by Matthew Syed
Kobayashi had eaten more than any competitor in history not because he had a surgically enlarged stomach or an extra esophagus (as some competitors alleged); rather, he triumphed via the aggregation of marginal gains. By failing in all sorts of small, well-measured, rigorously tested ways, he iterated his way to success. It was bottom-up rather than top-down, if you’ll forgive the expression.
And if this approach can be applied to eating salty tubes of sandwich meat, it can be applied to almost anything.
VI
To conclude this chapter, let’s examine the concept of marginal gains in visual form. The process of optimization can be compared to trying to get to the top of a summit. Suppose you start from a position below the summit of the smaller of two hills, Point A, and take a tiny step in a particular direction. You then test to see if you have gone up and, if you have, you take another small step, and test again.
In this way, by taking lots of small steps, each rigorously examined to see if it is taking you in the right direction, you will eventually end up at the smaller summit. Indeed, this method is so powerful that it will work even if you are wearing a blindfold, as the business expert Eric Ries has written in an excellent essay on the art of optimization.13
This is the potency of marginal gains. By dividing a big challenge into small parts, you are able to create rigorous tests, and thus deliver incremental improvements. Each may seem small or, as Brailsford often says, “virtually negligible,” but over time, and with discipline, they accumulate. You eventually reach the optimum point, the summit of the smaller hill. This is the Local Maximum.14 It is often the difference between winning and losing, whether in sports, business, or speed-eating hot dogs.
But this visualization also reveals the inherent limitations of marginal gains. Often in business, technology, and life, progress is not about small, well-delivered steps, but creative leaps. It is about acts of imagination that can transform the entire landscape of a problem. Indeed, these are sometimes the most important drivers of change in the modern world.
To see this difference, take Blockbuster. This was a business based around the renting of videos and later DVDs. As a concept it fared well for more than two decades, delivering an impressive rate of return. You can imagine a manager at the company using a marginal gains approach: altering the company’s logo, tweaking the design of the shelving at the stores, trialing different discount approaches like two-for-one, and so on.
Each of these tests would have been useful. Over time they would have accumulated, taking the company toward the top of the local optimization summit. But the problem is also obvious: the business model was eventually superseded by Netflix and the like, rendering videos and DVDs, to a large extent, obsolete.* The entire landscape fundamentally changed. And no amount of marginal gains (at least within a realistic time frame) would have helped Blockbuster to survive. The company was liquidated in 2013.*
In the diagram, the new landscape is represented by the taller hill. Marginal gains is a strategy of local optimization: it takes you to the summit of the first hill. But once you are there, taking little steps, however well tested, runs out of traction. To have stayed ahead of the competition, Blockbuster would have needed to move into an entirely new space, leveraging new technology and fresh insights.
There is an ongoing debate in the political, scientific, and business worlds about whether to focus on the bold leaps that lead to new conceptual terrain, or on the marginal gains that help to optimize one’s existing fundamental assumptions. Is it about testing small assumptions or big ones; is it about transforming the world or tweaking it; is it about considering the big picture (the so-called gestalt) or the fine detail (the margins)?
The simple answer, however, is that it has to be both. At the level of the system and, increasingly, at the level of the organization, success is about developing the capacity to think big and small, to be both imaginative and disciplined, to immerse oneself in the minutiae of a problem and to stand beyond it in order to glimpse the wider vista.
In this chapter we have looked at small steps and found that they are driven by discovering little failures. Marginal gains, as a philosophy, absolutely depends on the ability to detect and learn from small, often latent weaknesses. Now we are going to look at giant leaps, the audacious changes in technology, design, and science that transform our world.
And we will see that beneath the inspirational stories told about these shifts, the deepest and most overlooked truth is that innovation cannot happen without failure. Indeed, the aversion to failure is the single largest obstacle to creative change, not just in business but beyond.
Chapter 10
How Failure Drives Innovation
I
The headquarters of Dyson are in a futuristic building about forty miles west of Oxford. Outside the front entrance is a Harrier jump jet—not a replica, a real one—and a high-speed landing craft. They both hint at the unconventionality of what goes on inside.
James Dyson, the chairman and chief engineer of the company, works in a glass-fronted office just above the entrance. Along the back wall are the beautifully conceived products that have turned him into an icon of British innovation: super-efficient vacuum cleaners, futuristic hand dryers, and other devices yet to roll off the production line. In all, he has applied for more than four thousand patents.1
Progress is often driven not by the accumulation of small steps, but by dramatic leaps. The television wasn’t an iteration of a previous device, it was a new technology altogether. Einstein’s general theory of relativity didn’t tinker with Newton’s law of universal gravitation, it replaced it in almost every detail. Likewise Dyson’s dual-cyclone vacuum cleaner was not a marginal improvement on the conventional Hoover that existed at the time, it represented a shift that altered the way insiders think about the very problem of removing dust and hair from household floors.
Dyson is an evangelist for the creative process of change, not least because he believes it is fundamentally misconceived in the world today. As we talk in his office, he darts around picking up papers, patents, textbooks, and his own designs to illustrate his argument. He is tall, bright-eyed, and restless. A conversation scheduled for half an hour continues late into the evening, so that by the end the sun has gone down, and his expressive face is lit only by a table lamp (designed, incidentally, by his son: it contains an LED light that lasts for 160,000 hours rather than the usual 2,000).
He says:
People think of creativity as a mystical process. The idea is that creative insights emerge from the ether, through pure contemplation. This model conceives of innovation as something that happens to people, normally geniuses. But this could not be more wrong. Creativity is something that has to be worked at, and it has specific characteristics. Unless we understand how it happens, we will not improve our creativity, as a society or as a world.
Dyson’s journey into the nature of creativity started while vacuuming his own home, a small farmhouse in the west of England, on a Saturday morning in his mid-twenties. Like everyone else he was struck by just how quickly his cleaner lost suction. “It was a top-of-the-range Hoover,” he says. “It had one of the most powerful vacuum motors in the world. But it lost its suction within minutes. It started to let out this high-pitched scream. I had faced the problem before. Growing up, it had been my chore to vacuum the family home and the suction was a constant bugbear. But this time I just snapped.”
Dyson strode into his garden and opened up the device. Inside he could see the basic engineering proposition of the conventional vacuum cleaner: a motor, a bag (which also doubled as a filter), and a tube. The logic was simple: dust and air is sucked into the bag, the air escapes through the small holes in the lining of the bag and into the motor, and the dust (thicker than the air) stays in the bag. He says:
The bag was full of dust and so I assumed this was the reason that it had lost suction. So I ripped open the b
ag, emptied out the dust and Sellotaped it back up again. But when I went back to vacuum in the house, the efficiency was no better. The screaming started straight away. There was no suction.
I suddenly realized that the real problem was not that the bag was full; it was the thin lining of dust on the inside of the bag. The walls of the bag were clogged. The fine dust was blocking the filter. And that is why performance in conventional vacuum cleaners dips so rapidly; it is the very first dust that blocks them up.
This realization triggered a new thought: What if there were no bag? What if you could make an entirely bagless vacuum cleaner? “If you could find a way of removing the dust from the air another way, without using a conventional bag, you would no longer lose suction because of a blocked filter,” he says. “It would revolutionize vacuum cleaning.”
This idea percolated in Dyson’s mind for the next three years. A graduate of the Royal College of Art, he was already a qualified engineer and was helping to run a local company in Bath. He enjoyed pulling things apart and seeing how they worked. He was curious, inquisitive, and willing to engage with a difficulty rather than just accepting it. But now he had a live problem, one that intrigued him.
It wasn’t until he went to a lumberyard that the solution powered into his mind like a thunderbolt.
Nowadays you pick up wood from a merchant and just walk out. In the old days, they virtually had to cut and plane it for you. There was a lot of hanging about. As I stood there waiting I noticed this ducting going off the machines. It traveled along to this thing on the roof, thirty or forty foot tall.
It was a cyclone [a cone-shaped device that changes the dynamics of the airflow, separating the dust from the air via centrifugal force]. It was made of galvanized steel. And although a ton of dust was coming off the machines as they cut the wood, there was no dust coming out of the chimney at the top. I was intrigued. This thing was collecting fine dust all day long and it didn’t look as though it was blocking at all.
Dyson rushed home. This was his moment of insight. “I vaguely knew about cyclones, but not really the detail. But I was fascinated to see if it would work in miniature form. I got an old cardboard box and made a replica of what I had seen with gaffer tape and cardboard. I then connected it via a bit of hose to an upright vacuum cleaner. And I had my cardboard cyclone.”
His heart was beating fast as he pushed it around the house. Would it work? “It seemed absolutely fine,” he says. “It seemed to be picking up dust, but the dust didn’t seem to be coming out of the chimney. I went to my boss and said: ‘I think I have an interesting idea.’”
This simple idea, this moment of insight, would ultimately make Dyson a personal fortune in excess of £3 billion.
II
A number of things jump out about the Dyson story. The first is that the solution seems rather obvious in hindsight. This is often the case with innovation, and it’s something we will come back to.
But now consider a couple of other aspects of the story. The first is that the creative process started with a problem, what you might even call a failure, in the existing technology. The vacuum cleaner kept blocking. It let out a screaming noise. Dyson had to keep bending down to pick up bits of trash by hand.
Had everything been going smoothly Dyson would have had no motivation to change things. Moreover, he would have had no intellectual challenge to sink his teeth into. It was the very nature of the engineering problem that sparked a possible solution (a bagless vacuum cleaner).
And this turns out to be an almost perfect metaphor for the creative process, whether it involves vacuum cleaners, a quest for a new brand name, or a new scientific theory. Creativity is, in many respects, a response.
Relativity was a response to the failure of Newtonian mechanics to make accurate predictions when objects were moving at fast speeds.
Masking tape was a response to the failure of existing adhesive tape, which would rip the paint off when it was removed from cars and walls.
The collapsible stroller was a response to the impracticality of unwieldy baby carriages (Owen Maclaren, the designer, came up with the idea after watching his daughter struggling with a baby carriage while out with his granddaughter).
The wind-up radio was a response to the lack of batteries in Africa, something that was hampering the spread of educational information (Trevor Baylis came up with the idea after watching a television program on AIDS).
The ATM was a response to the problem of getting hold of cash outside of business hours. It was invented by John Shepherd-Barron while lying in the bath one night, worrying because he had forgotten to go to the bank.
Dropbox, as we have seen, was a response to the problem of forgetting your flash drive and thus not having access to important files.
This aspect of the creative process, the fact that it emerges in response to a particular difficulty, has spawned its own terminology. It is called the “problem phase” of innovation. “The damn thing had been bugging me for years,” Dyson says of the conventional vacuum cleaner. “I couldn’t bear the inefficiency of the technology. It wasn’t so much a ‘problem phase’ as a ‘hatred phase.’”
We often leave this aspect of the creative process out of the picture. We focus on the moment of epiphany, the detonation of insight that happened when Newton was hit by the apple or Archimedes was taking a bath. That is perhaps why creativity seems so ethereal. The idea is that such insights could happen anytime, anywhere. It is just a matter of sitting back and letting them flow.
But this leaves out an indispensable feature of creativity. Without a problem, without a failure, without a flaw, without a frustration, innovation has nothing to latch on to. It loses its pivot. As Dyson puts it: “Creativity should be thought of as a dialogue. You have to have a problem before you can have the game-changing riposte.”
Perhaps the most graphic way to glimpse the responsive nature of creativity is to consider an experiment by Charlan Nemeth, a psychologist at the University of California, Berkeley, and her colleagues.2 She took 265 female undergraduates and randomly divided them into five-person teams. Each team was given the same task: to come up with ideas about how to reduce traffic congestion in the San Francisco Bay Area. These five-person teams were then assigned to one of three ways of working.
The first group were given the instruction to brainstorm. This is one of the most influential creativity techniques in history, and it is based on the mystical conception of how creativity happens: through contemplation and the free flow of ideas. In brainstorming the entire approach is to remove obstacles. It is to minimize challenges. People are warned not to criticize each other, or point out the difficulties in each other’s suggestions. Blockages are bad. Negative feedback is a sin.
As Alex Faickney Osborn, an advertising executive who wrote a series of best-selling books on brainstorming in the 1940s and 1950s, put it: “Creativity is so delicate a flower that praise tends to make it bloom, while discouragement often nips it in the bud.”3
The second group were given no guidelines at all: they were allowed to come up with ideas in any way they thought best.
But the third group were actively encouraged to point out the flaws in each other’s ideas. Their instructions read: “Most research and advice suggests that the best way to come up with good solutions is to come up with many solutions. Free-wheeling is welcome; don’t be afraid to say anything that comes to mind. However, in addition, most studies suggest that you should debate and even criticize each other’s ideas [my italics].”
The results were remarkable. The groups with the dissent and criticize guidelines generated 25 percent more ideas than those who were brainstorming (or who had no instructions). Just as striking, when individuals were later asked to come up with more solutions for the traffic problem, those with the dissent guidelines generated twice as many new ideas as the brainstormers.
Further studies have shown that tho
se who dissent rather than brainstorm produce not just more ideas, but more productive and imaginative ideas. As Nemeth put it: “The basic finding is that the encouragement of debate—and even criticism if warranted—appears to stimulate more creative ideas. And cultures that permit and even encourage such expression of differing viewpoints may stimulate the most innovation.”
The reason is not difficult to identify. The problem with brainstorming is not its insistence on free-wheeling or quick association. Rather, it is that when these ideas are not checked by the feedback of criticism, they have nothing to respond to. Criticism surfaces problems. It brings difficulties to light. This forces us to think afresh. When our assumptions are violated we are nudged into a new relationship with reality. Removing failure from innovation is like removing oxygen from a fire.
Think back to Dyson and his Hoover. It was the flaw in the existing technology that forced Dyson to think about cleaning in a new way. The blockage in the filter wasn’t something to hide away from or pretend wasn’t there. Rather, the blockage, the failure, was a gilt-edged invitation to reimagine vacuum-cleaning.
Imagination is not fragile. It feeds off flaws, difficulties, and problems. Insulating ourselves from failures—whether via brainstorming guidelines, the familiar cultural taboo on criticism, or the influence of cognitive dissonance*—is to rob one of our most valuable mental faculties of fuel.
“It always starts with a problem,” Dyson says. “I hated vacuum cleaners for twenty years, but I hated hand dryers for even longer. If they had worked perfectly, I would have had no motivation to come up with a new solution. But more important, I would not have had the context to offer a creative solution. Failures feed the imagination. You cannot have the one without the other.”