Reframing Questions to Generate Better Answers

How to Reframe Questions to Generate Better Answers

A problem well defined is a problem half solved. —UNKNOWN

To land on Mars is to execute a perfect cosmic choreography. “If any one thing doesn’t work just right, it’s game over,” NASA engineer Tom Rivellini explained.

For one thing, Mars is a rapidly moving target. Depending on its alignment with Earth, the red planet is between 35 million and 250 million miles away, orbiting the Sun at over 50,000 miles per hour. Landing on a specific site, at a specific time, requires nothing short of an interplanetary hole in one.

But the most dangerous part of the interplanetary journey isn’t the six months it typically takes a spacecraft to travel from Earth to Mars during a suitable launch window. Rather, it’s the six minutes of terror at the very end of that journey, when the spacecraft enters, descends, and (hopefully) lands on the surface.

During its journey, a typical Mars-bound lander rests inside a two-part aeroshell—a cocoon of sorts—with a heat shield in the front and a back shell on the opposite side. When the spacecraft touches the Martian atmosphere, it is barreling through space at more than sixteen times the speed of sound. In about six minutes, it must bleed off its 12,000-miles-per-hour velocity to land safely on the surface. As the spacecraft tears through the atmosphere, the temperatures outside climb up to over 2,600°F (or roughly 1,400°C). The heat shield protects the spacecraft from bursting into flames as the atmospheric friction slows it down to about 1,000 miles per hour.

That’s still really fast. At about six miles above the surface, the spacecraft deploys a supersonic parachute and jettisons the heat shield. But the parachute itself isn’t enough to slow down the spacecraft. The Martian atmosphere is thin—its density is less than 1 percent of Earth’s atmosphere—and parachutes work by creating drag with air molecules. The fewer molecules, the less drag. As a result, a parachute can bring down the spacecraft’s speed to only about 200 miles per hour. We need something else to reduce that velocity so the spacecraft doesn’t strike the surface at race-car speed.

In 1999, when I started working on the operations team for what would later become the Mars Exploration Rovers mission, that “something else” was a three-legged lander with rocket motors. After a parachute had reduced its speed, the lander would deploy the three shock-absorbing legs that had been tightly stowed away during the journey. The lander would then fire up its rockets, and using a radar, navigate down to the surface for a soft, steady touchdown on its three legs.

That was the theory. But there was a practical problem. The 1999 Mars Polar Lander, which used this landing system, died a swift death. A NASA review board concluded that the Lander had probably plummeted to the surface after a premature shutdown of its rocket motors.

From our perspective, this accident presented a significant challenge. We were planning to use the same landing mechanism as the Mars Polar Lander, and that mechanism had just failed spectacularly. Our mission was grounded.

Initially, we asked the obvious questions: How can we innovate on the flawed design of the Mars Polar Lander? How do we design a better three-legged lander to ensure a smooth landing? But these questions, as we’ll discover, weren’t the right questions to be asking.

The book’s first part ("Launch") explores first principles, thought experiments, and ambitious goals. Here the focus shifts to an earlier decision: which problem to solve. The initial question may assume a solution before we have considered the alternatives.

Start by separating the desired outcome from the tool or tactic currently being considered. Then ask what other approaches could achieve that outcome, what other uses your resources have, and what changes if you reverse an assumption.

The Sentence Before the Verdict

The way that most people solve problems reminds me of a scene from Alice’s Adventures in Wonderland. In the scene, the Knave of Hearts is on trial for supposedly stealing tarts. After the evidence is presented, the King of Hearts, who’s presiding as a judge, says, “Let the jury consider their verdict.” The impatient Queen of Hearts interrupts and retorts, “No, no! Sentence first. Verdict afterwards.”

In solving problems, we instinctively want to identify answers. Instead of generating cautious hypotheses, we offer bold conclusions. Instead of acknowledging that problems have multiple causes, we stick with the first cause that pops to mind. Doctors assume they have the right diagnosis, which they base on symptoms they have seen in the past. In boardrooms across America, executives, eager to appear decisive, fall over each other to be the first to deliver the correct answer to a perceived problem.

But this approach puts the cart before the horse—or the sentence before the verdict. When we immediately launch into answer mode, we end up chasing the wrong problem. When we rush to identify solutions—when we fall in love with our diagnosis—our initial answer hides better ones lurking in plain sight. When the sentence is announced first, the verdict is always the same: guilty. The difficulty lies, as John Maynard Keynes said, “not in the new ideas, but in escaping from the old ones.”

When we’re familiar with a problem, and when we think we have the right answer, we stop seeing alternatives. This tendency is known as the Einstellung effect. In German, einstellung means “set,” and in this context, the term refers to a fixed mental set or attitude. The initial framing of the question—and the initial answer—both stick.

The Einstellung effect is partly a relic of our education system. In schools, we’re taught to answer problems, not to reframe them. The problems are handed to—well, more like forced on—students in the form of problem sets. The phrase problem set makes this approach clear. The problems have been set (einstellung), and the student’s job is to solve them—not change or question them. A typical problem declares “all of its constraints, all of its given information, comprehensively and in advance,” as high-school teacher Dan Meyer explains. The students then take the prepackaged and preapproved problem and plug it into a formula they memorized, which, in turn, spits out the right answer.

This approach is wildly disconnected from reality. In our adult lives, problems often aren’t handed to us fully formed. We have to find, define, and redefine them ourselves. But once we find a problem, our educational conditioning kicks in, launching us into answer mode rather than asking whether there’s a better problem to solve. Although we pay lip service to the importance of finding the right problem, we double down on the same tactics that have failed us in the past.

Over time, we become a hammer, and every problem looks like a nail. In a survey of 106 senior executives spanning ninety-one corporations in seventeen countries, 85 percent agreed or strongly agreed that their businesses were bad at defining problems and that this weakness, in turn, imposed significant costs. A different study by management scholar Paul Nutt found that business failures happen in part because problems aren’t defined properly. For example, when businesses spot an advertising problem, they search for an advertising solution, artificially excluding all other possibilities. In the study, managers considered more than one alternative in less than 20 percent of their decisions. This environment is hostile to innovation. “Preconceived solutions and limited searches for options,” Nutt concluded, “are recipes for failure.”

Consider another study. Researchers divided expert chess players into two groups and gave them a chess problem to solve. The players were asked to achieve checkmate using the fewest possible moves. For the first group of players, the board contained two solutions: (1) a solution that was familiar to any skilled chess player and would achieve checkmate in five moves and (2) a less familiar, but better, solution that would produce a checkmate in three moves.

Many experts in the first group couldn’t find the better solution. The researchers tracked the players’ eye movements and found that they spent much of their time tracing the familiar solution on the board. Even when they claimed to be searching for alternatives, the experts literally couldn’t keep their eyes away from what they knew. When they could see the familiar solution—the hammer to their nail—their performance was effectively reduced by three standard deviations.

For the second group of players in the study, the researchers changed the setup so that the familiar solution was no longer an available option. Instead, only the optimal solution would achieve checkmate. Without the familiar solution to distract them, the experts in this second group all found the best solution. In the end, the study confirmed a statement attributed to several world chess champions: “When you see a good move, don’t make it immediately. Look for a better one.”

When the Einstellung effect gets in the way—when we can’t see the better move—we can change our definition of the problem by questioning the question.

Questioning the Question

Mark Adler defies all engineer stereotypes. He’s charming and charismatic, with a pair of sunglasses often hanging from his neck—a relic of his upbringing in sunny Florida. He laughs often but also has a strong undercurrent of intensity. In his spare time, he flies small airplanes and goes scuba diving. And he talks as fast as he thinks: My interview with him lasted for over an hour, and I squeezed in three questions at best.

When the Mars Polar Lander crashed in 1999, Adler was an engineer at NASA’s Jet Propulsion Laboratory. Recall that our Mars mission was canceled because we were planning to use the same three-legged landing system as the one on the Mars Polar Lander. At the time, everyone involved in our mission—except Adler—was suffering from the Einstellung effect. Like the expert chess players, we focused on the familiar solution on the chess board, which, in our case, was the three-legged lander.

But Adler came up with a better problem to solve. When I asked him about his thought process, he told me it was “really, really simple.” The way Adler saw it, our problem wasn’t the lander. It was gravity. We were preoccupied with the obvious question: “How do we design a better three-legged lander?” Adler stepped back and asked, “How do we defeat gravity and land our rover safely on Mars?” The same force that causes an apple to fall from a tree also causes unhappy meetings between a spacecraft and the Martian surface unless you do something to cushion the fall.

Adler’s solution was to abandon the three-legged lander design. Instead, he proposed using giant airbags with our rover cocooned inside a lander. These balloons would inflate shortly before impact with the Martian surface. Cushioned by these big white grapes, our robot geologist would be released from a height of about ten meters, strike the surface, and bounce roughly thirty to forty times before coming to rest.

Yes, the balloons were crude. Yes, they were ugly as hell. But they worked. Airbags had successfully landed the Pathfinder spacecraft on Mars in 1997. Adler knew “they could work because they worked before.”

Adler took his proposal to JPL’s Dan McCleese, chief scientist for Mars exploration, and asked why it wasn’t being considered. McCleese said, “It’s because there’s no champion for that.” So Adler decided to become its champion. He pitched the idea to some of the best people at JPL and got them on board. In less than four weeks—a record time for designing a mission—they put together a mission concept using Pathfinder’s landing system. The proposal eventually became reality. NASA selected Adler’s design largely because it had the highest probability of getting the spacecraft safely to Mars.

“Every answer,” Harvard Business School professor Clayton Christensen says, “has a question that retrieves it.” The answer is often embedded within the question itself, so the framing of the question becomes crucial to the solution. Charles Darwin would agree. “Looking back,” he wrote in a letter to a friend, “I think it was more difficult to see what the problems were than to solve them.”

Think of questions as different camera lenses. Put on a wide-angle lens, and you’ll capture the entire scene. Put on a zoom lens, and you’ll get a close-up shot of a butterfly. “What we observe is not nature itself, but nature exposed to our method of questioning,” said Werner Heisenberg, the brains behind the uncertainty principle in quantum mechanics. When we reframe a question—when we change our method of questioning—we have the power to change the answers.

Research supports this conclusion. A meta-analysis of fifty-five years of research on problem finding across numerous disciplines found a significant positive relationship between problem framing and creativity. In one famous study, Jacob Getzels and Mihaly Csikszentmihalyi found that the most creative art students spend more time in the preparation and discovery stage than do their less creative counterparts. Problem finding, according to these researchers, doesn’t end with the preparation stage. Even after spending time viewing the problem from different angles, the more creative individuals keep an open mind as they enter the solution stage and stand ready to make changes to their initial definition of the problem.

In our Mars mission, Adler was like the more creative art students, spending more time in formulating the problem and seeing a question that everyone else had missed. But what happened next even Adler couldn’t foresee.

The Doppelganger

In many ways, Mars is Earth’s sister planet. It’s next in line from the Sun. Its seasons as well as its rotational period and the tilt of its axis are similar to ours. Although Mars is cold and desolate now, it was warmer and wetter in the past, with evidence that liquid water flowed across its surface.

These characteristics make Mars one of the few places in the solar system where extraterrestrial life could have existed—even thrived. After the final Apollo mission to the Moon in 1972, Mars naturally appeared to be the next frontier. The Mariner probes, a series of spacecraft launched between 1962 and 1973, had already snapped photos of the red planet from orbit. It was time to get down to the surface. If NASA astronauts could do what Armstrong and Aldrin did—put on a spacesuit and head to Mars with hammers, scoops, and rakes to collect samples—they would have done it. But from NASA’s perspective, that option wasn’t financially feasible. So NASA did the next best thing: Instead of human geologists, it sent robotic ones.

NASA’s first attempt to land on Mars came in 1975 with the launch of the Viking mission. Named after the Nordic explorers, the mission sent two identical space probes to Mars, unimaginatively named Viking 1 and Viking 2. The probes each contained an orbiter designed to analyze the planet from Martian orbit and a lander to study its surface. After the spacecraft arrived at Mars, the orbiters spent some time scouting suitable landing sites. When the landing sites were spotted, the landers detached from the orbiters and descended toward the surface.

The Viking 1 lander touched down on July 20, 1976—seven years from the date of the Eagle’s touchdown on Tranquility Base—and was followed by Viking 2 in September of that year. Designed to last for ninety days, both landers significantly outlived their warranty. The Viking 1 lander conducted science for over six years and Viking 2 for nearly four, beaming tens of thousands of images back to Earth.

Some of these images dotted the entryway of Cornell’s Space Sciences Building, where I spent much of my time as an undergraduate. A giant smile would automatically appear on my face as I strolled past them each day before heading up to my fourth-floor work space in the Mars Room. If there was a montage of my college life, the Viking images would feature prominently.

Sometime in 2000, I was busy designing operations scenarios in the Mars Room, simulating what would happen after our rover landed on Mars. This was after Adler’s brilliant insight with the airbags had brought us back from the dead. I heard the distinctive sounds of Steve Squyres’s boots clicking toward my colleagues and me in the hallway. Squyres, my boss and the principal investigator of our mission, walked into the room and announced that he had just gotten off the phone with Scott Hubbard at NASA headquarters.

When it comes to creating worst-case scenarios, my imagination is particularly vivid. Pessimistic thoughts immediately began clanging in my head. What had gone wrong this time? Were we being scrapped again?

But the news wasn’t bad. Hubbard was in charge of fixing NASA’s Mars exploration program after the Mars Polar Lander accident. He had just left a meeting with NASA administrator Dan Goldin, who had asked Hubbard to relay a simple question to Squyres.

“Can you build two?” Hubbard had asked Squyres on the phone.

Squyres replied, “Two what?”

Hubbard responded, “Two payloads.”

Dumbstruck, Squyres asked, “Why would you want two payloads?”

“For two rovers,” Hubbard said.

It was a simple question no one had thought of asking before: Can we send two rovers instead of one? After the Mars Polar Lander crash, we had narrowly focused on the problem with our lander and replaced it with Adler’s airbag design. But the risk wasn’t isolated to the landing system. Any number of random things could break our spacecraft while traveling nearly forty million miles through outer space and landing on a Martian surface littered with scary-looking rocks while getting whipped by strong winds.

Goldin’s solution for this uncertainty was to use a strategy we encountered earlier in the book: introduce a redundancy. Instead of putting all our eggs in one spacecraft’s basket and crossing our fingers that nothing bad would happen along the way, we decided to send two rovers instead of one. Even if one failed, the other might make it. What’s more, with economies of scale, the cost of the second rover would be pennies on the dollar. After Goldin came up with the idea, Adler and another JPL engineer, Barry Goldstein, were given all of forty-five minutes to estimate how much the second rover would cost. They came up with $665 million for two rovers, which was roughly 50 percent more than the $440 million price tag for one. NASA managed to find the extra cash and gave us the green light.

Just like that, our rover birthed a doppelganger.

This time, NASA decided to get more creative with the naming and held a Name the Rovers contest, allowing schoolchildren to submit essays with suggestions. The winner among ten thousand submissions was Sofi Collins, a third-grader from Arizona, who was born in Siberia and had lived in an orphanage until her adoption by an American family: “It was dark and cold and lonely,” she wrote in her essay describing her orphanage. “At night, I looked up at the sparkly sky and felt better. I dreamed I could fly there. In America, I can make all my dreams come true. Thank you for the ‘Spirit’ and the ‘Opportunity.’”

The primary scientific goal of the newly christened Spirit and Opportunity rovers was to determine whether Mars had once been capable of supporting life. Because water is a crucial ingredient for life as we know it, we wanted to go where water had once gone before. Double the rovers also meant double the science. Two rovers could examine two very different landing sites. If one site turned out to be a flop from a science perspective, the other site might save the day.

For Opportunity, we chose Meridiani Planum, a plain near the Martian equator. The area appeared promising because its chemical composition—specifically, the existence of a mineral called hematite—suggested a history of liquid water. What’s more, Meridiani Planum is one of the “smoothest, flattest, and least windy places” on the red planet, the Martian equivalent of a giant parking lot. In terms of landing sites, it would be hard to find a safer one.

With Opportunity heading to a site rich in chemistry, we picked Gusev, a landing site rich in its topography, for Spirit. Located on the opposite side of the planet from Meridiani, Gusev is a giant impact crater with a visible channel. Scientists suspected that the channel had been carved by water at some point in the past and that the crater once held a lake. Gusev was slightly riskier from a landing perspective: It had higher winds and higher rock density than Meridiani had. But with two shots on goal, we could afford to take a bit more risk with one of them.

Spirit was the first to arrive at Mars. After the spacecraft touched the Martian atmosphere, things went as planned. The parachute was deployed. The heat shield was jettisoned. The airbags were inflated, to be followed by lots of bouncing and tumbling on the Martian surface until the lander came to rest. Any lingering doubts about whether Adler’s airbag design would work disappeared quickly as the first photos of Mars began to flow in. After years of looking at photos of Gusev taken from orbit, it was surreal to see, for the first time ever, the inside of the crater from the Martian surface in all its high-resolution glory.

But the initial thrill of the landing began to wane as our group started analyzing the images in detail. Yes, we were safely on Mars, and yes, the achievement put us in the distinct minority of missions that had successfully landed there. But aside from the fact that we were looking at Mars, what we were seeing was less than thrilling. The images from the rover looked a lot like those taken by the Viking landers and hung in the Space Sciences Building as decorations: similar rocks, similar outlook, similar structure—similar everything.

This initial scientific whimper would later turn into a bang when Spirit began roving the terrain and arrived at Columbia Hills, a range of peaks three kilometers away from our initial landing site. The peaks were named after the seven astronauts who perished in the Space Shuttle Columbia disaster one year before our landing. In those hills, Spirit would eventually go on to find goethite—a mineral that forms only in water, strongly indicating that Mars once had water activity above the surface.

Three weeks later, Spirit’s twin, Opportunity, touched down on Mars. Meridiani Planum, which was Opportunity’s landing site, was like nothing we had ever seen before. Every photo ever taken of the Martian surface has chunks of rock scattered across the surface. But where Opportunity landed, there were no rocks. When the rover began to beam its first photos of the landing area to Earth, the mission support team at JPL began laughing, cheering, and crying. The flight director, Chris Lewicki, asked Squyres for a quick science overview of what they were seeing on the screen. But Squyres’s throat constricted. He slowly flipped the switch on his headset and said, “Holy smokes. I’m sorry, I’m just, I’m blown away by this.”

What they were seeing was an outcrop of bedrock right in front of the rover. Why would something as benign as bedrock leave a scientist speechless? An exposed, layered bedrock is the closest thing there is to time travel. A bedrock is like a history book. It shows us exactly what happened a long, long time ago, on this planet far, far away. Unlike Spirit—which had to climb a mountain, literally and figuratively, to find interesting science—Opportunity was handed scientific secrets on a silver platter, or in this case, on bedrock. Important early discoveries came within the first six weeks of the mission, thanks to Opportunity’s landing site—which was made possible by our decision to send two rovers.

Squyres didn’t realize it then, but his comments—including the “holy smokes” bit—were broadcast across the globe. They piqued the interest of a journalist in Seoul, South Korea, writing for the daily newspaper Munhwa Ilbo. The journalist wrote up the story of Opportunity’s historic Mars landing, summed up by the following headline: “The Second Mars Rover Lands, Sees Mysterious Smoke.” As another Korean journalist observed, it was fortunate Squyres didn’t say holy cow.

Like their Viking grandfathers, our rovers were designed to operate for ninety days. Both outlasted their planned missions, and Opportunity far outlived the Viking landers. Spirit lasted for more than six years until it got stuck on soft soil. It eventually lost communication with Earth after winter arrived and deprived its solar panels of their power source. A formal farewell was held for Spirit—complete with toasts and spirited eulogies for a rover that climbed mountains (which it wasn’t designed to do) and braved intense dust storms.

Opportunity—or Oppy, as we lovingly called it—kept going until June 2018, when a giant dust storm covered the rover’s solar panels, starving it of power. NASA officials sent hundreds of commands asking Oppy to call home, with no success. In February 2019, Opportunity was officially pronounced dead—over fourteen years into its ninety-day expected lifetime, having roamed a record-breaking twenty-eight miles on the red planet.

Holy smokes indeed.

In the end, two questions that reframed the problems ended up producing one of the most successful interplanetary missions of all time: What if we used airbags instead of a three-legged lander? What if we sent two rovers instead of one?

These questions may appear obvious, but they were obvious only in hindsight. How do you do what Adler and Goldin did and see the problem from a perspective others miss? One approach is to distinguish between two concepts—strategy and tactics—that are often conflated. To understand the distinction, let’s say goodbye to Mars (for now) and head over to Nepal.

Strategy and Tactics

Babies who are born too soon—before certain key organs develop—are called premature babies (or preemies). The account cites roughly one million premature-infant deaths worldwide each year, with hypothermia among the risks facing these babies. Because these babies are born with very little body fat, it is difficult for them to control their body temperature. To them, room temperature can feel like freezing cold water.

In developed countries, the solution is to place the baby in an incubator. The size of a standard crib, an incubator keeps babies warm while their bodies finish developing. The original incubators were fairly simple devices, but bells and whistles were added over time. Incubators now have arm ports for handling the baby, life support devices such as ventilators, and equipment to regulate humidity. The boost in tech also brought a boost in cost. A modern incubator is priced between $20,000 and $40,000, a price that doesn’t include the electricity required to make it work. Incubators are consequently hard to find in many developing countries, and the result is preventable deaths.

Four Stanford University graduate students set out to tackle this challenge in 2008 and build cheaper incubators. They were enrolled in a course called Design for Extreme Affordability, where students “learn to design products and services that will change the lives of the world’s poorest citizens.”

Instead of trying to innovate from the comfort of Silicon Valley, the team decided to take a field trip to Kathmandu, Nepal’s capital, to immerse themselves in the practices at a neonatal unit. They wanted to observe how incubators were being used in hospitals so that they could design cheaper equipment to work in local conditions.

But a surprise awaited them. The incubators in hospitals were collecting dust and sitting unused. Part of the problem was technical expertise. Incubators are often difficult to operate. What’s more, the overwhelming majority of premature babies in Nepal were born in rural areas. Most of these babies would never make it to a hospital in the first place.

The problem, therefore, wasn’t the lack of incubators in hospitals. Rather, it was the lack of accessible infant warmers in rural areas with no access to hospitals or, for that matter, reliable electricity. The traditional solution—send more incubators to hospitals or lower their cost—wouldn’t move the needle.

In light of this experience, the Stanford team reframed the problem. The team focused on providing warmth where incubators were inaccessible. That did not make other neonatal care unnecessary; it identified a particular need their device could address. The device to provide warmth had to be inexpensive and intuitive so it could be used by an often-illiterate parent in a rural environment without reliable electricity.

The result was the Embrace infant warmer. It’s a small, light sleeping bag that wraps around the infant. A pouch of phase-change material—which is an innovative wax—keeps the baby at the right temperature for up to four hours. You can “recharge” the warmer in only a few minutes by putting it in boiling water. And compared with the $20,000 to $40,000 price tag of a traditional incubator, the account gives a cost of $25 for Embrace. By 2019, the product had reached hundreds of thousands of premature infants in over twenty countries.

Often, we fall in love with our favorite solution and then define the problem as the absence of that solution. “The problem is, we need a better three-legged lander.” “The problem is, we don’t have enough incubators.” In each case, we pursue technology for the sake of technology. We lose the forest for the trees, the purpose for the method, the function for the form.

This approach mistakes tactics for strategy. Although the terms are often used interchangeably, they refer to different concepts. A strategy is a plan for achieving an objective. Tactics, in contrast, are the actions you take to implement the strategy.

We often lose sight of the strategy, fixate on the tactics and the tools, and become dependent on them. But tools, as author Neil Gaiman reminds us, “can be the subtlest of traps.” Just because a hammer is sitting in front of you doesn’t mean it’s the right tool for the job. Only when you zoom out and determine the broader strategy can you walk away from a flawed tactic.

To find the strategy, ask yourself, What problem is this tactic here to solve? This question requires abandoning the what and the how and focusing on the why. The three-legged lander was a tactic, and landing safely on Mars was the strategy. The incubator was a tactic, and saving premature infants was the strategy. If you’re having trouble zooming out, bring outsiders into the conversation. People who don’t regularly use hammers are less likely to be distracted by the hammer sitting in front of you.

Once you identify the strategy, it becomes easier to play with different tactics. If you frame the problem more broadly as one of gravity—not as a flawed three-legged lander—airbags can present a better alternative. If you frame the problem more broadly as the risk involved in landing on Mars—not just as a defective lander—sending two rovers instead of one decreases risk and increases reward.

Peter Attia, a physician who writes about longevity, is a master at distinguishing between strategy and tactics. I asked him what he does when patients come to him looking for the “right answers.” What diet should I follow? Should I take statins if I have high cholesterol? “I generally do not let patients fixate on tactics,” he told me, “and instead try to refocus them on strategy. When people are looking for the ‘right answers,’ they are almost always asking tactical questions. By focusing on the strategy, this allows you to be much more malleable with the tactics.” For Attia, whether to use a statin is “a tactical question that is in service of the much broader strategy” of delaying death from atherosclerosis.

To teach the difference between strategy and tactics to her students, Tina Seelig, then faculty director of the Stanford Technology Ventures Program, uses what she calls the five-dollar challenge. Students break up into teams, and each team gets five dollars in funding. Their goal is to make as much money as possible within two hours and then give a three-minute presentation to the class about what they achieved.

If you were a student in the class, what would you do?

Typical answers include using the five dollars to buy start-up materials for a makeshift car wash or lemonade stand and buying a lottery ticket. But the teams that follow these typical paths tend to bring up the rear in the class.

The teams that make the most money don’t use the five dollars at all. They realize that the five dollars is a distracting, and essentially worthless, resource.

So they ignore it. Instead, they reframe the problem more broadly as “What can we do to make money if we start with absolutely nothing?” One particularly successful team made reservations at popular local restaurants and then sold the reservation times to those who wanted to skip the wait. These students generated an impressive few hundred dollars in just two hours.

But the team that made the most money approached the problem differently. The students understood that both the five-dollar funding and the two-hour period weren’t the most valuable assets at their disposal. Rather, the most valuable resource was the three-minute presentation time they had in front of a captive Stanford class. They sold their three-minute slot to a company interested in recruiting Stanford students and walked away with $650.

What is the five-dollar tactic in your own life? How can you ignore it and find the two-hour window? Or even better, how do you find the most valuable three minutes in your arsenal? Once you move from the what to the why—once you frame the problem broadly in terms of what you’re trying to do instead of your favored solution—you’ll discover other possibilities in the peripheries.

Just as you can reframe questions to generate better answers, you can also reframe objects, products, skills, and other resources to put them to more creative uses. That requires thinking outside the box—in this case, the thumbtack box.

Thinking Outside the Thumbtack Box

What is a barometer for?

If you think the only answer is measuring pressure, think again.

Science professor Alexander Calandra—who was an advocate of unorthodox teaching methods—once penned a short story titled “Angels on a Pin.” In the story, a colleague asks Calandra to be the arbiter of a dispute between the colleague and a student over a question on a physics exam. The physics teacher believes the student deserves a zero, but the student demands full credit.

Here’s the question: “Show how it is possible to determine the height of a tall building with the aid of a barometer.” The traditional answer is clear: You take pressure measurements with the barometer at the top of the building and at the bottom and use the difference to compute the height.

But this wasn’t the answer the student gave. Instead, the student wrote: “Take the barometer to the top of the building, attach a long rope to it, lower the barometer to the street, and then bring it up, measuring the length of the rope. The length of the rope is the height of the building.”

The answer is certainly correct. But it’s a deviation from the norm. It’s not what the teacher had taught in class—the expected path to the expected outcome. A barometer is supposed to measure pressure, not serve as a makeshift weight for a rope.

The barometer story is a good example of functional fixedness. As psychologist Karl Duncker explains, the concept refers to a “mental block against using an object in a new way that is required to solve a problem.” Just as we treat problems and questions as fixed, we do the same with tools. Once we learn that a barometer measures pressure, we blind ourselves to other uses for it. Like the chess players whose eyes kept darting to the familiar solution on the chess board, our minds fixate on the function we know.

Perhaps the most famous example of functional fixedness is the candle problem, designed by Duncker. He devised an experiment where he seated participants at a table adjacent to a wall and gave them a candle, some matches, and a box of thumbtacks. He asked them to figure out a way to attach the candle to the wall so the wax doesn’t drip on the table below. Most people tried one of two approaches. They attempted to use thumbtacks to pin the candle to the wall or melt the side of the candle with a match to stick it to the wall.

But neither approach works. These participants failed in part because they focused on the traditional functions of the objects: Thumbtacks are for attaching things. Boxes are for storing things.

The successful participants disregarded the traditional function of the box. Instead, they reframed the box as a platform for the candle to stand on. They then affixed the box to the wall using thumbtacks.

We all encounter variations of the candle problem in our personal and professional lives. And often, we do what the unsuccessful participants did and view the box as a container, not as a platform. So how do we train ourselves to think outside the thumbtack box? How do we see the products or the services we provide from a different perspective? How do we take the skills we have in one field and recognize their value in another?

In a study conducted for the military, Robert Adamson attempted to answer these questions. He replicated Duncker’s candle experiment but with a twist: He divided the participants into two groups and slightly modified the setup for each. The second group far outperformed the first. Only 41 percent of the participants in the first group solved the puzzle, compared with 86 percent in the second.

What explained the stark difference in the outcome? For the first group, the three types of materials—the candle, the matches, and the thumbtacks—were placed in three boxes. The first group saw boxes being used as containers and, as a result, suffered from acute functional fixedness. They had a much harder time using the box for anything other than storing objects.

But for the second group, the objects were sitting on the table next to—not inside—the boxes, which were empty. With the objects out of the boxes, the participants could more easily see the boxes as potential stands for the candle. The results were similar to the conclusion of the study involving chess experts. In both cases, performance improved when the familiar solution was removed.

Functional fixedness arises from a set of assumptions we have about what a box or a barometer is supposed to do. We can reduce functional fixedness by taking out Occam’s razor—which we explored earlier in the book—and cutting our assumptions about the tool. If you didn’t know what you know, what else could you do with it? This can be as simple as blocking its obvious use—dumping the materials out of the box (as Adamson’s study did), removing the familiar solution from the chess board, or using the barometer for anything other than measuring pressure.

Combinatory play also helps. You can draw inspiration from how objects are used in other fields. For example, the airbags that landed my group’s rovers safely on Mars used the same mechanism that cushions a collision with your steering wheel in a car accident. The same fabric used in astronaut spacesuits is also used by Embrace to make a temperature-controlling swaddle. George de Mestral created Velcro after he saw his pants covered in cockleburs following a walk. He examined the cockleburs under a microscope and discovered a hooklike shape that he then emulated to create the hook-and-loop fastener called Velcro—with one side stiff like the cockleburs and the other side smooth like his pants.

It’s also helpful to separate function from form. When we look at an object, we tend to see its function. A barometer, we think, is for measuring pressure. A hammer is for driving in nails. A box is for storing objects. But this natural inertia toward the function also gets in the way of innovation. If we can look past the function to the form, we can discover other ways that the product, service, or technology can be used. For example, if you can view the typical barometer simply as a round object, it can also be used as a weight. If you view a box as a flat platform with sides, it can also be used as a stand.

In one study, participants were divided into two groups and asked to solve eight insight problems—including the candle problem—that required them to overcome functional fixedness. The control group received no training. The other group was taught to use function-free descriptions of objects—for example, instead of saying “a prong of an electrical plug,” they were taught to describe the prong as “a thin, rectangular piece of metal.” The group that received training solved 67 percent more problems than did the other participants.

The switch from function to form is also helpful in reframing the resources at your disposal. Consider, for example, the development of Amazon Web Services (AWS). As Amazon grew from an online bookstore to an “everything” store, it built up an immense electronic infrastructure, including storage and databases. The company realized that its infrastructure wasn’t simply an internal resource. It could also be sold to other companies as a cloud-computing service, to be used for storage, networking, and databases. AWS became a major source of operating profit for Amazon, generating roughly $17 billion in revenue in 2017. Its operating profit exceeded that of Amazon’s retail business; its revenue did not.

Amazon reframed the thumbtack box again with its purchase of Whole Foods Market. The purchase befuddled many observers. Why was the internet giant buying up a struggling brick-and-mortar grocery store chain? One answer was based on the reframing of the physical Whole Foods Market stores. Instead of seeing them simply as grocery stores, Amazon reframed them as distribution centers located in densely populated urban hubs. These centers could enable fast delivery of products to Amazon Prime customers.

In both cases, Amazon looked beyond the function to the form. The function of Whole Foods stores was to sell groceries, but the stores took the form of a massive real estate footprint with storage and refrigeration that could be repurposed for distribution. The function of Amazon’s computing infrastructure was for internal support, but its form—a massive data center—could provide a highly profitable service to companies such as Netflix and Airbnb.

If you’re having a hard time switching from function to form and seeing the thumbtack box as a candle platform, there’s another approach you can try: Reverse the box.

What If We Did the Reverse?

On Friday, October 4, 1957, the Soviet Union launched Sputnik, the first artificial satellite in Earth orbit. Russian for “fellow traveler,” Sputnik orbited the Earth roughly every ninety-eight minutes. If you doubted that humankind had created its own moon, you could walk outside with a pair of binoculars after sunset and see it flying overhead.

You could not only see Sputnik, but hear it as well. At the time, two young physicists by the name of William Guier and George Weiffenbach were working in the Johns Hopkins Applied Physics Laboratory in Maryland. They were curious whether the radio signals emitted by Sputnik could be received on Earth. In a matter of hours, Guier and Weiffenbach had locked in on a series of signals emanating from the satellite.

Beep. Beep. Beep.

This easily detectable signature was no oversight on the Soviets’ part. Masters at propaganda, the Soviets had intentionally engineered the Sputnik to broadcast a signal that could easily be picked up by anyone on Earth with a shortwave radio.

Beep. Beep. Beep.

As Guier and Weiffenbach listened to the Red broadcast, they realized they could use the signal to calculate Sputnik’s speed and trajectory. Just as the siren of an ambulance buzzing past you decreases in pitch, the beeps from Sputnik changed as the satellite moved away from the scientists’ location. Using this phenomenon—called the Doppler effect—the two men plotted the entire trajectory of Sputnik.

Sputnik’s launch inspired awe but also whipped Americans into a frenzy. “If the Russians can deliver a 184-pound ‘Moon’ into a predetermined pattern 560 miles out in space,” wrote an editorial in the Chicago Daily News, “the day is not far distant when they could deliver a death-dealing warhead onto a predetermined target almost anywhere on the Earth’s surface.”

Frank McClure was also thunderstruck by Sputnik, but for a different reason. McClure was then the deputy director at the Applied Physics Laboratory. He called Guier and Weiffenbach into his office and asked them a simple question: “Can you guys do the reverse?” If the two men could calculate the unknown trajectory of a satellite from a known location on Earth, could they find an unknown location on Earth using the known location of a satellite?

This question may sound like a theoretical riddle, but McClure had a very practical application in mind. At the time, the military was developing nuclear missiles capable of being launched from submarines. But there was a problem. To strike a precise location with a nuclear missile, the military had to know the precise location of the launch site. In the case of nuclear submarines swimming through the Pacific Ocean, their precise location was unknown. Hence the question: Can you discover the unknown location of our submarines through the known location of a satellite we’ll launch into space?

The answer was a resounding yes. Within three years of Sputnik’s launch, the United States began launching satellites for Transit, a navigation system designed around a constellation of five operational satellites. Transit preceded the global positioning system, or GPS, whose development continued through the 1980s. GPS was a distinct successor system, not a new name for Transit.

McClure’s approach illustrates a powerful way of reframing questions: taking an idea and flipping it on its head. This method dates back at least to the nineteenth century, when the German mathematician Carl Jacobi introduced the idea with a powerful maxim: “Invert, always invert” (Man muss immer umkehren).

Michael Faraday applied this principle to generate one of the greatest scientific discoveries of all time. In 1820, Hans Christian Ørsted—who coined the term thought experiment—discovered the connection between electricity and magnetism. He noticed that a compass needle deflected when a wire carrying an electric current passed over it.

Later, Faraday came along and reversed Ørsted’s experiment. Instead of passing a wire with electric current over a magnet, he passed a magnet around a coil of wire. This generated an electrical current that grew bigger the faster he spun the magnet. Faraday’s work on electromagnetic induction underlies the generators in modern hydroelectric and nuclear power plants. A turbine supplies mechanical rotation, and relative motion between a magnetic field and conducting coils generates electricity.

Across disciplines, in biology, Darwin adopted the same reversal mantra. While other field biologists looked for differences between species, Darwin searched for similarities. He compared, for example, the wing of a bird with the hand of a human. Exploring the similarities between otherwise vastly different species eventually culminated in the theory of evolution.

The power of inversion extends far beyond science. To cite a business example, the clothing company Patagonia reversed an industry best practice in a 2011 advertising campaign. The company asked, “Instead of telling people to buy from us, what if we told them not to buy from us?” The result was a full-page ad in the New York Times that ran on Black Friday—the Friday after US Thanksgiving, when Americans flock to stores to take advantage of deep discounts for the holiday shopping season. The ad featured a Patagonia jacket with the headline “Don’t buy this jacket.” With this ad, Patagonia became “the only retailer in the country asking people to buy less on Black Friday.” The ad worked, partly because it supported Patagonia’s mission of reducing consumerism and lowering environmental impact. But the contrarian ad also ended up helping the company’s bottom line by attracting customers who shared the same mindset.

In athletics, reversing conventional wisdom landed Dick Fosbury the Olympic gold medal. At the time, if you had met Fosbury in person, you wouldn’t have thought he was an athlete. He was awkward, scrawny, and tall, with a significant acne problem that he couldn’t seem to shake. When Fosbury was training to be a high jumper, athletes would use a technique called the straddle method, where they would jump face down over the bar. At the time, the straddle method was considered beyond improvement. There was no need to experiment or come up with something new.

But the straddle method never worked for Fosbury. As a high-school sophomore, he was performing at a junior high level. On a bus ride to a track meet, Fosbury decided to do something about his mediocrity. The rules allowed athletes to clear the bar any way they wanted as long as they jumped off one foot. The straddle method was a mere tactic. But clearing the bar was the strategy. So instead of jumping face down to the bar, Fosbury did the reverse. He jumped backward.

His approach initially invited ridicule. A newspaper called him “The World’s Laziest High Jumper.” Many fans laughed at him as he cleared the bar like a fish flopping in a boat.

The laughs eventually turned into cheers as Fosbury proved his critics wrong and took home the gold medal at the 1968 Summer Olympics—by doing the exact opposite of what everyone else was doing. The Fosbury flop, as it has come to be known, is now the standard method used at Olympic high jump events. Fosbury came home to a ticker-tape parade and appeared live on the Tonight Show, where he taught Johnny Carson how to perform the Fosbury flop.

The serial entrepreneur Rod Drury calls this approach the “George Costanza theory of management.” In an episode of Seinfeld, Costanza sets out to improve his life by doing the exact opposite of what he had done before. Drury, who founded and led the accounting software company Xero, would outsmart his far bigger competitors by asking himself, “What is the exact opposite of what an incumbent would expect us to do?” Asking this question in 2005, Drury went all in on using a cloud-based platform when his competitors were all still stuck on desktop applications.

Drury knows a secret missed by many business leaders: The low-hanging fruit has already been picked. Copying a stronger competitor can leave you competing on their terms. Considering the opposite approach may reveal an opening, though reversal alone does not make a strategy sound.

Instead of adopting a common best practice or the industry standard, reframe the question by asking, “What if I did the reverse?” Even if you don’t execute, the simple process of thinking through the opposite will make you question your assumptions and jolt you out of your current perspective.

The next time you’re tempted to engage in problem solving, try problem finding instead. Ask yourself, Am I asking the right question? If I changed my perspective, how would the problem change? How can I frame the question in terms of strategy, instead of tactics? How do I flip the thumbtack box and view this resource in terms of its form, not its function? What if we did the reverse?

Breakthroughs, contrary to popular wisdom, don’t begin with a smart answer.

They begin with a smart question.