Why the Most Innovative Companies Don’t Bother with Promotions | Next Big Idea Club
Magazine / Why the Most Innovative Companies Don’t Bother with Promotions

Why the Most Innovative Companies Don’t Bother with Promotions

Career Creativity
Why the Most Innovative Companies Don’t Bother with Promotions


  • The real reason why Pan American Airlines went out of business
  • What water teaches us about good leadership
  • Which simple, groundbreaking innovation led to Walmart’s success

Safi Bahcall is a physicist turned biotech entrepreneur, and the author of the Next Big Idea Club Official Selection, Loonshots: How to Nurture the Crazy Ideas That Win Wars, Cure Diseases, and Transform Industries. He recently sat down with world-leading business thinker Whitney Johnson to discuss why office politics could be killing your organizational culture, and how to make your team more focused and innovative than ever before.

This conversation has been edited and condensed. To listen to the full version, click here.

Whitney: What is a loonshot exactly?

Safi: The big ideas that change the course of science, business, or history rarely arrive with blaring trumpets and red carpets, dazzling everybody with their brilliance. They’re usually dismissed for years, even decades, their champions written off as crazy. There wasn’t really a good word to describe those big ideas, so I made one up—“loonshots.”

Whitney: Very clever. In the book, you tell these great stories about Juan Trippe and Robert Crandall, and the different kinds of loonshots. Could you tell us about that?

Safi: I talk about the two types of loonshots, and it’s important to understand what those two types are because almost everyone has a blind spot for one or the other, and missing that blind spot can get you killed. Here’s what I mean by that:

Juan Trippe was a classic product-type innovator—he would champion what I call “P-type” loonshots, new products or technologies that everybody says could never work. For example, when the idea for the telephone came up, [people said,] “Oh, you can’t do that, and even if you can, it won’t be important.” Or the transistor: “There’s no way you can make a switch out of solid-state materials.” Or personal computers, or digital cameras… Those are products that everybody said couldn’t work.

The other kind [of loonshot] is a small shift in strategy that people say won’t matter very much. A good example of that is when Sam Walton decided to move his retail stores to rural America, which he did kind of on a whim. He wanted to open a retail store in St. Louis, but his wife said, “I don’t like living in big cities.” He liked being married, and he also liked quail hunting, so he found one region of the country where there were four quail seasons—four states met at a point, so you could go quail hunting all year round. So he located his store in Bentonville, Arkansas.

Who knew that there was such enormous demand out in rural America? His store, Walmart, ended up dominating the retail industry and wiping out all these other players. [That idea] was a loonshot, but a loonshot with no new technology. There were no fancy gadgets—he just moved somewhere else, and sold stuff a little bit more cheaply.

You asked about the story of Juan Trippe and Bob Crandall from the aviation industry. Juan Trippe was an engine guy, a product guy. When he was young, he got his first plane, then raised some money, took off the propeller, and put on a bigger propeller. He also carved out an extra seat to make it a two-seater, ferrying people from Manhattan to Long Island.

“Underlying culture is structure, which drives patterns of behavior.”

And he ended up starting a little airline called Pan American Airways, which grew into the largest, most dominant airline in the world. The Beatles arrived in the United States on Pan Am. James Bond flew Pan Am. In a 2001 Stanley Kubrick movie, there was a Pan Am spaceship with Pan Am stewardesses wearing Pan Am clothes. It was this dominant, awesome airline, and Trippe did it by going after these product innovations, one after the other after the other. He added radio navigation so that he could circumnavigate the globe, and he was the first guy to develop jet engines when no one said it could work.

He kept trying to get bigger, faster, better planes, and bigger, faster, better engines. And then he heard about this super powerful engine, the 747, and he bet the company on it—a billion dollars. But the problem was that there weren’t enough passengers to fill that [huge plane] because, by that time, jet engines had become commoditized, and there were a lot of competitors. And then when the fuel crisis and airline deregulation hit, he was left flying these giant planes [with no passengers inside]. He missed the small, crucial changes in strategy.

Now Bob Crandall was not a product guy. He was running American Airlines, and he came up with these small changes in strategy that didn’t require any new technology—something called Frequent Flyers. He found that people loved it, so all of a sudden people started going with American Airlines.

Here’s another strategy that everybody said was crazy: Taking the reservation systems that agents used inside American Airlines, and giving it away to everybody—all travel agents in the United States—for free. Suddenly, bookings for American Airlines went up. And when airline deregulation hit, Pan Am went bankrupt, and so did every other competitor—but American Airlines survived.

So who made the cover of magazines? Juan Trippe, because he brought jet engines to the masses. He crossed the globe, bringing the old world and the new world together. It was an incredible triumph, and it was sexy. Movie stars were flying in Pan Am jets. The stuff that Bob Crandall was doing, on the other hand, was much less glamorous. You don’t see Frequent Flyers making the cover of magazines; you don’t see an airline reservation system getting a Time magazine cover. But American Airlines survived and Pan Am went down the drain, in part because Bob Crandall had mastered both types of loonshots, and Juan Trippe had a blind spot for one type.

Whitney: So interesting. In your book, you talk about phase transitions and control parameters. Can you talk briefly about what a phase transition is?

Safi: Imagine you have a glass of water—you can stick your finger in and swish it around. But as I gradually change the temperature, all of a sudden, at 32°F, the behavior of those water molecules completely changes. They become totally rigid, and the water freezes into ice. Why? It’s exactly the same molecules, so how did they know to suddenly change behavior? There’s no CEO molecule with a bullhorn and a thermometer saying, “Okay, it’s 31 degrees—everybody line up!”

That, in science, is what’s called a “phase transition,” a change in structure that’s triggered when you adjust something called a “control parameter.” Every phase transition is a result of two competing forces. In the case of water, one of those forces—called “entropy”—makes the molecules want to run around and be free. The other force is binding energy, which wants to lock every water molecule rigidly in place.

At high temperatures, the entropy force—the running around—wins. Binding energy is very weak. But as you adjust the temperature, the relative strength of those two forces [begin to reverse] until, right at 32 degrees, they exactly break even. And then boom, the system snaps, that tug-of-war flips sides, and the rigid side wins, [creating ice].

Now, when you bring a team or a company together, you also create two competing forces. You can think of those two forces as stake in outcome and perks of rank.

Let’s say you bring a small number of people together around developing a new cancer drug. Everybody has as an enormous stake, because if the drug works, everyone’s a hero and a millionaire. If it fails, everyone’s unemployed.

Now imagine that you grow the company. Once you’re 10 or 20 people, you need to add some team captains, and maybe some vice presidents. So you start to have another motivating force, which is the perks of rank. When your company is really small, [like] five people, it doesn’t really matter who’s a team member and who’s a team captain—you just roll up your sleeves and make the project work. But if your company is 500 people, there’s a big difference between being an associate and a senior vice president. The perks of rank become important.

As you gradually adjust the properties of the company—like size, for example—at some point the perks of rank become more important than the stake in outcome. And at that point, you have a phase transition. The incentive shifts from, “Hey, let’s all roll up our sleeves and make this thing succeed,” to, “What can I do to get promoted?” You have a bigger stake in promotions than you do in the outcome. When that happens, you shift from an innovative culture to a political culture.

There’s so much out there about culture, culture, culture, but culture is what you see on the surface, like political versus innovative. Underlying culture is structure, which drives those patterns of behavior. So if you reward rank, you’re going to get a very political culture. Everyone is going to try to stab their neighbor in the back. If they come up with some crazy idea, you’ll be whispering to your boss, “I think that’s really crazy… Look at all these flaws.” On the other hand, if you celebrate results and uniting around risky, promising ideas, you’ll get an innovative culture.

If you try to get a bunch of molecules to line up rigidly one by one, that’s very hard. But adjusting the temperature to get them to do that [on their own] is much easier. [In the same way,] fixing culture is very hard—forcing people to watch two-hour videos and hold hands and sing “Kumbaya” doesn’t have much effect. But changing incentives can be very easy.

“How can you design systems where everybody’s got more skin in the game?”

Once you understand a phase transition, you can begin to control it, to manage it. For example, with water, you will get a phase transition as you lower the temperature—it’ll go from fluid to rigid, and it will freeze. But the temperature [at which that happens] isn’t fixed in stone; it can be controlled.

When it snows overnight, you sprinkle salt on your sidewalks. Why? Because it lowers the freezing temperature by making the binding energy a little bit weaker, and it keeps things fluid a little bit longer. [Similarly,] with companies you can adjust the control parameters to make them more innovative, [even if they’re very large].

One control parameter is equity fraction, which is the percentage of your incentives that are associated with results versus rank. [Think about what happens if] you pay somebody just based on [rank]: “You make $50,000 dollars if you’re at this level, then $80,000, then $120,000, then $200,000.” Their focus is going to be entirely about getting promoted, which means elbowing their neighbors and stabbing them in the back—it becomes a political culture. Instead, try to focus all of their incentives around the success of their projects. In those cases, people are going to focus on making their crazy loonshots succeed.

[Another control parameter] is management span, or how many direct reports [managers have]. If you imagine a company with 1,000 people, and everybody has two or three reports, you’re going to have many, many layers [of management]. So everybody’s going to be thinking about promotion. On the other hand, if you have 1,000 people, and each manager has 40 direct reports, promotions happen almost never—there are only two layers beneath the CEO, so it’s not even worth your time thinking about a promotion. Instead, you might as well focus on your projects. So if you want more innovative employees who are more focused on results and less on politics, you want those much wider-span groups.

Also think about the salary growth rate, a measure of how much salary increases as you go up each layer in the company. If my salary triples every time I get a promotion, what am I going to be thinking about day and night? Getting promoted. Now, if my salary goes up by 1% every time I get promoted, am I going to care much about being promoted? No, not really. If, instead, I’m paid based on the success of my projects, why would I care about promotions? I’m just going to focus on [getting results].

The last [control parameter] is what I call “organizational fitness.” It’s essentially a measure of, “How much does the company reward politics, versus how much does it reward skill?” Let’s say you’re an average employee, and you’re working on, say, a coffee machine design. How much does lobbying your boss matter for whether you get promoted or your bonus? At the more political companies, lobbying can have a lot of effect. Individual bosses have a lot of power, or a lot of say over how much their reports get rewarded or promoted.

At other companies, like Google or McKinsey, the manager actually has very little say in how his or her direct reports are compensated. Some independent person flies in and conducts interviews, then flies out and offers a recommendation. When the manager is taken out of the equation like that, there’s much less politics—lobbying your boss for a promotion doesn’t matter if your boss isn’t deciding on promotions. So return on politics is a measure of exactly that, “How much, on average, does politics matter in a company?”

Now, “How much does skill matter?” is a measure of, “How well are employees matched with their projects?” Let’s say your job is to design coffee machines, and like me, you’re aesthetically challenged. I would not be a very good coffee machine designer. So if I’m assigned to the project of making coffee machines, and I’ve got one more hour in my day, am I going to spend that last hour on my coffee machine project, or am I going to spend it trying to schmooze my boss, and convince him that my work is good and my neighbor’s work is bad? I’m not a very good designer, so if I work another hour on my coffee machine, I’ll still make the same lousy product—it’s just not worth my time. I might as well be investing my time in politics. So companies where employee skills are not well-suited to their job—either they’re not very well-trained, or they haven’t been optimized—will be more political.

[Ideally,] employees are stretched just enough, like a string that’s not too tight, not too loose. You don’t want to put Frank Lloyd Wright on that coffee machine project. Because after a few hours, he’ll be done. He’ll have made a home-run, Museum-of-Modern-Art coffee machine, and then what’s he going to do? Nothing. He might as well walk the halls, and talk about great he is, and shoot down his colleagues so that he gets the next big promotion. That’s why the best organizations have the lowest return on politics, and the highest amount of fit.

Whitney: Now, if someone wants to get this right, which lever would you suggest people start with?

Safi: Thinking about incentives is a pretty straightforward one to start with. If you’re paying everybody just based on rank and you’re a large company, you’re encouraging politics. So how can you design systems where everybody’s got more skin in the game? It’s not easy, but nothing about building companies is easy.

I think part of the problem is that we’ve been kind of lazy. We’ve said, “It’s a 1,000-person company. Everybody had a good year, so let’s write everybody a check for 10%. Or we had a bad year, so let’s write everybody a check for zero.” The problem with that is, what’s your motivation if you’re four or five levels down? You really can’t affect the success of the entire company—you can only affect the one little thing that you’re working on. So if you want to start with something, think about designing incentives where what people get out is more connected to what they put in.

Ready to learn more about loonshots? Join the Next Big Idea Club today!

the Next Big Idea App

app-store play-market

Also in Magazine