On Business Taste

SpaceX went public these past couple of days, and many of its old stories have been trending.

I saw that back in 2006, when it was on the verge of bankruptcy, it received a life-saving investment from Peter Thiel—money that turned into more than 50 billion dollars on the day SpaceX went public. I think Peter Thiel’s taste in investing and business is remarkably distinctive, and the fact that he has stuck to his taste consistently over all these years and achieved such enormous success is genuinely rare.

Mimetic Desire

This taste of Peter Thiel’s stems from René Girard’s theory of “mimetic desire.” The theory holds that between the subject and the object of desire, there is often a “mediator.” We desire something not because of the object itself, but because we want to imitate this mediator, and that is what gives rise to the desire.

This theory of “mimetic desire” not only inspired Peter Thiel’s investment in Facebook; it also includes a category called “competitive mimicry”: when you imitate that mediator, the two sides eventually grow gradually homogeneous, sinking into an extremely brutal state of competition.

This is what inspired Peter Thiel’s pursuit of “anti-competition.” So he later spent years championing a business philosophy: we should pursue monopoly rather than fall into homogeneous competition. And the best path to monopoly is innovation. His book Zero to One is all about encouraging us to do innovative, zero-to-one things, rather than going from 1 to 100 through simple copying and competition.

From Zero to One

Zero to One is probably one of the books on my shelf that I’ve read the most—I’ve flipped through it at least three times all told, sometimes the e-book version, sometimes the print version.

The two things that left the deepest impression on me are these:

1. If you’re going to start a company, you should begin in the smallest possible domain, monopolize that domain, and only then consider other things.
2. If you get caught in competition, you’ll inevitably fall into a price war and forever struggle on the edge of survival (in a perfectly competitive market, profits all trend toward 0). This state is always especially painful, and it is not a good business model.

A good business model has a landmark endpoint, rather than forcing you to maintain the status quo with no end in sight. It should be like launching a rocket: at the start you have to work incredibly, incredibly, incredibly hard. But once you exceed the first cosmic velocity and reach outer space, you can just keep drifting up there. You don’t need to keep pushing yourself forward with fuel like an airplane, or keep flapping your wings nonstop like a bird. As long as you “floor the gas” past that threshold of the first cosmic velocity, you no longer need to keep maintaining that high-intensity state afterward. You’ll drift up there on your own, and stay there with ease.

In Zero to One, four traits of monopoly are mentioned:

1. Proprietary technology. Your technology has to be ten times better than the competing products or substitutes along some very important dimension—only then can it form a real barrier; being just a little bit better isn’t enough. Excellent technological taste is that first cosmic velocity: as long as you work hard to cross that point, everything afterward gets easier and easier.
2. Network effects. The more users you have, the larger your marginal returns become, continually reinforcing the entire network. The most typical example is social media. And this kind of business model can only succeed if it starts from a very small market—for instance, Facebook started at Harvard. Once your penetration within a certain group reaches a certain proportion, the network effect grows stronger and stronger, forming something like gravity that makes subsequent expansion very easy.
3. Economies of scale. A good monopoly business gets stronger as it gets bigger; its marginal cost approaches zero. Software and content are classic examples: once it’s built, the marginal cost of distribution is nearly zero, and it can expand without limit. By contrast, many service industries (like massage or hair-washing) have no way to scale up, because their marginal costs remain high and don’t decrease as scale increases.
4. Brand. A strong brand can also constitute a monopoly—Apple and Moutai are classic examples. Once a brand is established, its moat becomes very strong. This includes the recent case of Pop Mart, which has actually gradually formed its own moat as well. Likewise, once penetration reaches a certain proportion, things become easy.

In recent years it’s been popular to say “do the hard, right things,” but I think there should be one more clause: do the hard, right things, and once you cross a certain critical point, things will get easier and easier.

Warren Buffett and Duan Yongping

On the subject of taste in business models, I think of two extraordinarily impressive value-investing masters: one is Warren Buffett, the other is Duan Yongping.

I’ve read a lot of what they’ve said. Their judgment of a good company almost always follows a process of their own, but the first step is invariably to look at the business model first. In fact, whether a business actually makes money is already decided the moment its business model is laid out.

The effort that comes afterward is more like the “0”s after the “1.” There are plenty of hardworking people, but what truly determines whether a business is easy to run, how grueling it is, and how much money it makes—ultimately, it always comes down to its business model. It’s like the most fundamental model, and the effort and various tactical maneuvers that follow are really just embellishments on the skeleton, making it better.

Besides the business model, they also consider “company culture.” The way I understand it, company culture actually determines how long this model can last. If a good business model ends up drifting far from its culture and values, it cannot sustain itself over the long term.

So the business model and company culture form two dimensions:

1. The business model determines its shape;
2. The company culture determines how long that shape can be sustained.

If both are satisfied, it’s essentially what Duan Yongping often calls a “long slope with thick snow” track.

In short, combining this with the earlier discussion of Peter Thiel, I’ve come to realize that taste in business models is the single most important thing, whether for investing or for starting a company. Because it determines how deeply you understand things. If the business model is bad from the start, then no amount of subsequent effort may be anything more than wasted motion.