploding on national television and earning itself nicknames like Flopnik, Kaputnik, and Stayputnik.2 The Soviets were quick to rub salt in the Americans’ cosmic wound. They inquired whether the United States was interested in receiving foreign aid earmarked for “undeveloped countries.”

In August 1959, the unmanned rocket Little Joe 1 got a little too excited. Because of an electrical problem, it decided to launch itself half an hour before schedule, as NASA personnel watched dumbfounded.3 It crashed after flying for twenty seconds. In November 1960, the launch of the Mercury-Redstone rocket became known as the “four-inch flight.” The rocket lifted just four inches off the ground before settling back down on the pad.4

Numerous mishaps occurred in manned missions as well. To cite one memorable example, a problem on Gemini 8 nearly claimed the life of Neil Armstrong three years before he stepped foot on the Moon.5 Gemini 8 was a complex mission that would mark the first time two spacecraft docked in orbit. A radio-controlled target vehicle, called the Agena, would be launched into orbit first, followed by Gemini 8, which would rendezvous and dock with the Agena.

The successful docking was followed by panic. Long before the movie Apollo 13 made the line famous, astronaut Dave Scott radioed to Houston, “We have serious problems here.” Gemini 8 had started spinning wildly—more than one revolution per second—blurring the astronauts’ vision and threatening vertigo and loss of consciousness. As the spacecraft continued to spin out of control, a cool and collected Armstrong jettisoned the Agena, switched to manual controls, and fired the opposite thrusters to slow down the spin.

THE “FAIL FAST, fail often, fail forward” mantra is all the rage in Silicon Valley. Failure is viewed as inspirational fodder, a rite of passage, a secret handshake shared by the insiders. Countless business books instruct entrepreneurs to embrace failure and flaunt it as a badge of honor. There are conferences, such as FailCon, dedicated to celebrating failure and FuckUp Nights, where thousands have gathered in more than eighty-five countries to toast their failures.6 There are funerals for failed start-ups, complete with bagpipes, DJs, sponsorships by liquor companies, and slogans like “Putting the Fun in Funeral.”7

Most rocket scientists would bristle at this cavalier attitude toward failure. In rocket science, failure can mean the loss of human life. Failure can also cost taxpayers hundreds of millions of dollars. Failure means that decades of work goes up in smoke—literally and figuratively. No one celebrated the numerous explosions and mishaps that occurred during the race to the Moon. They were embarrassing. They were catastrophic. And they weren’t taken lightly.

In this chapter, I’ll use a rocket-science framework to explain why it’s as dangerous to celebrate failure as it is to demonize it. Rocket scientists apply a more balanced approach to failure. They don’t celebrate it; nor do they let it get in their way.

In the first and second parts of this book (“Launch” and “Accelerate”), we explored how to ignite, refine, and test breakthrough ideas. Pursuing audacious ideas means daring greatly, and daring greatly means some of those ideas are going to fail when they collide with reality. So we begin this last stage of the book—“Achieve”—with failure.

You’ll learn why most of us think about failure the wrong way and how we can redefine our relationship with it. I’ll reveal how elite companies build failure into their business models and create an environment where employees willingly reveal their mistakes, instead of concealing them. I’ll share with you one of the biggest misconceptions about rocket science that appears in a Hollywood blockbuster and what the development of Viagra can teach you about failure. You’ll walk away from the chapter with science-backed ways to fail gracefully and create the right conditions for learning from failure.

Too Afraid to Fail

We’re wired to fear failure. Centuries ago, if we didn’t fear failure, we became prey to a ravenous grizzly bear. Growing up, failure got us into the principal’s office. Failure meant getting grounded or getting our allowance cut. Failure meant dropping out of college or not getting our dream job.

There’s no denying it; failure sucks. In most aspects of life, there are no participation trophies. When we fail a class, go bankrupt, or lose our job, we’re in no mood to celebrate. We feel worthless and weak. Unlike the high of success, which quickly dissipates, the sting of failure lingers—sometimes for a lifetime.

To ward off the bogeyman of failure, we keep a safe distance from it. We stay off the edges, avoid healthy risks, and play it safe. If we aren’t guaranteed to win, we assume the game isn’t worth playing.

This natural tendency to avoid failure is a recipe for failing. Behind every rocket unlaunched, every canvas unpainted, every goal unattempted, every book unwritten, and every song unsung is the looming fear of failure.

Thinking like a rocket scientist requires redefining our troubled relationship with failure. It also requires correcting one of the greatest misconceptions about rocket science popularized by a Hollywood blockbuster.

Failure Is an Option

In Apollo 13, there’s a scene where a group of rocket scientists are gathered in a room after learning that the spacecraft suffered an oxygen tank explosion on its voyage to the Moon. The spacecraft’s power is dangerously low, and the astronauts’ days are numbered. The scientists in mission control must figure out a way to get them back before their power runs out. “We never lost an American in space. We’re sure as hell not going to lose one on my watch,” roars Gene Kranz, the flight director, before adding the punch line: “Failure is not an option.” Kranz later wrote an autobiography with the same title, where he described the tagline as “a creed that we all lived by” in mission control.8 NASA gift shops quickly capitalized on the credo and began selling T-shirts emblazoned with the words Failure Is Not an Option.

The mantra makes sense when you’ve got human lives at stake. But as a descriptor for how rocket science works, it’s misleading. There’s no such thing as a zero-risk rocket launch. You still have to compete with physics. You can plan for some mishaps, but the cosmic banana peel is always around the corner. Accidents are inevitable when you’re creating a controlled explosion in a machine as complex as a rocket.

If failure weren’t an option, we never would have dipped our toes into the cosmic ocean. Doing anything groundbreaking requires taking risks, and taking risks means you’re going to fail—at least some of the time. “There’s a silly notion that failure’s not an option at NASA,” Elon Musk says. “Failure is an option here [at SpaceX]. If things are not failing, you are not innovating enough.”9 It’s only when we reach into the unknown and explore ever-greater heights—and in so doing, break things—that we move forward.

The same is true for scientists working in a lab. For them, without the ability to be wrong, they could never be right. Some of their experiments succeed and others don’t. If things don’t work as planned, it’s a hypothesis proven wrong. They can tweak the hypothesis, try a different approach, or abandon it altogether.

British inventor James Dyson described the inventor’s life as “one of failure.”10 It took Dyson fifteen years and 5,126 prototypes to get his revolutionary bagless vacuum to work. Several of Einstein’s attempts to devise a proof for E = mc2 failed.11 In some fields—for example, pharmaceutical drug development—the average failure rate is over 90 percent. If these scientists lived by the “failure is not an option” mantra, the self-loathing, the shame, and the embarrassment would all cripple them.

A moratorium on failure is a moratorium on progress.

If you’re in the business of taking moonshots—if you’re going to experiment with bold ideas—you’re going to miss more often than you connect. “Experiments are by their very nature prone to failure,” Jeff Bezos explained. “But a few big successes compensate for dozens and dozens of things that didn’t work.”12

Remember the Amazon Fire phone? The company lost $170 million over that misfire.13 Or Google Glass, designed by X, Google’s moonshot factory?14 The Glass was supposed to be the next best thing after the smartphone, but it flopped. It’s one thing to carry a smartphone in your pocket, consumers thought, and something else to attach one to your cornea. This was one piece of hardware that was decidedly uncool to sport. People wearing it were dubbed “glassholes.”

These failures are built into X’s business model. To X, killing projects is a “normal part of doing business,” as the company’s head Astro Teller puts it. It’s not unusual for X to kill over a hundred ideas in a single year.15 “Because X is premised on the idea of pursuing highly risky projects,” Kathy Cooper of X explains, “there’s just an understanding that a lot of them aren’t going to work. So it’s not seen as surprising or the fault of anyone if something doesn’t work.”16 By normalizing failure, X makes moonshot thinking the path of least resistance.

Not everyone can afford to swing and miss to the tune of $170 million, as Amazon did with the Fire. The size of your investment may differ dramatically, but the underlying principle remains the same: Treating failure as an option is the key to originality. “When it comes to idea generation,” Adam Grant writes in Originals, “quantity is the most predictable path to quality.”17 Shakespeare, for example, is known for a small number of his classics, but in the span of two decades, he penned 37 plays and 154 sonnets, some of which have been “consistently slammed for unpolished prose and incomplete plot and character development.”18 Pablo Picasso produced 1,800 paintings, 1,200 sculptures, 2,800 ceramics, and 12,000 drawings—only a fraction of which are noteworthy.19 Just a handful of Einstein’s hundreds of publications had real impact.20 Tom Hanks, one of my all-time favorite actors, admits, “I’ve made an awful lot of movies that didn’t make any sense, and didn’t make any money.”21

But when we judge the greatness of these individuals, we don’t focus on their troughs. We focus on the peaks. We remember the Kindle, not the Fire. We remember Gmail, not the Glass. We remember Apollo 13, not The Man with One Red Shoe.

It’s one thing to acknowledge that failure is an option. But it’s something else entirely to celebrate it. To take the sting and shame out of failure, Silicon Valley overcorrected. The pendulum swung too far in the other direction.

The Problem with Fail Fast

The fail-fast mantra has no place in rocket science. When each failure is horrifically expensive—in terms of both money and human lives—we can’t rush to the launch pad with a crappy rocket and fail as fast as possible.

Even outside rocket science, the fail-fast refrain is misguided. When entrepreneurs are too busy failing fast and celebrating it, they stop learning from their mistakes. The clinking of champagne glasses mutes the feedback they might otherwise receive from failure. Failing fast, in other words, doesn’t magically produce success. When we fail, we’re often none the wiser.

Consider a study of nearly nine thousand American entrepreneurs who founded companies between 1986 and 2000. The study compared the success rates—defined as taking a company public—of first-time founders and founders who had previously failed in business. You might expect that the experienced founders—having launched a business before and presumably learned from their failure—would be much more likely to succeed than those who had never started a business before. But that’s not what the study found. The success rate of first-time entrepreneurs was nearly equal to the success rate of entrepreneurs who had previously failed in business.22

Another study is also on point. Researchers examined 6,500 cardiac procedures by seventy-one surgeons over a ten-year period. They found that the surgeons who botched a procedure performed worse on later procedures.23 The results suggest that the surgeons not only failed to learn from their mistakes but also ended up reinforcing bad habits.

What explains these counterintuitive results?

When we fail, we often conceal it, distort it, or deny it. We make the facts fit our self-serving theory rather than adjust the theory to fit the facts. We attribute our failure to factors beyond our control. In our own failures, we overestimate the role of bad luck (“Better luck next time”). We blame the failure on someone else (“She got the job because the boss likes her more”). We come up with a few superficial reasons for why things went south (“If only we had more cash reserves”). But personal culpability seldom makes the list.

“What’s a little white lie?” you might ask. After all, putting a positive spin on failure can help us save face. But here’s the problem: If we don’t acknowledge we failed—if we avoid a true reckoning—we can’t learn anything. In fact, failure can make things worse if we get the wrong messages from it. When we attribute our failures to external factors—the regulators, the customers, the competitors—we have no reason to change course. We throw good money after bad, double down on the same strategy, and hope the wind blows in a better direction.

Here’s what most people get wrong about persistence. Persistence doesn’t mean repeatedly doing what’s failing. Remember the old adage about the futility of doing the same thing over and over again and expecting different results? The goal isn’t to fail fast. It’s to learn fast. We should be celebrating the lessons from failure—not failure itself.

Learn Fast, Not Fail Fast

The hardest part of getting to Mars is clearing a hurdle right here on Earth. NASA doesn’t build and operate Martian spacecraft all on its own.24 When it’s planning a new mission, it makes a formal announcement that describes, in general terms, the spacecraft that NASA intends to send and the science it expects the spacecraft to conduct. The announcement solicits proposals from anyone interested in sending scientific instruments to space. The number of great ideas far exceeds the funding, so NASA uses a Darwinian process to select only the strongest proposal. All others fail. This competitive system is as it should be: A cheap mission to Mars costs American taxpayers half a billion dollars.

My former supervisor, Steve Squyres, began writing proposals in 1987 to lead a Mars mission.25 Over the ensuing ten years, every single one of his ideas was shot down. “It was bitterly disappointing when you put years of effort and hundreds of thousands of dollars into writing a proposal,” Squyres recalls. But he doesn’t accuse NASA of failing to see the genius in his proposals. Rather, he places the blame squarely on himself. “The early [proposals] weren’t good enough,” Squyres admits. “They didn’t deserve to be selected.”

There are two responses to negative feedback from a credible source: Deny it or accept it. Every great scientist chooses the latter, and Squyres did the same. Each proposal he submitted to NASA was better than the one that came before it.

After ten years of learning, tweaking, and improving, Squyres’s proposal—which would eventually become the 2003 Mars Exploration Rovers mission—was finally selected in 1997. But selection didn’t guarantee flight. The mission was scrapped and brought back to life three times—most recently, after the crash of the Mars Polar Lander, which, as described in earlier chapters, used the same landing mechanism our group was planning to use. The mission was salvaged by two questions that reframed the problem: What if we used airbags instead of a three-legged lander? And what if we sent two rovers instead of one?

After we doubled up on rovers—named Spirit and Opportunity—and got our launch ticket back, malfunctions occurred about every month. During testing, the parachutes displayed a problem called squidding. For unknown reasons, they would pulsate like a squid, opening up and flapping closed repeatedly—a problem not seen in a parachute like ours in thirty years.26 One of the cameras on board the rovers developed an inexplicable “speckling” problem that overwhelmed the images with static.27 Two months before launch, we blew the fuse on Spirit.

In late June 2003, I flew down to Florida for Opportunity’s launch. In advance of the launch, we got together on Cocoa Beach for a private team meeting with no agenda, our heads gazing skyward toward Destination Mars. As we were popping champagne corks to commemorate the occasion, we learned that the cork on our rocket had popped too.28 The cork, which provides thermal insulation to the rocket, wouldn’t stick and kept peeling off. Our launch was delayed for several days as we scrambled to find a solution. We came dangerously close to the end of our allotted launch window until someone came up with the ingenious idea of using a resilient superglue called red RTV, available at Home Depot. With red RTV to the rescue, we took off for the red planet.

Each failure proved to be an invaluable learning opportunity. Each failure revealed a flaw that required correction. Each failure was followed by progress toward the ultimate goal. Although these failures took their toll on us, we couldn’t have landed safely on Mars without them.

These failures are what business school professor Sim Sitkin calls “intelligent failures.” They happen when you’re exploring the edges, solving problems that haven’t been solved, and building things that may not work.

We often speak of intelligent failures as losses. “I lost five years of my life.” “We lost millions of dollars.” But these are losses only if you call them that. You can also frame them as investments. Failure is data—and it’s often data you can’t find in a self-help book. Intelligent failures, if you pay them proper attention, can be the best teachers.

These errors can have staying power that lessons from success often lack. Intelligent failures can produce a sense of urgency for change and produce the shock necessary to unlearn what we know. “Give me a fruitful error any time, full of seeds, bursting with its own corrections,” Vilfredo Pareto wrote, “You can keep your sterile truth for yourself.”29

Thomas Edison recounted the story of a conversation with an associate who lamented that after thousands of experiments, he and Edison had failed to discover anything. “I cheerily assured him that we had learned something,” Edison recalled. “For we had learned for a certainty that the thing couldn’t be done that way, and that we would have to try some other way.”30

Learning can also take the stigma out of failure. “The best thing for being sad,” the author T. H. White wrote, “is to learn something. That’s the only thing that never fails. You may grow old and trembling in your anatomies, you may lie awake at night listening to the disorder of your veins, you may miss your only love, you may see the world about you devastated by evil lunatics, or know your honour trampled in the sewers of baser minds. There is only one thing for it then—to learn. Learn why the world wags and what wags it.”31

Without opportunities to learn why the world wags and what wags it, failure has no upside. But if you’ve learned something—if this failure means you’re more likely to succeed when you try again—failure won’t hit you as hard. Learning takes despair and turns it into excitement. With a growth mindset, you can maintain forward momentum even as the explosions pile up, the work gets hard, and the obstacles begin to appear insurmountable. As Malcolm Forbes, the founder of Forbes magazine, put it, “Failure is success if we learn from it.”

Squyres’s failed proposals for Mars missions are still sitting on his desk. “I can look at those old proposals,” he says, “and I can look at the things we did wrong, and I can look at the lessons that we learned and how we made things better, and I can see why on our fourth try we finally got selected.”

Just a few years after our rovers set sail for the red planet, it would take another group of rocket scientists four attempts to get it right.

The Opening and the Finale

The third time’s the charm.32

In August 2008, this is what SpaceX employees were telling themselves as they awaited the third launch of Falcon 1, the company’s very first rocket. At the time, outside observers were already busy drafting the obituary of what they thought was Musk’s vanity project. When Musk started SpaceX, he invested a hundred million dollars of his own money in the company—enough for three launches.

The first two failed.

Falcon 1’s maiden flight in 2006 lasted for all of thirty seconds. A fuel leak caused an unexpected fire in the engine, shutting it off and sending the rocket plummeting into the Pacific. “The first launch failure was heartbreaking,” SpaceX executive Hans Koenigsmann recalls. “We learned a lot of things we did wrong, and learning sometimes hurts.” The leak was blamed on corrosion around an aluminum nut that secured the fuel line. To correct the problem, the company replaced the aluminum fasteners with stainless steel ones that were more reliable and had the benefit of being cheaper.

The Falcon 1 was back on the launch pad a year later in 2007 for a second attempt. This flight got farther—clocking in at 7.5 minutes—but also failed to reach orbit after fuel stopped flowing into the engine. The failure “didn’t feel anywhere as harsh as the first time,” Koenigsmann says. “The vehicle actually flew very far, and then didn’t make orbit, but at least it flew out of sight.” Despite the ultimate failure, most mission objectives were met: Falcon 1 could launch and reach space. The anomalies that caused the trouble were quickly diagnosed and fixed.

The third attempt came a year later. Although 2008 was a bad year for many people, Musk says it was the worst of his life. His electric car company, Tesla, was flirting with bankruptcy, the world had spun into a financial crisis, and Musk had just gotten divorced. He was borrowing money from his friends to pay rent. He had plunged much of his fortune into SpaceX, and the two Falcon 1 failures had eaten into his investment. What was left of it was sitting on the launch pad awaiting a perilous flight.

On the third attempt, Falcon 1 rumbled to life and took off carrying three satellites and the ashes of James Doohan, the actor who played Scotty in Star Trek: The Original Series (think “I’m givin’ her all she’s got, Captain!”). It soared into the sky, executing a perfect flight of its first stage (recall that rockets are built in stages stacked on top of each other). After the first stage took the spacecraft into space, it was time for stage separation—the critical point in the flight where the rocket’s first stage detaches and falls away after running out of fuel. That’s when the smaller, second stage kicks in to take the spacecraft into orbit. The stages separated as scheduled, but the first stage didn’t stop. It fired again and bumped into the second. “We rear-ended ourselves,” SpaceX’s president, Gwynne Shotwell, recalls. “It was almost Monty Pythonesque.”

The problem was missed during testing because SpaceX had failed to follow the test-as-you-fly principle. The engine pressure that resulted in the unexpected boost of thrust was below the ambient pressure in SpaceX’s ground testing facility, so it had barely registered. But in the vacuum of space, the same pressure produced enough of a kick to cause a catastrophic collision.

For SpaceX, this failure was strike three. Hundreds of shell-shocked SpaceX employees, who had been working seventy- to eighty-plus-hour weeks for six years, awaited word from their boss at SpaceX’s factory in Hawthorne, California. “The mood in the building hung thick with despair,” recalls former SpaceX employee Dolly Singh. Musk emerged from the control room, where he was commanding the mission along with senior engineers. He walked past the press to address his troops, who had just lost their third consecutive major battle.

Musk told them that they knew the project was going to be hard. He reminded them that what they were trying to do was, after all, rocket science. The company’s rockets had reached space, accomplishing what major countries had failed to accomplish. Then came the surprise: Musk announced he had secured an investment that would get SpaceX two more launches. This wasn’t the end. As Shane Snow describes it, Musk told his troops they would “learn what had happened tonight and they would use that knowledge to make a better rocket. And they would use that better rocket to make even better rockets. And those rockets would one day take man to Mars.”33

It was time to get back to work. “Within moments,” Singh recalls, “the energy of the building went from despair and defeat to a massive buzz of determination as people began to focus on moving forward instead of looking back.” The likely culprit for the failure was identified within a matter of hours. “When I saw the video, it was like, ‘OK. We can figure it out,’” Shotwell explains. The solution was simple: introduce a longer delay before stage separation to prevent a collision. “Between the third and the fourth flight we changed one number, nothing else,” Koenigsmann says.

In less than two months, SpaceX was back on the launch pad. “Everything hinged on that launch,” recalls Adeo Ressi, Musk’s college friend. “Elon had lost all his money, but this was more than his fortune at stake—it was his credibility.” If the fourth launch failed, “it would have been over. We’re talking Harvard Business School case study—rich guy who goes into the rocket business and loses it all.”

But the rocket didn’t fail. On September 28, 2008, SpaceX’s Falcon 1 launched out of the atmosphere and into the record books, becoming the world’s first privately built spacecraft to reach Earth’s orbit.

When SpaceX survived its baptism by fire on its fourth try, everyone took notice—particularly the bureaucrats at NASA looking to sustain the American space program after the expected retirement of the space shuttle in 2010. In December 2008, three months after Falcon 1’s successful voyage, NASA handed SpaceX a lifeline in the form of a $1.6 billion contract for resupply missions to the International Space Station. When NASA officials called to give him the good news, an otherwise stern Musk broke out of character and screamed, “I love you guys!” For SpaceX, Christmas had come early.

To paraphrase F. Scott Fitzgerald, there’s a difference between a single failure and final defeat.34 A single failure, as SpaceX’s story illustrates, can be the beginning, not the end. Many outside observers called the three Falcon 1 crashes failures—mistakes committed by a team of amateurs led by a rich kid playing with expensive toys. But labeling these crashes failures was like calling a tennis match before it’s over. “I’ve come from behind too often,” the great tennis champion Andre Agassi writes, “and had too many opponents come roaring back against me, to think that’s a good idea.”35

The opening doesn’t have to be grand, as long as the finale is.

Time morphs how we view events. Something that looks like a failure in the short term changes when we zoom out and put on a broader lens. Pixar’s former president Ed Catmull calls the initial ideas behind the studio’s blockbuster animation films “ugly babies.” All their films start out “awkward and unformed, vulnerable and incomplete.”36 But if the game doesn’t end until the film is released, an early version gone wrong isn’t a catastrophe. It’s a momentary blip. A temporary glitch. A problem to be solved.

Breakthroughs are often evolutionary, not revolutionary. Take a look at any scientific discovery, and you’ll find there is no magical it. No single aha moment. Science weaves from failure to failure, with each version better than the one that came before. From a scientific perspective, failure isn’t a roadblock. It’s a portal to progress.

We embodied this mindset as children. When we learned how to walk, we didn’t get it right on the first try. No one told us, “You’d better think hard about how you take that very first step because you get one step and that’s it.” We repeatedly fell. With each fall, our bodies learned what to do and what not to do. By learning not to fall, we learned how to walk.

Nothing comes to existence perfectly formed. Rome, as the saying goes, wasn’t built in a day. The Apollo 11 spacecraft that put Armstrong and Aldrin on the Moon didn’t just spring out of the factory. It took numerous iterations—through the Mercury, Gemini, and earlier Apollo missions—to get it right.

For scientists, each iteration is progress. If we get a glimpse into the dark room, that’s a contribution. If we don’t find what we thought we’d find, that’s a contribution. If we change a single unknown unknown to a known unknown, that’s a contribution. If we ask a better question than the ones asked before, that’s a contribution, even if the answers elude us.

Which inexorably brings us to Matt Damon. In the movie version of the terrific book The Martian, Damon’s character, Mark Watney, teaches astronauts-in-training what to do in case of impending doom. “At some point, everything’s gonna go south on you and you’re going to say, ‘This is it. This is how I end,’” Watney says. You can either accept that as a failure—or you can get to work. “You do the math. You solve one problem. And you solve the next one. And then the next. If you solve enough problems, you get to come home.”

If you solve enough problems, you get to land your rovers on Mars. If you solve enough problems, you get to build the Roman Empire. If you solve enough problems, you get to land on the Moon.

That’s how you change the world. One problem at a time.

Changing the world one problem at a time requires delaying gratification. Most things in life are “first-order positive, second-order negative,” as Shane Parrish writes on his website Farnam Street.37 They give us pleasure in the short term but pain in the long. Spending money now instead of saving for retirement, using fossil fuels instead of renewable energy, guzzling sugar-laden beverages instead of water are all in that category.

When we’re focused on first-order outcomes, we look for the instant success, the instant best seller, the instant fill-in-the-blank. We search for shortcuts, life hacks, and advice from self-proclaimed gurus. We “applaud the wrong things: the showy, dramatic record-setting sprint,” Chris Hadfield writes, “rather than the years of dogged preparation or the unwavering grace displayed during a string of losses.”38 What’s more, failure is expensive in the short term. When we’re trying to maximize our profits and comfort tomorrow, we discount the value that failure brings in the long term. As a result, failure hits us hard. To boost our short-term pleasure, we avoid doing things that might fail.

Those who get ahead in life flip this perspective. “A real advantage is conferred on people who can do things that are first-order negative, second-order positive,” Parrish writes.39 These people delay gratification in a world that has become obsessed with it. They don’t quit simply because their rocket blew up on the launch pad, they had a bad quarter, or their audition fell flat. They reorient their calibration for the long term, not for the short.

When it comes to creating long-lasting change, there are no hacks or silver bullets, as venture capitalist Ben Horowitz says. You’ll need to use a lot of lead bullets instead.40

Inputs over Outputs

Think back on the failures you’ve had in your life. If you’re like most people, you’ll picture the bad outcomes—the business that never took off, the penalty kick you missed, or the job interview you bombed. Poker players, as Annie Duke explains in Thinking in Bets, refer to this tendency to “equate the quality of a decision with the quality of its outcome” as “resulting.”41 But as Duke argues, the quality of an input isn’t the same as the quality of the output.

Focusing on outputs leads us astray because good decisions can lead to bad outcomes. In conditions of uncertainty, outcomes aren’t completely within your control. An unforeseeable dust storm can cripple a perfectly designed Martian spacecraft. A bad wind can misdirect a perfectly shot soccer ball. A hostile judge or jury can derail a great case.

If we engage in resulting, we reward bad decisions that lead to good outcomes. Conversely, we change good decisions merely because they produced a bad outcome. We start shaking things up, reorganizing departments, or firing or demoting people. As one study shows, National Football League (NFL) coaches change their lineup after a one-point loss, but don’t change it after a one-point win—even though these minor score differences are often poor indicators of player performance.42

Most of us act like American football coaches, treating success and failure as binary outcomes. But we don’t live in a binary world. The line between success and failure is often razor thin. “Failure hovers uncomfortably close to greatness,” wrote James Watson, the codiscoverer of DNA’s double-helix structure.43 The same decision that produced a failure in one scenario can lead to triumph in others.

The goal, then, is to focus on the variables you can control—the inputs—instead of the outputs. You should ask, “What went wrong with this failure?” and if the inputs need fixing, you should fix them. But this question isn’t enough. You must also ask, “What went right with this failure?” You should retain the good-quality decisions, even if they produced a failure.

Consider Amazon’s response to its Fire phone fiasco. Viewed through the standard metrics of output like profitability, the Fire was a colossal failure. But Amazon looked beyond the outcome. “When we try a new project, we look at the inputs,” says Amazon’s Andy Jassy.44 “Did we hire a great team? Did the team have thoughtful ideas? Did they think the idea all the way through? Did they execute in a timely fashion? Was the quality high? Was the technology innovative?” Even if the project fails, you can take the inputs that worked and use them in future projects. “Not only did we take the learning from [the Fire] technology,” Jassy explains, “but we also took all of the technology we built and applied it to a bunch of other services and capabilities.”

Inputs aren’t sexy. The word input might be better reserved for a boring database software. But an input-focused mind is the mark of anyone who has achieved anything extraordinary. The amateur focuses on getting hits and expects short-term results. The professional plays the long game and prioritizes inputs, perfecting them for years with no immediate payoff. This is why the tennis player Maria Sharapova describes focus on outcomes as the worst mistake that beginning tennis players make.45 Watch the ball as long as you can, Sharapova cautions, and zero in on the inputs. By taking the pressure off the outcome, you get better at your craft. Success becomes a consequence, not the goal.

This reorientation toward inputs has another upside. If you find yourself resenting the inputs, you might be chasing the wrong output. There’s a question that frequently shows up in self-help books: What would you do if you knew that you could not fail? This isn’t the right question to ask. Instead, do as Elizabeth Gilbert does, and flip the question on its head: “What would you do even if you knew that you might very well fail? What do you love doing so much that the words failure and success essentially become irrelevant?”46 When we switch to an input-focused mindset, we condition ourselves to derive intrinsic value out of the activity. The input becomes its own reward.

With an input-focused mindset, you’re free to change your destination. Goals can help you focus, but that focus can also turn into tunnel vision if you refuse to budge or pivot from your initial path.

For example, when Google Glass was roundly dismissed as a pointless product, X found a different path. Once the product hit the consumer market, the company realized that the Glass wasn’t a consumer product at all. Instead, X learned from that failure and reinvented the Glass as a tool for businesses.47 You can now find Google Glass on countless workers, including Boeing employees working on aircraft and doctors looking through a patient chart using a fancy attachment to their faces.48

Consider another example from the pharma industry. In 1989, Pfizer scientists developed a new drug called sildenafil citrate. Researchers hoped the drug would expand blood vessels to treat angina and high blood pressure associated with heart disease. By the early 1990s, the drug appeared to be ineffective for its intended purpose. But the participants in the trials reported an interesting side effect—erections. It wasn’t long before researchers abandoned their initial hypothesis to pursue the astonishing alternative. And Viagra was born.49

Focusing on inputs has another upside. You avoid the wild swings of misery and euphoria that come with chasing outcomes. Instead, you become curious—no, fascinated—about tweaking and perfecting the inputs.

How Fascinating!

Mike Nichols was the prolific film director behind many classics, including The Graduate.50 Although people tend to remember Nichols’s hits, many of his films were flops. Some of these duds would appear from time to time—as flops do—on late-night television. Whenever Nichols came across one of his failures, he would park himself on his couch and watch the whole thing from start to finish.

As he sat and watched, what’s important is what he wouldn’t do. He wouldn’t cringe. He wouldn’t look away. He wouldn’t blame the damn critics.

He’d simply watch and think, “That’s so interesting, how that scene didn’t work out.” Not “I’m a loser.” Not “This is awful.” Not “What a complete embarrassment.” Instead, with no judgment, he’d ponder, “Isn’t it funny how sometimes things work and other times they don’t?”

Nichols’s approach reveals the secret to taking the sting out of failure. Curiosity takes a failure, turns the volume of drama all the way down, and makes failure interesting. It provides emotional distance, perspective, and an opportunity to view things through a different lens.

In their terrific book, The Art of Possibility, Rosamund Stone Zander and Benjamin Zander offer a practical method for putting this mindset to practice. Every time you make a mistake, every time you fail at something, you should throw your arms in the air and say, “How fascinating!”51

Fair warning: If you’re anything like me, you’ll grumble when you first do this. As you try to put your arms in the air, they’ll go up ever so slowly—as if you’re doing an imaginary bench press with really, really heavy weights. And the phrase “How fascinating!” will sound more petulant than joyous.

That’s okay. Do it anyway. As you bask in the glory of your fascination, start asking some questions. What can I learn from this? What if this failure was actually good for me?

If you need inspiration, just picture Mike Nichols, sitting on his couch—not complaining about how the gods have turned on him by broadcasting his biggest failures on television for the world to see—but smiling, nodding, and knowing that watching this failure with curiosity means he’ll do better the next time.

Flying Blind

Failure, as we’ve seen, is the portal to discovery, innovation, and long-term success. But most organizations suffer from collective amnesia over their failures. Mistakes remain concealed because employees are too afraid to share them. Most companies tell their employees, explicitly or implicitly, that if you succeed—according to short-term, quantifiable metrics like profits—you get a big pot of money, a better office, and a better title. If you fail, you get nothing. Or worse, you get shown the door.

This incentive scheme only exacerbates the deeply ingrained inertia against owning up to our failures. When we reward success and punish failure, employees will underreport failures, overreport successes, and reframe anything that falls in between in the best possible light. When we shoot the messengers, people stop delivering messages—particularly if they work for us. In one study, 42 percent of surveyed scientists across nine federal agencies (including NASA) feared retaliation for speaking out.52 Of more than forty thousand employees surveyed at a tech company, 50 percent believed it wasn’t safe to speak up at work.53

But failures transmit invaluable signals. Your goal should be to pick up these signals before your competitors do. But in most environments, these signals are elusive whispers that don’t rise above the noise. If you can’t hear them, if you suppress them, or if you shed them before they stick, you can’t learn from them.

This is why airplanes carry flight recorders called black boxes. They record everything, including conversations in the cockpit and data from the airplane’s electronic systems. The name black box is actually a misnomer because the box is bright orange to make it easier to find after a crash. The box is also fireproof, shockproof, and waterproof because the data it holds is crucial to uncovering why an accident happened.

We omit black boxes from our life to our detriment. Let’s return for a moment to the crash of the 1999 Mars Polar Lander. Recall from earlier that the Lander crashed most likely because its engines shut down prematurely. But we don’t know what happened for sure. Because money was tight, the lander lacked a way to communicate with mission control during its descent to the Martian surface. The team members had to cut corners, and this particular corner they cut deprived them—and all future rocket scientists—of the ability to extract critical lessons from this $120 million mishap.54

The omission resulted in part because the Mars Polar Lander was viewed myopically as a single project. If management had viewed the lander as part of a comprehensive whole—one probe among many interplanetary probes—then a communications device crucial for long-term learning should have been included.

To facilitate learning from failure, NASA catalogs mistakes in human spaceflight in a document called “Flight Rules.”55 The rules are a record of the past to guide the future. They’re a body of knowledge culled from decades of missteps and miscalculations to ensure that the lessons endure. The document contains thousands of anomalies that have come up during manned spaceflight since the 1960s and the solutions to them. The book preserves this institutional knowledge for future generations, giving each failure shape and purpose as part of a larger story. It also obviates the need to reinvent the wheel and allows employees to focus on new problems. But as with any set of rules, these should be guardrails, not handcuffs. They should guide but not constrain. As we saw earlier, historical processes can rigidify into inflexible rules that impede first-principles thinking.

NASA’s “Flight Rules” document works in part because the failures of others are the best catalyst for our own understanding. Our approach to failure is hypocritical. We explain away our own failures by blaming them on external factors. But when others stumble, we point to internal factors—they were careless, incompetent, or not paying enough attention. Our tendency to catalog the personal failings of others is why their mistakes can be a great source of learning. In one study, cardiac surgeons who observed their colleagues’ blunders got significantly better at performing the procedure.56 They homed in on the other surgeons’ mistakes and learned not to repeat them.

Although companies pay lip service to tolerating and documenting failures, they often fail in practice. When I speak to corporate executives on failure, some of them argue that if failure is tolerated, then failures will multiply. Failure means fault, and fault needs to be assigned. If these executives don’t discipline the responsible party, they assume they’ll end up nurturing an anything-goes culture, where it becomes okay to fail.

These beliefs are out of step with a generation of research. As you’ll see in the next section, you can create an environment of intelligent failures without complacency. You can allow people to take high-quality risks, but you can also set high standards. You don’t have to tolerate sloppy failures—repeatedly making the same mistakes or failing because of a lack of care. You can reward intelligent failures, sanction poor performance, and accept that some errors are going to be inevitable when you’re building things that may not work. People should be held accountable not for failing intelligently, but for failing to learn from it.

“There are two parts to failure,” Pixar’s former president Ed Catmull writes. “There is the event itself, with all its attendant disappointment, confusion, and shame, and then there is our reaction to it.” We don’t control the first part, but we do control the second. The goal, as Catmull puts it, should be “to uncouple fear and failure—to create an environment in which making mistakes doesn’t strike terror into your employees’ hearts.”57

Rewarding intelligent failure sounds simple in theory, but it’s difficult to implement in practice. A superficial commitment to “innovation” or “taking risks” won’t create a culture of intelligent failures. In the next section, we’ll explore how to create this ideal environment in the context of medicine, which provides a close analog to rocket science. The challenges on the operating room table aren’t that different from those on the launch pad. The stakes are high. The pressure is on. The tiniest mistake can prove fatal. In this environment, creating a culture of intelligent failure is difficult, but as we’ll see, it’s not rocket surgery.

Psychological Safety

Medication errors in hospitals—where the wrong drug is given to the patient—are shockingly common. A 1995 study found 1.4 medication errors per patient per hospital stay. And of those errors, roughly 1 percent produced complications and harmed the patient.58

Amy Edmondson, a professor at Harvard Business School, wanted to explore the cause of these medication errors.59 She asked herself, “Do better hospital teams make fewer medication mistakes?” To Edmondson, the answer seemed obvious. Better teams, with better-performing members and leaders, should make fewer errors.

But the results were the exact opposite. Better teams were making more mistakes, not less. What could explain this counterintuitive outcome?

Edmondson decided to dig deeper, sending a research assistant into the wild to observe the teams on the hospital floor. The assistant discovered that better teams weren’t making more mistakes. Instead, they were simply reporting more mistakes. The teams that had a climate of openness—where the staff felt safe to discuss mistakes—performed better because employees were more willing to share failures and actively work to reduce them.

Edmondson refers to this climate as “psychological safety.” I have to admit, when I first heard the term, I instinctively dismissed it as woo-woo. It conjured images of employees sitting around a conference table joining hands and sharing their feelings. But after studying the research, I backed down. The supporting evidence is rock solid. Psychological safety means, in Edmondson’s words, “no one will be punished or humiliated for errors, questions, or requests for help, in the service of reaching ambitious performance goals.”60

Research shows that psychological safety stimulates innovation.61 When people feel free to speak up, ask provocative questions, and air half-formed thoughts, it becomes easier to challenge the status quo. Psychological safety also increases team learning.62 In psychologically safe environments, employees challenge questionable calls by superiors instead of obediently complying with the commands.63

The best-performing hospital team in Edmondson’s study was led by a hands-on, highly accessible nurse manager who actively facilitated an open environment. During interviews, the manager explained that a “certain level of error” is expected on her team and that a “nonpunitive environment” is essential to uncovering the error and addressing it. The nurses working in the unit confirmed the manager’s statements. One noted that “people feel more willing to admit errors here, because [the manager] goes to bat for you.” On this team, it was the nurses themselves who shouldered the responsibility for mistakes. As the manager explained, the “nurses tend to beat themselves up about errors; they are much tougher on themselves than I would ever be.”64

The two worst-performing hospital teams had very different climates. In these teams, making a mistake meant getting punished. A nurse described one incident where she had inadvertently hurt a patient while drawing blood. The nurse manager put her “on trial,” she explained, “it was degrading, like I was a two-year-old.” Another nurse explained that “doctors condescend, and they bite your head off ” if you make a mistake. One nurse said it was like “being called into the principal’s office.” As a result, if a medication error happened, nurses didn’t advertise it, to save themselves from short-term embarrassment and anguish. By doing so, however, they discounted the long-term consequence of remaining silent—namely, injury or death of the patient because of a mistake.

This environment, in turn, led to a vicious cycle. The worst-performing teams—those that were in most need of improvement—were also the least likely to report errors. And if errors aren’t reported, the team can’t improve.

To encourage the reporting of failures, Google’s moonshot factory, X, takes an unusual approach.65 In most companies, it’s a senior leader who decides to pull the plug on a faltering project. But employees at X are empowered to kill their own projects as soon as they realize, for one reason or another, that the project isn’t viable.

Here’s the interesting part: For this act of hara-kiri, the entire team receives a bonus. Recall from an earlier chapter that X led a project called Foghorn to convert seawater into fuel by sucking carbon dioxide out of the ocean water. Although the technology was promising, it wasn’t economically viable, so the team decided to shut down its own project. “Thank you!” X’s head, Astro Teller, announced at an all-hands meeting. “By ending their project, this team has done more to speed up innovation at X this month than any other team in this room.”66

The notion of giving bonuses for failing might strike you as odd. It’s one thing to tolerate failure, but something else to reward it. But there’s genius in this incentive scheme. It’s more expensive for nonviable projects to continue; they waste money and resources.67 If a project has no future, shutting it down frees up precious resources for other moonshots that have better odds of landing. The resulting environments—where people are constantly generating intelligent failures—“remove the fear and make it safe for people to kill their project,” X’s Obi Felten explains.

Amazon follows a similar approach. If the quality of the inputs on a failed project was outstanding, then the team is rewarded through assignments in great new roles in the company—not punished. Otherwise, Amazon’s Andy Jassy says, “you’ll never get great people to take chances on new projects.”68

This mindset translates to a six-word mantra: “Reward excellent failures, punish mediocre successes,” as a seminar attendee once told author Tom Peters.69 There must be a clear commitment to supporting intelligent failure and well-intentioned risk taking. People must know that intelligent failure is necessary for future success, that they won’t be punished for it, and that their careers won’t be ended for it. If the signals are mixed, employees will err on the side of caution and hide their mistakes instead of revealing them.

There’s another component to psychological safety. If employees are to share their mistakes, the leaders must do the same.

Advertise Your Failures

It’s not easy for smart, competitive people to own up to their blunders, particularly when no one else has noticed them. But astronauts are expected to advertise their own missteps and put them under a microscope for everyone to see.70 Talking openly about screwups is mandatory because one astronaut’s admission of a boneheaded move can save another’s life.

Even where lives aren’t at stake, advertising our failures can facilitate learning and develop psychological safety. This is why I started the “Famous Failures” podcast, where I interview the world’s most interesting people about their failures and what they learned from them. As you might imagine, asking guests to appear on the show has made for some interesting conversations.

“Hey, Dan, I have a podcast where I interview failures. You’d be perfect for it.”

Surprisingly though, most people I’ve approached have been eager to appear on the show because they know firsthand what many others fail to recognize: Anyone who has done anything meaningful has failed in some fashion. Having interviewed numerous titans on the podcast—including top entrepreneurs, Olympic medalists, and New York Times best-selling authors—I have found one common thread: Everyone—and I mean everyone—is a walking imperfection. Even genius isn’t blunder-proof.

Einstein spoke openly about his biggest blunders. As astrophysicist Mario Livio writes, “More than 20 percent of Einstein’s original papers contain mistakes of some sort.”71 Sara Blakely, the founder and CEO of Spanx, highlights her own oops moments at company-wide meetings.72 Catmull, the former president of Pixar, talks about the mistakes he has made at new-employee orientations: “We do not want people to assume that because we are successful, everything we do is right,” he explains.73 Economist Tyler Cowen wrote a detailed analysis of how, in the lead-up to the 2008 financial crisis, he “badly underestimated the chance that something systemic had gone wrong in the American economy.” Cowen admitted his remorse: “I regret that I was wrong, and I regret that I was overconfident in my belief that I was right.”74

If these individuals now appear more endearing to you, you’re experiencing what researchers call the “beautiful mess effect.”75 Exposing your vulnerability can make you more desirable in the eyes of others. But there’s one caveat. You must establish your competence before revealing your failures. Otherwise, you risk damaging your credibility and coming across as a mess—and not a beautiful one.76

Despite the beautiful-mess effect, most of us are terrible at owning up to our blunders. Our public image is synonymous with our self-worth. We puff ourselves up and create curated portrayals of our imperfect lives. We round off the edges, airbrush the negatives, and present a perfect image to the world devoid of any failures. Even when we talk about our failures, we do so in a flattering light.

I get it. It’s painful to fail, and airing your failures can compound the pain. But the opposite approach—denial and avoidance—makes things worse. To learn and grow, we must acknowledge our failures without celebrating them.

This advice is particularly important for leaders. People pay close attention to the leader’s behavior since they depend on the leader for recognition.77 Studies also show that people look to the leader to initiate change.78 If leaders fail to acknowledge their failures—if there’s a perception that the leader can do no wrong—it’s unrealistic to expect employees to take the risk of challenging the leader or revealing their own mistakes.

Consider a study of sixteen hospitals with top-tier cardiac surgery departments that implemented a new technology for surgery.79 The technology upended how surgery was conducted. Each team had to unlearn ingrained habits and adopt different ones from scratch. The teams that learned quicker than the others shared three essential characteristics, one of which is particularly relevant here. They were led by surgeons who were more willing to acknowledge their own fallibility. For example, one surgeon repeatedly told his team, “I need to hear from you because I’m likely to miss things.”80 Another surgeon would say, “I screwed up. My judgment was bad in this case.”

What made these messages effective was their repetition. Entrenched behaviors don’t change with one impassioned speech. As team members heard these messages over and over again, they felt psychologically safe to speak up—even in an environment as hierarchical as an operating room. “There are no sacred cows,” a member of one surgery team explained. “If somebody needs to be told something, then they are told—surgeon or orderly.”81

Whether you’re in the operating room, the boardroom, or the mission control room, the principle is the same. The road to success is filled with potholes. You’re better off acknowledging them than pretending they don’t exist.

How to Fail Gracefully

Not all failures are created equal. Some are more graceful than others. Rocket scientists use a constellation of tools to contain failures so the blunders don’t create a cascade of damage. We covered some of these tools in earlier chapters. For example, rocket scientists conduct thought experiments where a failure produces no tangible damage. They build in redundancies so the mission doesn’t fail even if a component fails. They use tests to lower the stakes because failures on the ground prevent far more disastrous ones in space.

Beyond rocket science, you can also use tests to fail more gracefully. Instead of rolling out an innovative policy across the entire company, you can use one division or a subset of your customers as a laboratory or an experiment. If one division breaks, the company still stands. If a subset of customers hates the policy, the damage is contained. For example, Starwood Hotels—which includes hotel brands like Westin and Sheraton—often used its W Hotels brand as an innovation lab, a testing ground for new ideas like signature scents and a living room experience in the hotel lobby. If the ideas worked in the smaller pilots in W Hotels, the company would roll them out to the other hotels in its portfolio.82 If the ideas didn’t work, the damage would be contained.

Testing has another upside. By definition, it allows you to practice failure in a relatively safe environment. Rocket scientists fail regularly, but for many of us—particularly in newer generations—failure can be an unfamiliar experience. As Jessica Bennett writes in the New York Times, “Faculty at Stanford and Harvard coined the term ‘failure deprived’ to describe what they were observing: the idea that, even as they were ever more outstanding on paper, students seemed unable to cope with simple struggles.”83

Overcoming this fear requires exposure therapy. In other words, we must expose ourselves to failure regularly. Think of this as vaccination: Just as introducing weak antigens can stimulate “learning” in our immune system and prevent future infection, exposure to intelligent failures can allow us to recognize and learn from them. Each dose builds resilience and breeds familiarity. Each crisis becomes training for the next one.

This doesn’t mean imposing catastrophic failures on ourselves. We don’t have to be masochists. Rather, it means giving ourselves the breathing room to push boundaries, tackle thorny problems, and, yes, to fail. Let yourself fall on the grass. Give yourself permission to botch a song on the piano and write ghastly first drafts of book chapters (as I keep telling myself).

Parents can take a cue from Sara Blakely. She went from selling fax machines door-to-door to becoming the world’s youngest self-made woman billionaire. She credits her success partly to a question that her father would ask her every week when she was growing up. “What have you failed at this week?” If Sara didn’t have an answer, her father would be disappointed. To her father, failing to try was far more disappointing than failure itself.

WE OFTEN ASSUME that failure has an endpoint. We fail until we succeed and then stop failing to reap the benefits of our newly minted position in the pecking order. But failure isn’t a bug to get out of our system until success arrives. Failure is the feature. If we don’t develop a habit of failing regularly, we court catastrophe. As we’ll see in the next chapter, where failure ends, complacency begins.

Visit ozanvarol.com/rocket to find worksheets, challenges, and exercises to help you implement the strategies discussed in this chapter.

9

NOTHING FAILS LIKE SUCCESS