Careers
Choosing the game, not just the outcome
A salary pays for your time, while ownership pays for what your work keeps producing. Before choosing a career by its expected income, decide which of those games you want to play.
- A high salary and a high-leverage career are not the same thing. The difference is how closely your earnings are tied to your own time.
- Ownership lets you capture value from something that scales beyond your own labor. But uncapped upside does not promise a good financial outcome.
- A career choice is not only a bet on expected income. It also decides how much uncertainty you live with and which problems fill your days.
- Being recognized as exceptional is different from building something exceptionally valuable.
- Don’t start a company for its own sake. Find a problem worth solving first, then a way of solving it that can grow beyond what one person could do alone.
General pediatricians in the United States earned about $274,000 on average in 2025, according to Doximity’s 2026 Physician Compensation Report.
That’s a lot of money by almost any standard, and it takes years of medical school and residency to earn. The work matters to patients and families, and the career is far more predictable than a startup founder’s. By most measures, medicine is an excellent career.
Yet the size of a salary is the wrong thing to compare. What matters more is the structure of the career that produces it.
A physician’s pay rises with experience. But a physician who mainly sees patients is limited by time. There are only so many patients one person can see in a day.
Physicians can also own their practice, so the real line isn’t between doctors and entrepreneurs. It’s between earning mainly through your own labor and owning something that creates value beyond it.
That raises a larger question. What if the most predictable path to success leads to a kind of success you don’t actually want?
Income and wealth are not the same thing
Naval Ravikant’s How to Get Rich (without getting lucky) draws the clearest version of this line. Naval separates earning money by renting out your time from building wealth through ownership and leverage.
Imagine two people. The first earns $300,000 a year providing a professional service. The second owns a large stake in a software company whose product is used by thousands of organizations.
In any given year the first person may well earn more. They may also have more financial security, and they can build real wealth by saving and investing. But their professional income still depends on their continuing to work.
The second person’s company can earn revenue from thousands of customers without the founder serving each one. The founder’s payoff is a share of what the business is worth, if that share can ever be sold.
This is leverage. The same hour of work can serve one customer or thousands.
Software is the clearest case, because serving one more user costs almost nothing. Code written once can run millions of times, and a small team can build something used around the world. A book or a video works the same way.
Leverage lets output grow faster than effort. Ownership decides how much of that output comes back to you. Creating value and capturing value are two different problems.
An employee can do much of the work that makes a company succeed and be paid only a salary. An investor can own a stake that grows in value without working there at all. That says nothing about whose work matters more. They have different contracts.
So look at how a career pays you, not only how much. Does your income stop when your hours stop? Will any part of what you build keep earning after you move on? Neither answer is wrong, but you should know which one you are choosing.
Uncapped upside is not a better bet
Many professions have a well-defined path. You finish the training, then climb a ladder that institutions built long before you arrived. There is plenty of room for excellence on that path, and a great teacher can matter far more than their salary suggests.
Starting a company works differently. There is no residency for founders and no promotion schedule, and no institution decides whether you’re qualified to try. You find a problem and try to convince people that your solution is worth paying for.
In principle the upside is uncapped. What you build might reach a million people. It might also reach nobody.
Count the odds of failure and the years of salary you give up while you try. A startup with enormous upside can still have a poor expected financial return. We also hear mostly about the founders who succeeded. Companies that never find customers or never raise money rarely get written about.
So the case for starting a company can’t be that it pays better than a conventional career. For many people, it doesn’t.
The better argument is that people can rationally prefer different distributions of outcomes. Some want a stable career with a clear trajectory. Others will accept much more uncertainty for a chance at an extraordinary result. Neither preference is wrong, but they lead to very different careers. Work out which one you hold before you compare salaries.
Autonomy is another form of compensation
Salary comparisons also leave out freedom, in particular the freedom to try an idea without waiting for an institution to approve it.
Academic research follows the grants. Medicine runs on standards of care, most of which exist to protect patients. Employees at large companies work within a strategy that someone else set.
Starting a company doesn’t remove constraints. Founders still answer to investors, and in the end to whether customers pay. A founder who can’t find customers is no freer of outside judgment than a researcher waiting on a grant review. In some ways a founder carries more obligations than an employee. Employees and customers depend on the founder’s decisions, and investors often hold contractual rights over the major ones.
So the freedom isn’t absolute. The difference is that you get more say in which constraints you take on. You decide which problem is worth pursuing, and you can change direction when customers tell you it isn’t working.
If autonomy matters to you, count it as part of the pay. A lower expected income with more control over your work can be the better offer.
Recognition is not the same as value
Ambition has an uncomfortable side. It’s easy to confuse wanting to build something extraordinary with wanting to be seen as extraordinary. From the outside, the two look the same.
A paper in an elite journal can be a real achievement. It can also become a stand-in for one. An academic can be highly cited for work that changes little in practice. A startup can raise a lot of money and get plenty of press without ever becoming a sustainable business.
The reverse happens too. Someone can spend years on an obscure but essential technology that improves thousands of lives and never be publicly recognized.
A useful test is to ask which you would pick, a little-known company that improves people’s lives or wide recognition for work that matters less. Recognition and wealth are fine things to want. They hold up better when they follow from building something valuable than when they replace it. The goal is to create exceptional value, not to look exceptional.
Economic value and social value also don’t always line up. A product can sell well without making anyone better off, and some of the most important problems in healthcare and science are hard to make money from. So financial upside can’t be the only test for what to build. Weigh it against what kind of value you would be creating.
The gap between a discovery and a product
Scientists have a particular version of this choice. Research shows how much work sits between a discovery and anything that changes a patient’s care. I’ve seen it from human genetics and computational biology, and now from AI. A promising result in a paper is not a diagnostic test. A model that scores well on a benchmark has not yet improved anyone’s treatment.
Those gaps are hard problems, and they are also opportunities. Someone who understands both the science and what it takes to deploy it in a clinic could build something valuable. That takes different work from research, though. You have to find out whether anyone would pay for it, which no paper ever has to show.
Starting a company isn’t more impressive than doing science. The reason to try is to find out whether what you know can become something that scales, and to carry useful knowledge out of papers and into practice.
Find the problem before the company
An obsession with leverage can put things in the wrong order. You start by wanting to own a large company, then go looking for something to build.
Having investors doesn’t make a company valuable. A company is valuable when it solves a problem someone cares about, in a way that pays for itself. Half of a company nobody wants is worth very little. A small stake in a lasting business can be worth a great deal. So the question isn’t how much equity you can own. It’s what valuable thing you can help build, and what ownership makes sense for building it.
That makes the first step smaller than founding anything. Find a problem you believe is important and learn it unusually well.
- Talk to the people who have the problem, and watch how they handle it today.
- Build the roughest prototype that could prove you wrong.
- Ask what they use now and what would make them switch.
- Be ready to learn that an idea you loved doesn’t work.
Don’t spend years defending a hypothesis just because it was yours. If the evidence changes, change with it.
In a previous essay I wrote about optimism in the face of uncertainty, and about choosing experiments for what they teach you. The same rule applies here.
Find the problem before the company. Don’t make the biggest bet first. Run small experiments that teach you something even when they fail, and commit more only as the evidence comes in.
Choosing the game
No career is right for everyone. The path that feels confining to one person gives another exactly the stability and meaning they want. Starting a company, whatever its upside, is a bad choice for someone who values predictability or doesn’t want to run a business.
It isn’t the only way to build wealth either. A high salary invested steadily over decades can buy real financial independence, and many professionals do enormous good through their work without owning a company.
But a career is more than a bet on expected income. It’s how you spend a large part of your life.
- Which problems do you want to spend your time on?
- How much uncertainty and responsibility are you willing to carry?
- How much does autonomy matter to you?
- What outcome would make the effort worth it?
A high salary buys security and, in time, freedom. Both are worth a lot. Ownership offers something different, which is value that isn’t capped by a pay scale and can grow beyond what you contribute yourself. My own answers to those questions point toward ownership. Yours may point elsewhere, and that is just as reasonable.
Whichever game you choose, choose it on purpose. If it’s ownership, start with the unglamorous part. Talk to the people who have the problem, and find out whether any of them would pay to solve it. That will tell you more than any salary table.
Further reading
- Naval Ravikant, How to Get Rich (without getting lucky), on why renting out your time won’t make you rich.
- Doximity, 2026 Physician Compensation Report, on pay across medical specialties.
- Paul Graham, How to Get Startup Ideas, on finding startup ideas through firsthand knowledge of a problem.
- Chen-Yang Su, Optimism in the face of uncertainty, on exploration and choosing experiments for what they teach you.