David Epstein is a journalist and the #1 New York Times bestselling author of Range: Why Generalists Triumph in a Specialized World. He has previously worked as a senior writer for Sports Illustrated and as an investigative reporter for Pro Publica. His new book, Inside The Box: How Constraints Make Us Better, publishes the day this episode airs.
David joined host Robert Glazer on the Elevate Podcast to discuss his remarkable writing career, the advantages generalists enjoy, and why constraints are a valuable asset in business, creativity, teamwork, marketing and other major disciplines.
—
Listen to the podcast here
David Epstein On Range And His New Book, Inside The Box
Our quote for this episode is from Alexander Graham Bell. “We often look so long and so regretfully upon the closed door that we do not see the one which is open for us.” My guest is David Epstein. David is a journalist and the number one New York Times bestselling author of Range, one of my favorite books. He’s previously worked as a senior writer for Sports Illustrated and an investigative journalist for ProPublica. His new book, Inside the Box: How Constraints Make Us Better, publishes the day that this episode airs.
David, welcome. It’s great to have you on the show.
Thanks so much. Happy to be here.
Congrats on pub day.
Thanks. It’s a lot.
I know the feeling. I say to people it’s like taking six months trying to get a year’s worth of demand in one week. That’s how I described it.
That’s a good way to put it.
From Freak Injury To French Mnemonics
Inside the Box opens with your childhood, which makes this a natural place to start. Can you walk us through the injury that changed your trajectory and the unexpected academic discovery that it led to?
I appreciate that because nobody has asked me about it yet.
I like hearing that.
I was a sports fanatic when I was a kid. Football, basketball, and baseball. I would wake up early in the morning to watch the same SportsCenter segment play twice in a row. One day in eighth grade, while playing schoolyard football at gym class, I had a good throwing arm, so I was the person who threw for the kickoff. You threw instead of kicking in this schoolyard football. On the throw, my arm snapped on the follow-through in a spiral. It completely separated the bone.
Is that a Tommy John thing?
No. Tommy John’s was a ligament tear. In this case, the bone snapped fully.
I thought you heard a tendon snap.
The bone snapped in a spiral, completely through the upper arm, the humerus.
I have no medical aptitude. That was my problem there. We don’t have any science genes in our family.
I’ve only seen this happen one other time to a pitcher. He had his arm amputated because he had some kind of cancer. What exactly was going on for me, we’ll never know. I got to the hospital, and at first, they were telling me to move my hand for X-rays. I felt sick. I was nauseous. I was saying that I was
with my eyes closed, and they were like, “No, you’re not.”
Eventually, I looked and realized that because the bone was separated, I was rotating my shoulder. I was feeling my hand was in a certain place, because that’s where your brain thinks it is, but it wasn’t there. They thought there was no way I could have broken my arm because I didn’t get hit. It turned out I had this freak injury, and we’ll never know why.
The doctor who set the bone said if there hadn’t been witnesses, they would have thought one of my parents had busted my arm until it snapped in a spiral. He said, “Maybe you had some air pocket in your bone or some kind of weakness. We’ll never know because once it breaks, whatever that evidence is, is gone. Who knows?” It was devastating for my life at the time because sports were all I cared about.
I had to have my arms strapped to my body, not just a cast. I was immobilized. I couldn’t play sports. School was terrible. I couldn’t put my arm through a sleeve. There were a few interesting things that happened, but one was that I had a French class at the time. Our tests were where you’d have to listen to someone speaking French on a recording. You had a worksheet that would follow along with the same text, but then it had blanks. You had to catch the words that you would fill in for the blanks while you were going.
My writing hand was strapped to my body, so I couldn’t write fast enough to keep up with these tests. I started memorizing the words as they would fly by and using mnemonic devices. There was usually some sports-related idea that I would attach to the words so that I could remember it. I would go back at the end and write very slowly with my left hand. I started doing much better than I had done before. I started using mnemonic devices for all sorts of things. These days, if I give an hour-long keynote talk, I memorize it largely using mnemonics.
Decades after that, I came across one of the most famous studies of memory research ever done, where a Carnegie Mellon student was taken from being able to memorize seven digits in a row to being able to memorize 80 digits in a row. It was done with sports-related mnemonics. He would attach these number sequences to sports things so he could remember them.
It also got me because I couldn’t play contact sports for a year. I couldn’t play football my freshman year in high school, which I wanted to do. I ended up running cross country to stay in shape, and that got me interested in running. Eventually, I went on to be a Division I 800-meter runner, university record holder, and all this stuff.
This horrible freak thing that happened that devastated me in the short-term ended up leading to these things, both academically and athletically, that I never would have done had I not had these constraints foisted upon me. What looked terrible in the short-term ended up causing me to explore in ways that I would not have done if not forced.
When did you make the deliberate shift into full-time writing and journalism?
For my academic background, I studied Geology and Astronomy in college. I went to grad school. I was living in a tent in the Arctic, studying the carbon cycle. I thought I was going to be a scientist. When I was a national-level runner, one of my training partners died at the end of a race from this condition called hypertrophic cardiomyopathy.
That one, I’ve heard of.
I realized that we could save some people by raising awareness of this. I decided I wanted to merge my interest in sports and science and write about this. I wrote Sudden Cardiac Death in Athletes for Sports Illustrated. I left the track to become a scientist and started taking any job I could get in journalism. I worked my way to Sports Illustrated and did exactly that. I became the science writer at Sports Illustrated. I left grad school, training to be a scientist, got into writing, and eventually made my way to Sports Illustrated, where I became the science writer.
The Tiger Vs. Roger Path: Why The “Sampling Period” Wins
I can see how this arc works. Usually, it takes a little more work. Let’s dive into Range first because it’s the foundation for what follows and the new book. The central argument in Range is that late specialization after a period of broader general training produces better outcomes than early focus. You open with this story in the book, which a lot of people remember, of Tiger Woods and Roger Federer.
Woods was hyper-trained as a golf prodigy, one could argue, from birth. Federer was a good athlete and then settled into tennis later. Why is that pairing such a useful frame for the argument? Not to pick on Tiger, but the argument seems to be getting better with age when you look at what’s happened to him and the cost that he paid for all that intense specialization.
I try not to draw a causal link between his very narrow childhood and adult problems. I use him, Tiger, in that opening of the book in the introduction called Roger versus Tiger as an example of success. I’m talking about his athletic success. Then, I contrast it with Roger Federer, who played a dozen different sports and delayed specialization.
The reason I picked them both is that they’re both examples of success. My question was, which one of these is the norm, and which is the exception? I’d like to contrast those two because they’re equally famous, same era, everything like that. The science shows that the Roger path is by far the norm, and the Tiger path is the exception. Even tennis enthusiasts don’t know anything. Everyone knows Tiger was on a developmental path. It’s probably the most famous story of modern development. Maybe Mozart would be second.
The science shows that the Roger Federer path to success is by far the norm, and the Tiger Woods path is the exception.
The Williams sisters were a similar version of Tiger.
That’s what I thought. I gave a talk to a small group about some of the stuff in Range, and Serena Williams was sitting in the front row for this talk. I’m like, “This is terrible.” I can marshal all the data I want and show what the norm is and what the exception is. The norm is this broader, early so-called sampling period before you pick. If the goat stands up and says, “You’re an idiot,” it’s going to be a bad day in front of that audience, no matter how much data you have behind it.
She raises her hand to ask the first question. I’m sweating bullets. She says, “I think my father was ahead of his time. He had me do track and field, taekwondo, gymnastics, and ballet. He had me and Venus learn how to throw a football for the overhand snapping motion of a serve and all this stuff.” I was thinking I was a Senior Writer at Sports Illustrated.
I’d never heard that, not that they didn’t have a lot of tennis folks practice when they were young. Even that story gets more complicated when you look closer at it. After that, I asked her, “Could I catch up with you later in the day and interview you about this?” I added a little bit about it to the afterword of the paperback of Range because she then sat for an interview with me.
The Resilience Of A Diversified Identity
Have you studied at all whether there’s success, and what the arc of it looks like? When we look at Tiger, how does the peak hold? How does it fall off? I’m making this up, and you can tell me if there’s any science. If there was one thing that was your whole purpose in life, and you started hitting some roadblocks in it, I could assume that that would come apart a little bit more, rather than if you had lots of different things that filled your needs and wants.
There are multiple different lines of evidence that support what you’re saying, some of it about building your identity house. You want multiple rooms in your identity house. You don’t want to be living in one all the time.
You want multiple rooms in your identity house. You don’t want to be living in one all the time.
This was business people. They sell their business, retire, and then die or fall apart within a year.
If you’re asking sports-specific, there is research showing that often, when kids are deemed very talented early on, their development will start to get accelerated. More resources will pour into them or whatever. They’ll be moved into what some of these scientists call a more restrictive environment or a much more programmed and structured plan, much more like an adult.
When they then have their first loss, which may be other people catching up, like biologically maturing, and so they’re not as dominant, the quitting rates are enormous. It’s even higher among girls. Everything is in this one basket. It can impact your resilience. For adults, there’s this line of research showing that if adults have a hobby unrelated to their work, it improves their self-efficacy, feeling of competence, and ability to overcome a challenge. If all their activities are too closely related to their work, it decreases their self-efficacy. There are all these lines of research showing benefits for diversifying your identity in terms of resilience.
The Industrialization Of Youth Sports
The people who didn’t read Range are probably the ones leading or sending their kids to club sports. I live more in a soccer area. There are these 10 and 11-year-olds starting club soccer for whom they can’t do other sports or play on other things. They’re doing it five days a week. We’re seeing the consequences. We’re seeing repetitive injury. We’re seeing burnout. They get tired of it in high school, and they’re walking away entirely.
Why are people going in the opposite direction of the data? There’s this false belief that this is the great hope of getting their kids to college sports or pros, and the odds have not improved. When you look at the industrial Asian complex of youth sports, there are no more real college slots or pro slots, but there’s a massive funnel underneath it.
Some of this has nothing to do with what’s best for the kids.
I would argue very little of it.
When I was in New York City, there was a U8 travel soccer team that met near me. Is there a human being in the world who thinks that 7-year-olds can’t find good enough competition in a city of 9 million people that they have to travel?
My friends’ kids are on these travel teams, even fourteen-year-olds. They’re not the best in New York, but they travel to Texas for these exhibition competitions. I understand if you were the best team in New York, and then you were in the club finals. It is unnecessary to travel out of state.
In that case, at least with the team that I saw in New York, this was about an adult locking in a customer pipeline. If the kids aren’t on the U8, then they can’t be on the U9. It sets up a system where if you don’t play the game, you don’t even have a chance to participate. There are places that do this differently, like Norway, which is probably the greatest, at least, per capita sports country in history.
Also, the happiest place to live in the world. They’re constantly in the top five.
Finland won, but Norway was out there.
Generally, Scandinavian.
Scandinavia does very well. Norway won 50% more gold medals in the US at the Olympics. They dominated.
That guy was an absolute beast, the cross-country guy.
They want a ton of stuff. In the summer Olympics, they won beach volleyball. You don’t usually think about that for Norway. They have people at the top of soccer, golf, tennis, and track and field. This is a country that’s the size of Metropolitan Atlanta. They do not allow kids to be deselected before age thirteen. They’re not allowed to keep scores. They’re not allowed to rank kids. They view mass participation as aligned with elite performance.
It may be that your parents started you in elite soccer. You love basketball, but there’s no time to play basketball.
It could be that you’re better at basketball. The earlier you choose, the more likely you are to put the wrong person in the wrong spot, especially when that’s pre-puberty.
The earlier you choose, the more likely you are to put the wrong person in the wrong spot, especially when that’s pre-puberty.
Do you know what’s interesting? As you said this, I hadn’t thought about it before. Maybe we’ll make an intellectual discovery here. If you’re thinking about what you get from cross-cultural opportunities, the one thing historically people have always said about sports was you learn how to be a good teammate and a team member.
A lot of teams are competing against your town and winning the high school championship. These kids are playing in games where, honestly, the team doesn’t matter. They’re playing for coaches and spots. They’re losing one of the biggest things that could probably help them in other areas, which is about being a good teammate.
I worry about the way youth sports in this country are developing because some places do better. There are virtues of sport that we give lip service to that we are not supporting in practice. That includes what you’re talking about, which is being devoted to a specific team that you can’t move away from, and not only pursuing your own individual goals. The structure of things is not shaping up to support those values that we want. In the US, we can get away with having poor development on an individual basis because we have so many athletes. Take my sport, track and field.
We have so many sports in the US. In a lot of countries, there are three things people do.
For my sport, track and field, there are 30,000 to 40,000 young adults being supported in pretty serious training by the NCAA system. If I had to guess, I would guess that’s probably roughly equal to the rest of the world combined in terms of people being supported in doing pretty serious training. We can do all sorts of stuff wrong with individual development in lots of our sports and still do well at the elite level. To get to some of the ideas of Inside the Box, the countries that are more constrained, like Norway and Australia, which have six pro sports and not a big population, have to be a lot more concerned with individual development. They have much more holistic development pipelines.
Last question, and then we’ll get there. You made a strong case in Range that, outside of sports, the best problem solvers, companies, and organizations draw on ideas outside their own field. I remember reading about The Medici Effect. The whole Renaissance was based on this concept of people from all different disciplines working with each other. It’s the same thing in the workplace. There’s a pressure for people to specialize versus go learn, bring back stuff, and think about a problem in a different way.
Spillover Effects And The Superpower Of Being “Behind”
Two things to say about that. One, since you’re mentioning the Renaissance, there’s a famous econ paper by an economist named Ed Glaeser, who is a brilliant guy. He wrote this famous paper on something called the spillover effect. There was this debate from two very prominent econ camps about what causes cities to drive so much innovation and make people so much more productive than in non-cities.
The two competing theories were regional specialization. You focus on a particular industry in an area, and that draws talent to it, infrastructure, etc. The other was spillover, where you cram lots of different types of organizations together, and people and solutions start jumping from one industry and organization to another. It’s the mixing. What he showed is that spillover is the answer. You want to cram all these different people together. They take analogies from one industry or one problem and bring them over to another
Once you get that broad toolbox, you have to focus it on achievement and satisfaction.
He did an amazing profile of Detroit back in the Henry Ford days, where it was like Silicon Valley. There are people working on boats and carriages, and then you have people like Henry Ford who would say, “I can combine that motor from that with that carriage over there and make something new.” When it became a one-industry town, that’s when it was on the path to dying, where people were not having that spillover effect anymore. That’s the one thing.
There’s a new paper out in the journal Science, 1 of the 2 most prestigious scientific journals in the world. Recent Discoveries on the Acquisition of the Highest Levels of Human Performance is the name. It looked at about 30,000 careers of musicians, scientists, and athletes. It backed up the thesis of Range. They cited some of the same papers I did.
What they found was that the things that predict top youth performance, including early narrow specialization, become negative predictors of adult performance. Almost none of the top youth people were the top. There were some, but the 90% of the top adults were different from the top youth people. They showed this, to your point, outside of sports.
They looked at scientists. The Nobel laureates, in the research they looked at, progressed more slowly in their careers than their peers because they were more interdisciplinary early on. It gave them this power later on, but it meant they got tenure later. They didn’t publish high-impact papers as early, and all this kind of stuff. That breadth becomes a superpower, but it makes you look like you’re behind early on.
In a world that doesn’t value delayed gratification, that’s a problem. It occurs to me that this correlates to the data people say where SAT and test scores correlate to how people do in the 101 classes in college. As they move up to the classes and it’s more about participation, not just getting the right Xs and Os, that correlation tends to fall apart a little bit.
That’s interesting. I don’t know that particular research. I’ll take your word for it.
I’m pretty sure I’ve read that. It’s interesting. What a line. That brings us to the new book, Inside the Box. In some ways, it seems a little bit, to me, like the sequel to Range.
I’m glad you said that.
Inside the Box: Turning Curiosity Into Achievement
If Range made the case for the generalist path, the new book is how you turn that into achievement. What pulled you towards that question?
That’s how I feel. On the face of it, Inside the Box talking about how constraints make us better can seem contradictory to Range. To me, there were two main things that drew me to this topic. The most common question I was getting from people after Range was, “I’ve got this broad toolbox. Now what? Sometimes, I’m so curious and so wide-ranging. What do I do with it? I want to bounce around.” In some ways, this was my answer to it. Once you get that broad toolbox, you have to focus it on achievement and satisfaction. I wanted to look at the obvious follow-up question, which was how do you put boundaries around your projects and your life in a way that can lead to achievement?
I always say the same amount of energy in a prism can refract and be a lot of pretty light, but in a laser, it can cut glass. How do you not spread yourself too thin?
With Range, the older I’ve gotten, the more I realize there’s a big dose of me-search in my books, not just research. I’m trying to understand my own life. I’m a generalist. I’ve zigzagged all over the place. I have also been terrible at putting constraints around my own work. For my first 2 books, I wrote a 150% the length of a book to get 1 book because I didn’t draw good boundaries around the project, and I have this expansive curiosity.
In my first book, I cut a trip to Arctic Sweden for goodness’ sake, that I would have realized that if I’d done more planning and had boundaries upfront, it wasn’t going to fit. After Range, I became a dad. I’m like, “I cannot be taking trips to Sweden. It’s not going to be useful for the book. I need to be more efficient with my time.” I wanted to learn how to draw better boundaries around my own work day, in my life, and around my projects. This time around, I went through this long process where I forced myself to write on one single page a structural architecture for the book before I started.
I got this idea from a character in the book, Tony Fidel, who was the lead designer of the iPod and co-founded the company Nest. His main advice to entrepreneurs is to write the press release before you start the project because it gives you this vision of where you’re going, but also, he calls it a bounding box. This shows you what your priorities are. You have it written down, and you don’t move the goalposts. I made this architecture for the book before I started writing. It’s one page only.
Is it pretty close to where you ended up?
Yeah. I ended up writing as small as possible as if I was trying to cheat my own system. If it’s not on that page, it’s not in the book. Consequently, the book ended up 20% shorter than my previous two. I didn’t overwrite it. The writing’s much tighter. The book is more coherent because of that. I would never not do it this way again.
You took the Mark Twain quote. “Sorry, I would have written you a shorter letter if I had had more time.”
That’s very often the case. The things I’m writing about are things that I want to be better at.
That makes sense. We’re justifying why our path is the right one. You’re a deep researcher. One thing that jumped out was how you spent hours with Pixar Founder, Ed Catmull, watching a documentary and discussing elements of it. Talk about the value of that deep, but meandering research. I think that’s part of the story.
General Magic Vs. Pixar: The Danger Of Unlimited Resources
In the first two chapters of the book, after the intro, I’m contrasting two companies. One is this one called General Magic that had all the talent and resources in the world. They had the vision of the iPhone many years before it existed. They went public in the first so-called concept IPO. They had such an amazing idea and collection of talent that Goldman Sachs took them public without a product, just an idea. It ends up becoming a complete disaster. They have so many resources and so much talent that they can build anything, so they do. The project gets bigger and spirals out of control.
I contrast that with Pixar, which at the same time had this equally big vision of making the first computer-animated feature film. This was a time when computer graphics were solid objects rotating on a black screen. They end up pulling it off. One of the things I thought was interesting was that Ed devoted his career to putting constraints in place that help channel people’s creative ideas.
There are tons of ideas in organizations. It’s a process for setting up the bumpers in the bowling alley and getting them to trundle in the right direction. That’s much more rare. General Magic, the company I wanted to contrast, has a documentary about it. I wanted to know what Ed would think about General Magic. The best way I could do it was to sit next to him while he watched the movie and ask him to talk out loud, because then, you can write in a way.
He hadn’t seen it before, so you were getting his raw reaction to it.
The reason I like to do that is that that’s the closest you can get to being able to narrate someone’s actual thoughts. Instead of me asking him, “I saw this in this movie. What do you think about it?” it is him watching it and then being like, “Ed, please speak your thoughts out loud.” I can write what he’s thinking while he’s watching that. I thought it was a useful reporting technique.
No one’s heard of General Magic. Why don’t you tell us what it is that they were trying to do, because that’s interesting.
This was a company founded by three guys. It was 2 of whom were designers of the original Apple Macintosh, and one of whom had a job inside of Apple, this guy named Marc Porat. His job at Apple was to see the future of technology after personal computing. We’re talking late ‘80s. I read his PhD dissertation in my reporting. On the first page, he coins the term information economy. He created that term.
I was reading the PhDs from 1976, and it is amazing how prescient it was. He’s like, “There’s a communications and information technology revolution coming. Most people have no idea it’s going to change everything about work. It’s going to lead to problems with misinformation. Commerce is going to be virtual,” and all this stuff. The internet did not exist. Almost nobody even had computers at home.
He was clearly a visionary.
That’s why his job at Apple was being a visionary. He was visionary with General Magic. I saw it in his notebooks. In 1989, he sketched a thin glass rectangle with no protruding buttons, just a touchscreen, where you could download apps. This is pre-Internet. It would be a phone and a computer. You could play games, get airline tickets, and send messages. He saw the future.
One of the reasons they went public so early was to create a heaven for engineers where they could only be limited by their imaginations. He said, “What else could anyone ask for?” It turned out the answer was a little less freedom. They had so much freedom and so much talent that they could do anything. Everyone who had a good idea, they built it.
It turned out the answer was a little less freedom. They had so much freedom and so much talent that they could do anything.
One of the emblematic interviews for me was with this guy, Steve Perlman. He was an engineer at General Magic. He was in charge of creating a calendar function. He creates it to go from 1904 to 2096. He checks it in and says, “Fine. I’m done.” Someone else comes to him and says, “Steve, someone might make apps that go way farther back in time. You’ve got to make this calendar app bigger.”
He makes it to go from year one instead and checks it in. Done. Another team comes to him and says, “Steve, why are you starting with an arbitrary religious context? You should go back to the beginning of astronomical time. What kind of program has a calendar function go from the beginning of the universe into the future?
None of this being any functional use cases.
It would have been four lines of code if he’d stuck with 1904 to 2096, but instead it took months. This was emblematic of everything at General Magic. They called their user Joe Sixpack. Nobody bothered to define who the heck that was. After several years of missed deadlines, they realized nobody knew the guy. They had not identified a clear problem they were solving. When the device appeared, their personal communicator, it had so much stuff on it that the user experience was choppy and confusing. The battery life was terrible. It was expensive.
Meanwhile, people inside General Magic who bit off tiny pieces and were focused on it did some important things. One low-level service engineer, when the internet exploded while they were missing deadlines, started a website called Auction Web that facilitated online auctions. He went to the chief financial officer and general counsel and said, “Look at this. It’s facilitating commerce between strangers. Isn’t that what we want to do?”
He offered it to them. They said, “Too small. We have a much bigger vision.” He took it and left, and that became eBay. Pierre Omidyar was a low-level service engineer. They had a third-party app developer who created an app where you could use a stylus and make strokes, and it’d be turned into writing. When it was clear that things weren’t going great at General Magic.
It sounds like PalmPilot.
The Palm Pilot Vs. AI “Work Slop”
It was PalmPilot. He took it out. When it was clear that General Magic wasn’t going to work, he decided to make his own device. He said, “I’ve identified a clear customer problem. It’s busy professionals who want to sync their contacts and calendar on the go with their computer and take it on the go. I’m going to make one that’s just calendar contacts and a memo pad.” General Magic said, “Our partners are AT&T, Motorola, and all of that. There’s no chance you can’t beat us.” he destroyed them. The PalmPilot also became a hit. They could not figure out how to do something instead of everything.
I know you must have written the book a couple of years ago at this point, but I can’t help thinking about the AI industry and how it benefits from some more constraints, and maybe the lost chapter, if you were writing it now. They seem to be operating with a blank check. Unlimited data centers and unlimited money. I’m curious whether you think it’s good or bad.
I had a thesis. A couple of months ago, we did an episode talking about it with my team. I said, “I don’t think all these models can build for everything.” You saw OpenAI dump Sora. That’s even from a cost computing standpoint. What’s the danger when everyone in AI is trying to act like everything’s a win-all race and spending money like it’s growing on trees?
Even if the technology is transformative, it’s a bubble. There’s a lot of mismatch between the sexiness of the tool and solving people’s problems. There’s this new term coming out. It started with some research at MIT. It’s called work slop, where people are creating an incredible amount of mediocre stuff that is causing more burden on the people they’re sending it to.
Even if the technology is transformative, it’s a bubble. There’s a lot of mismatch between the sexiness of the tool and solving people’s problems.
Have you seen this cartoon where one guy’s saying, “AI is great. I gave it these four bullet points, and it turned it into a six-page memo, and then emailed it.” The person on the other side says, “AI is great. It took this six-page memo and turned it into four bullet points.” This MIT research on work slop is that it’s not adding to productivity the way that we might have thought. There’s an issue with these tools being so expansive and not being targeted at specific problem-solving. I’ve seen this in organizations, too.
There’s one particular AI company that I embedded with a little bit. The organizations that use them are saying, “We need AI,” because they feel behind if they don’t have it, but they haven’t identified a specific problem that they should match the tool to. They end up implementing stuff, and it’s a problem because it didn’t well define the problem they were trying to solve. They implement something. You create all this work stuff and have to figure out how to reel it back.
With these mass models, I think, “Who’s the audio company? Who’s the video company? Who’s the work one?” The way I’ve made it work personally and professionally is with customization and putting constraints around it. It is much better when you reduce what it’s supposed to do and narrow its focus than let it go off.
I come from the marketing world. You’re going to have to say, “Either this becomes a marketing book or a company launch book,” or do the little derivative. The two problems that companies have, and one is a war thing, like building your front. They don’t pick a defined user audience that gets them launched. We tried to sell large global projects at our company, and we couldn’t, even though we had a global platform, because there was a lot of politics. We started to get into one company and get a global MSA, and then we would grow. If we tried to get the whole thing, we wouldn’t get it if we got over the wall.
1) The focus on the customer and finding the one thing that they’re willing to turn over money for. 2) I’ll bring some examples if you don’t have them, because they might be interesting for your research, is that companies that historically didn’t have enough marketing money did amazing things. They got super gritty. That seems to be two requirements for success in building a company.
Indigestion Vs. Starvation
When you don’t have resources, you’re forced to get resourceful. The venture capitalist, Bill Gurley, who famously invested in Uber, Zillow, and all these things, was a big fan of Range. We came into contact through that. He said, “Constraints? We have a saying in venture. More startups die of indigestion than starvation.” Too much, not too little. Tony Fidel said that same thing. He was like, “That’s my line.”
More startups die of indigestion than starvation. Too much, not too little.
Multiple people are claiming that they came up with this line. It is interesting. Ed Catmull of Pixar said, “I found if we put too many resources early, things get sloppy.” You want to keep it tight and lean early. This gets at one of the major underlying points in Inside the Box. As the cognitive scientist Daniel Willingham puts it, our brains are not made for thinking. They’re made for preventing us from having to think whenever possible because thinking is energetically costly.
Given total freedom, we’ll default to what cognitive psychologists call the path of least resistance. You’ll do the convenient thing. You’ll do the thing you’ve seen before, the thing you’re used to, or the thing that seems easiest. In order to get creative, we have to have constraints. You have to have the typical solution or the one that you would do with complete freedom blocked. That’s the only way that you can be creative because of how our brains work.
I’m curious. Do you know the Honest Tea and Stonyfield yogurt examples?
No.
The Ingenuity Of The Underdog
They could be case studies in your book. They came to me because I heard both their CEOs speak. Honest Tea was struggling early on. They didn’t have a lot of money and were very constrained. They set up these unmanned pop-up stores in cities with a bunch of bottles and a donation box. You could take them and the suggested donation. What they did was they videoed it and created a, “Which city is more honest?” benchmark. The news got all over it. They got millions of impressions. They got double-digit sales growth in every market because the news covered it.
It was called the Honest Index. It was a shoestring budget thing that would have worked much better. The other one is that Stonyfield had gotten into the Chicago market. In order to stay there, they were told that they had to triple their market share in three months to stay at the supermarket or whatever it was. Gary Hirschberg, they told them it was going to cost $50,000 in advertising to do that. He was like, “I don’t have that.”
He decided to go to the transportation authority and gave out 80,000 cups of yogurt to people on an L with a coupon celebrating them for taking the train. He made a whole initiative out of taking the train. He ended up getting in the market share in weeks, not months or whatever he got to. It was 1/5 of what they said. These are examples that could have been in there. Ingenuity is core. When you have unlimited money, it’s hard to believe people make good choices or think about what could work.
That reminds me, too, of business necessity. There are a few business necessity cases I talk about in the book like the one that I like with Stan Lee. He was the editor of Atlas Comics, and their whole strategy was pumping tons of volume in different genres. When their rival became their distributor and the rival limited them to a few issues a month, he said, “We can’t do volume anymore. What are we going to do? We’re going to have to do sustained storytelling.” That’s when they started doing superheroes with real-life adult problems, anger issues, and all these other sorts of things. It was the necessity of, “We can’t do volume now.” Southwest Airlines, you know that one?
Is this is the alcohol story?
No, I don’t know that one.
You tell me yours, and then I’ll tell you mine because it’s one of my favorite business stories of all time.
A year after Southwest started, they were dead. They had $140 in the bank or something. In order to make payroll, they had to sell 1 of their 4 planes. All of their routes were in Texas. They were like, “We have to sell one plane to make payroll, but we’re dead without that because then, we can’t fly as many flights.” They decide they have to sell the plane.
This VP of ground operations does the math and realizes they can still do all the flights if they can turn around the planes in 10 minutes instead of an hour. They’re going to line up passengers on the tarmac and bring them in one door while people are going out the other door. The flight attendants are going to be pulling bags down as soon as it lands. The pilots are going to help clean the cabin and all this stuff. They end up turning around the planes in 10 minutes when normally, it had been an hour. That was the first of 47 straight years of profit. They may have been profitable up until the pandemic.
They had more profit than the whole airline industry over a twenty-year period. Their whole business was all about constraints. They’re losing it. They were non-standard airports. They were like, “We’re not going to do the first class.” It was all the things that they weren’t going to do. I try to be a student of Southwest. There is a famous story where this woman kept writing to Herb Kelleher, telling him everything she hated about the airline. He wrote back and said, “It sounds like we’re not the right choice for you.”
I knew about their turn advantage, but I didn’t know that story. I’ll match you with the Jim Beam. Apparently, a little further down the road, they were getting going, and an American came in on one of their routes. It was $80. Americans started charging $60. They were going to try to put them out of business. They were flooding the market with these low-cost fares, and Southwest was like, “We can’t compete. They have more money than us. We need to do something.”
This is one of the most famous Southwest stories. Maybe you can add this. It’s still one of my favorite business stories ever. Someone looked at it and was like, “The people flying this route are mostly business travelers. We’ll give them a choice. They can pay the $60 or pay the $80, and they get a free handle of Jim Beam.” They knew that people were expensing this. They became the largest distributor in Texas of Jim Beam for four months. Ninety percent of the people took the alcohol thing because they were expensing the ticket, and they got free booze. American gave up the route. Southwest is also still alive because of that story. There’s a history of constraints.
There’s one being resourceful off a constraint after another.
That’s their DNA. There’s also this anecdote you have in the book about scientific research that I thought was interesting, describing this trap. Scientists conduct an experiment, sift through the data, and ultimately draw a conclusion that may be random by chance. How does this happen, and how does this relate? I thought that was a little different.
The Replication Crisis: Limits As Truth-Tellers
This chapter is about the so-called replication crisis in science, the fact that a lot of famous research has not held up on further review. A main problem has been that researchers have too much freedom in how they make conclusions. What led to this is that researchers were doing things like making some predictions. I feature Nutrition Research quite a bit in this chapter because it’s a mess. A researcher might say, “If you’re watching an action film versus a romance film, you’re going to eat more calories while you’re snacking.” That’s their theory. They take the data, and it doesn’t hold up.
It’s too wide.
If you stuck to your original hypothesis. What typically will happen is they’ll say, “That didn’t hold up. We have all this data, so let’s look for other stuff. Let’s start sifting and looking for other correlations.” It’s very counterintuitive that it’s wrong to do that. For statistical reasons, if you don’t stick with your initial prediction and you’re retrospectively coming up with your prediction. Which is called harking or hypothesizing after the results are known. You’ll always find a false positive result by sifting through your data.
It’s the genesis of a conspiracy theory. You can take all these points and point them to some theory that doesn’t exist.
I opened that chapter in the book by looking at how this one segment of the National Institutes of Health had people studying drugs and dietary supplements for cardiovascular health. Before the year 2000, most of the drugs and supplements had positive results for health, and then starting in 2000, almost nothing has had a positive result. Did medicine stop working all of a sudden? No.
In 2000, they required the researchers to register ahead of time the prediction that they’re making. That simple constraint caused almost all of the results to turn negative, which means that in the past, people were making predictions. They were turning out negative, and then they were sifting the data to find some other positive thing, which means it’s all false positives. There’s a ton of drugs and stuff out there now that, for sure, are based on false positive results.
Interestingly, maybe it’s a little bit meta, but I did see an article that said the least reputable research all came from higher education. Maybe that’s feeding the problem.
That’s interesting. From higher education, meaning from academic research?
From academic research and studies. They were the least replicable when they went back and asked them to prove it.
One of the reasons that we figured this stuff out is that academic researchers started calling their colleagues on it. Now, science is in a much better place, but there’s a ton of nonsense out there. One of the studies I cite in that chapter is often referred to as the everything in your fridge causes and prevents cancer study. These researchers gathered up all the studies of different foods and found that everything had been found both to cause and prevent cancer many times, except for bacon. It had only been found to cause cancer. It’s very sad.
Bacon has not been getting a lot of good press.
The point of the chapter, ultimately, is that the replication crisis is very interesting. I get into how normal people and businesses can use the scientific testing framework to make better decisions.
It would seem that the more things you try to look at are variable and more complicated the thesis. The harder it would be to replicate it, the narrower the scope, the more you’d be able to test what you’re looking for.
That’s why in that chapter, I cite this research where startup founders were randomized to different types of training for market research. Some of them got training framed as the scientific method, where it was like, “You’re going to make a hypothesis that has to be well-defined, and then you’re going to find a way to test it.” That forced them to define the problem they thought they were solving explicitly.
They’re very good at telling stories.
Also, moving the goalposts, if the story seems like it should change. In those cases, a lot of those founders ended up realizing that some of their assumptions were wrong once they were forced to explicitly define the problem they thought they were solving.
They didn’t think they could escape their way out of it. The slack pivot is one of the most told, but it is the total survivor bias. The one out of a million that works, where you can change the whole business premise if it wasn’t what you thought it was going to be.
That was way into things. In the research that I cite in this chapter, people are randomized at the beginning. They’re early startup founders, and they’re randomized. A lot of the ones that use the scientific method of market research end up realizing that the problem they think they’re solving for end users is not the problem that the end users want solved. It’s some other problem. They end up pivoting very early and end up making a lot more money. Whereas the other startup founders more likely retrofit their story instead of pivoting and do a lot worse.
I’m curious if you have a thought on this. I have a thesis, but I am not a scientific researcher. Between COVID and the ZIRP decade or the Zero Interest Rate Policy decade, there was an entire group of executives and leaders who grew up with no constraints. You solve a problem and add more people. All the layoffs that people are blaming on AI. That’s about a quarter true. It’s undoing all this dramatic adding of resources since COVID.
My theory is that I don’t think it’s that easy to pivot from someone who is used to operating with no constraints for a decade of the formative part of their career. People act like it’s a flip you can switch. If I’m hiring someone, this is one thing I’m looking at. Did they operate with free money, and did they know how to do otherwise?
I agree. I’m speculating, but your intuition is right about that. Some of the companies that have said, “Due to AI, we’re firing,” Block was one. People internally started writing articles saying it wasn’t AI.
They have hired 500% growth in employees since 2000 or something like that.
AI was a convenient excuse.
The public market likes that.
You’re right that if you haven’t had to work under constraints, you haven’t been forced to be resourceful. That’s going to show up later.
Related to that, a piece of advice that he had is that if you don’t have constraints, make up constraints. How do you recommend leaders do that for their teams?
Tony Fadell’s “Styrofoam” Constraints
That was another one that Tony Fidel told me, the Nest Cofounder. This was the first time I interviewed him. I interviewed him several times. He was almost yelling at me on the phone, “You don’t have constraints. Make up constraints.” He’s a very enthusiastic guy. At Nest, for example, he forced his team to work inside a literal box where he had them prototype the packaging before they had the product. He said, “This is what it’s going to look like on a shelf to an end user.”
“By prototyping this packaging, we’re figuring out what are the things that we want to communicate in this tiny space to the end user. If it doesn’t fit on there, then it’s not one of our priorities this time around.” As he told me, with these ultra constraint-based things, it slows you down, but it forces this kind of thing. That’s why he advises writing the press release also ahead of time.
He had been in companies where things drag on forever, so he showed Steve Jobs a styrofoam model of the iPod in March of 2001. He got the green light and said, “Ten weeks for a first version.” A clear user problem was growing music collections in the MP3 era. There was no way to carry them on the go. After ten weeks, you do what some designers call Design Freeze.
Whether you’re done or not, everyone stops, collects their lessons, and then you go to the next version. They ended up going from a styrofoam model to shipping in nine months. He picked Christmas as what he calls external heartbeats, where he wants some external benchmark so everyone can see it’s not just the boss making stuff up for when they’re going to ship. It was using constraints.
There are a number of constraints if you want me to throw out some thought exercises that I’ve used with executives. The main mindset shift I hope Inside the Box engenders is from viewing limits as only bad to viewing them as opportunities to clarify priorities and launch productive exploration. A few of the thought experiments I’ve used when I’ve been starting out talking about this stuff with people are to pick one behavior. If you could only pick a single behavior that you wanted more of in your organization, what would it be? That can be clarifying for people working on their culture. If you had half the time or budget, how would you do this?
One of the first people to read the book was this guy, Ed Hoffman, who was the Chief Knowledge Officer at NASA. It’s like the head psychologist. In reading it, it was like you did with Southwest. He was like, “I have to tell you about this NASA mission called L-Cross,” where the team ended up with half the time and budget they wanted.
First, they complained, and then they said, “If we were going to get this done, how would we get it done anyway?” It forced them to borrow imaging equipment from Army tanks and engine temperature sensors straight from NASCAR. They pulled them right out of NASCAR, put them on a probe, and confirmed water on the moon. He said they never would have done this unless they were forced to do it.
In doing those kinds of thought experiments, what’s called a preclude constraint, take your most familiar tool or oldest tool and say you can’t use it. If you’re going into the next client meeting, if we weren’t allowed to propose the thing we would always propose, what would we propose instead? That can be a generative prompt.
Tell me what you think about this one. There’s one executive that told me he has a constraint that he’s found useful. He calls it the legacy constraint. He said he told his team, “If we were going out of business in two years and only we knew it, and our clients and customers didn’t, and we couldn’t tell them, what would we do differently?” He said immediately, people started saying, “I’d stop focusing on that, and I’d focus on this.” His answer was, “You should do that.”
The Legacy Constraint
Financial advisors will give clients, “If you had a year to live, a day to live, and ten years to live.” You have no idea whether you have a year or a decade to live. As you remove the time thing, people got very specific about what mattered to them most.
It’s like the Stoic philosophy ‘Memento Mori.’ Every morning, you’re supposed to remember your mortality and that you have limited time. It helps you prioritize. Humans have hardwired what’s called additive bias. We always add things in order to solve problems, even when taking away would be better. I describe one organization doing this in the book. They put all of their current commitments on Post-it notes and put them up on the wall.
Humans have hardwired what’s called additive bias. We always add things in order to solve problems, even when taking away would be better.
Usually, what you’ll see when you do that is that you’re overcommitted. That some things are not as important as other things. I did that in my own life with my own projects and then said, “If I had to cut something within the next 90 days, what would it be?” It doesn’t mean you have to kill that thing forever, but you can move it out of the current funnel. That can be useful.
You’ll see if you miss it. David, where can people learn more about you and your work? I’m assuming they can buy Inside the Box anywhere books are sold.
They can buy Inside the Box anywhere books are sold. They can go to DavidEpstein.com. I have a free Substack newsletter that’s linked on my website as well.
I am a subscriber. It’s very good. Everyone should subscribe. We’ll add that to the recommended list on my Substack, too.
I have a few tips. Some of these exercises, I talked about on my website as well.
Thank you for joining us. I’ve been a fan of your work for years. I’m excited to have this opportunity to chat with you.
It’s a pleasure. Thanks for telling me some of the Southwest story. That Jim Beam one is hilarious.
I might have Jim Beam wrong. It was some cheaper alcohol.
I’ll look it up.
There are famous articles on it. I’ll send you the Substack I wrote on it. You can learn more about David and Inside the Box on the Episode page at RobertGlazer.com. If you enjoyed this episode or you think someone else could benefit from it, the favor I have is do you mind texting it to him or sharing it with them. That’s how new users discover the show. If this episode made you think or see differently, we’d love it if you could share it with a friend. Thanks again for your support. Until next time, keep elevating.
Important Links
- David Epstein
- David Epstein on Substack
- Range
- Inside the Box: How Constraints Make Us Better
- The Medici Effect
- Episode – Thinking Thursdays: How Should Leaders Use and Limit AI?
