How to get good at something useful: Part 7 Career capital: how to invest in yourself

Kate wanted to make a difference, so she did the obvious thing — she went to work at a nonprofit straight out of university. However, she quickly hit a ceiling in terms of how far she could advance.

When we talk to leaders in the nonprofit sector, they often recommend getting trained up elsewhere first. Kate ended up returning to the corporate sector for several years, and thinks she would have ended up ahead if she’d done that straight away. We know many other people who feel like they wasted years of their career.

Don’t make these mistakes. Although it’s great to make a difference right away, you also need to invest in yourself to maximise your impact and fulfilment in the long term. This requires building what we call career capital: skills, connections, and credentials that put you in a better position to make a difference in the future.

However, it’s also possible to make the opposite mistake: generically accumulating credentials for years when it would have been possible to do something meaningful much sooner. Given the urgency of the world’s problems, and how rapidly the world could change due to AI, you need to balance these two types of error.

Here we’ll cover three common mistakes people make when building their careers, and explain how to strike the right balance between investing in yourself and doing your dream job as soon as possible.

Reading time: 10 minutes

Mistake 1: Failing to invest in yourself

People like to lionise the Mozarts, Malala Yousafzais, and Mark Zuckerbergs of the world: those who achieved great success while young. There are all sorts of awards for young leaders, like the Forbes 30 Under 30 and Time‘s Next Generation Leaders.

But these stories are interesting precisely because they’re the exception. Most people reach the peak of their impact in middle age. Income usually peaks in your forties or fifties — on average at more than double what it was at age 25 — suggesting that it takes around 20 years for most people to reach peak productivity.¹

Similarly, expert researchers and creatives tend to peak somewhere between the age of 30 and 60, producing several times as many important works per year as they did age 25.² The peak is earlier in more conceptual fields, and later in those that are more experimental or require accumulated knowledge. If anything, the age of peak output seems to be increasing over time.³

Even among software startups — famous for their young founders — if you look across the entire industry, the average age at which people start companies is 40,⁴ and older founders are about twice as likely to succeed.⁵ In part this is about skills, but it’s also about accumulating connections. That’s why fields like politics and management take longest to peak. It’s rare for anyone to make it to Congress in their twenties, and your odds are over four times higher in your sixties than your thirties.⁶

Here’s a rough estimate of the age of peak output in a variety of fields:⁷

The study of expert performance was pioneered by late Swedish psychologist K. Anders Ericsson. After 30 years of research, he concluded:

I have never found a convincing case for anyone developing extraordinary abilities without intense, extended practice.⁸

For Mozart to succeed so young, he needed to start young. Mozart’s father was a famous music teacher and trained him intensely from the time he was a toddler.

This may sound like a bit of a downer: becoming skilled takes time, especially in established fields. But consider the flip side: it also means you can become much more skilled than you are today.

Lots of people come to us and say, “I’m not sure I have any useful skills to contribute.” That’s often somewhat true — especially if you’ve just graduated. More than likely, you’ve spent the last four years studying Moby Dick, metaphysics, or Molière, and your future job is unlikely to involve any of those.

But Ericsson’s research also suggests that anyone can improve at most things with focused practice. Sure, other factors are important too — if you’re seven feet tall it’s going to be a lot easier to get good at basketball — but that doesn’t mean short people can’t improve their game.

In fact, it’s likely there are ways you can improve a lot, especially if you can find something that matches your talents. We’ve already seen that output rises 2–5 times as people advance in a career path. But you can increase your ability to contribute even more by opening up career paths that are entirely new.

Imagine you’re stuck in a path you don’t think is very impactful, but by doing a master’s you could open up research jobs you think contribute far more. Sometimes the right job, crash course, or connections can increase your impact many times over.

Among people we advise, we’ve seen lots of examples where people become far more successful, happy, and capable by investing in themselves, often in areas they never thought they could succeed in.

Despite the high returns that come from investing in yourself, we also meet many people who want to skip it to do something important and meaningful as soon as possible. I was the same when I first graduated — I wanted to do whatever seemed most likely to have an impact in the next few years.

But as we saw with Kate, that can mean getting stuck in a path where it’s hard to advance. Treading water for a couple of years doesn’t only mean you lose those years, it also delays when you reach your peak impact. So, the first key point is to compare the next job you could take both in terms of its immediate value, and in terms of how much it’ll improve your career capital.

Mistake 2: Not figuring out what will advance you most efficiently

After finishing a psychology master’s in Amsterdam, and despite a nagging concern it wasn’t right for her, Lauren Kuhns continued into a PhD in the same subject. And indeed, she didn’t enjoy it, leaving academia as soon as it was finished. To make things worse, she then discovered the PhD wasn’t especially helpful in getting her jobs she was interested in.

We’ve advised a lot of people who have started a PhD and gone on to realise they hate academia and that their PhD won’t help much when pursuing other options. They end up feeling like they’ve wasted several years.

It’s common to get sucked into other generally prestigious, ‘safe’ options, most classically things like consulting, finance, law, professional services, and even socially-focused programmes like Teach First or Teach for America. A good test is to imagine you can never tell anyone you’ve done the job. Would you still want to do it? If not, you’re probably pursuing it for prestige, rather than the skills you will actually learn.

Often these kinds of options are really just an expensive form of procrastination, because there are usually much more direct routes into the most impactful careers. For instance, we often come across people who already have a graduate degree, or a lot of work experience, and want to transfer into policy. But they mistakenly think they need to get another degree, this time in policy, before entering.

Instead it’s usually possible to transition after several months of well-targeted reading and networking by applying directly to fellowships and other positions. Rashida Polk was a nurse for eight years before she got inspired to work on pandemic prevention. She was able to join the Horizon Fellowship, which helps people transition into technology policy. In under two years, she was able to work on committee hearings on the origins of COVID-19 and oversight of high-risk virus research.¹⁰

Ask yourself, for the options you’re actually interested in, which steps will get you there most efficiently? For the global problems you think matter most, what do they most badly need, and how could you help with those needs the fastest?

The best way to figure this out is to speak to people a few steps ahead of you in the field. Ask them what you could do over the next year that would enable you to contribute most. If you can speak to people who themselves advanced unusually quickly, and ask how they did it, that’s even better.

Usually this means doing something as close as possible to actually doing the work, rather than generic skill-building. Fresh out of college, Daniel Ziegler knew how to code, and wanted to do research on technical AI safety. Instead of finding a normal software engineering job, he emailed Dario Amodei — at the time a leading AI researcher at OpenAI — about how he could most effectively enter the field.

Based on their conversation, Daniel spent six weeks studying deep reinforcement learning intensively with a housemate. They read 20–30 key papers together and practised implementing core algorithms. At the end of the sprint, he had enough knowledge, and evident determination, to start a research engineering position on OpenAI’s safety team, and has since advanced from there, publishing several papers in the field.¹¹

Daniel’s story illustrates principles anyone can apply: he identified the skills most essential to the role (real-world ML engineering), devised an efficient way to learn them (co-working with a housemate), and found a way to demonstrate his skills to employers (GitHub projects). By doing this, he was able to accelerate his career by at least several years.

What about if you’re not yet sure where you’d like to end up in the long term? Then ask yourself, “Where could I maximise my overall rate of learning?” Any position where you’ll learn a tonne can be a good call, even if you’re not sure where it’ll lead. It’s usually better to be in a random organisation with an amazing mentor and co-workers than at a fancy graduate school with a supervisor who hates students.

Mistake 3: Waiting too long to have an impact

It’s rare to have a big impact right out of college, so at the very start of your career, gaining career capital should (normally) be your biggest priority. However, it’s also possible to make the opposite mistake: accumulating generic credentials or lingering in your corporate job for years when you could have done something meaningful much sooner.

Consider two options:

1. Work at a nonprofit for two years.
2. Work at Deloitte for two years, then switch to a nonprofit.

You should only do the second if you’ll end up in a significantly more senior nonprofit role as a result. Otherwise, you’ve missed out on two years of impact for nothing.

People often fail to appreciate that the career capital they’ll gain from the impactful option can sometimes be better than the safe corporate path. By working at the nonprofit, you’ll learn more about how to tackle an important global problem and make connections with others who want to do the same, rather than with random accountants.

But let’s suppose you believe the Deloitte option will in fact help you build better career capital than the more impactful job. Should you take the detour?

The answer depends on your situation. If you’re facing this tradeoff, here are two key questions you can ask yourself.

First, how long is your time horizon? The earlier you are in your career, the more time you have to use your career capital in the future. Learn public speaking at age 22 and you can give talks for decades. Learn it one year before retirement, and you’ll be the most charismatic person at Bingo night.

Not only do younger people have more time ahead of them, people’s work skills usually increase the most from about 25–35, then slow down as they approach their peak.¹² This suggests the first few investments in career capital pay off the most, and the returns diminish once you’re already experienced.

Your time horizon can also be shorter depending on issues you think are most pressing. The best opportunities to help in global health or climate change are gradually disappearing as more progress is made, making it important to help sooner than later.¹³

However, the rate they’re disappearing is relatively slow, so if the world remains relatively normal, it could easily be justified to spend a decade investing in yourself in order to contribute on a greater scale later.

But if you think everything is going to change due to AI in the next five years, and that will create never-to-be-repeated opportunities for impact, then your time horizon could be a lot shorter. And that means your bar for delaying your impact will need to be a lot higher.¹⁴

The second question is simply how much career capital do you expect to gain from taking the detour compared to the immediately impactful path? If you can put yourself in a far better position, that will often be the deciding factor. But if the potential career capital is around equal, you should choose the impactful option instead.

In other cases, it will be unclear which side wins. But note that even if you have a very short five-year time horizon, it can still be worth a delay. If you can spend one year to make yourself 30% more productive for the next four, then, roughly speaking, you’ll come out ahead. For example, people who finish college earn around 50% more than high school graduates.¹⁵ Income isn’t a perfect measure, but it definitely suggests that finishing college increases your career capital significantly.

In the case of AI specifically, it’s likely that the most consequential moments will come when truly advanced systems are deployed in society, perhaps irreversibly, or key treaties are signed. Therefore your focus should be on getting to the best possible position to be useful in those transitions before they happen. Additionally, you shouldn’t be certain AGI will arrive soon, so you still need to consider the other scenarios where your time horizon is longer.

A final note: if you do decide to do something that temporarily disconnects you from the impact you want to make, try to stay involved in other, small ways, such as by attending conferences, donating 1% of your income, or volunteering. Otherwise, it can be all too easy to drift, losing sight of the good intentions you had at the start of your career.

How can you get the best possible career capital?

Researchers disagree about what’s most crucial for developing valuable expertise, but if you read what they say carefully, they pretty much all agree that the following are all important:

1. Talent match. You’ll be able to learn some skills much faster than others, and reach higher peaks of performance. This is so important it’s the subject of all of part 10.

2. Learning environment. Ericsson finds that almost every expert had a great mentor, and also emphasises the importance of clear feedback.

3. Time spent practising. Your skills improve over time, especially in the first 10 years learning them.

4. Being in the right place at the right time.

The last point is also important. My career was totally shaped by meeting Toby Ord in 2009 when I (Benjamin) was an undergraduate. I discovered Toby because I used to go to a lot of talks outside of my course. In hindsight, that greatly increased my chances of coming across someone interesting.

High-flyers with unusual success have often got lucky somewhere in their story. They found a company, scene, industry, or technology that was taking off and got in early. This gave them valuable skills and connections that might otherwise have taken much longer to build.

Although luck isn’t something we control, there are steps you can take to increase your chances of getting lucky. Ask yourself the following:

• Who are the most talented people you know? What are they into?
• What areas are new and rapidly growing?
• Have you found a project or idea that feels like it has natural momentum? One that keeps succeeding more than you expect?
• What are people unfairly biased against working on?

All of these can be signs to work on an area, even if you’re not sure where it’ll lead. In contrast, if you go into an established field like law, academia, or the arts, then you’re going to be in for a long slog competing against many other qualified people.

In order to gain valuable skills, then, you want to find a skill that matches your talents, an environment that helps you learn rapidly, and conditions that maximise your chances of getting lucky.

Once you have some skills, you can use them to notch up achievements to impress employers, to earn and save money, and to make connections with inspiring people of good character — giving you the other aspects of career capital. In short, the goal is to get good, and get known.

But before you start building skills, which skills are actually worth having?

Take a break

Read next:  Part 8: Which skills will be most valuable in the future?

Continue →

Or see an overview of the whole guide.

Get the whole guide as a book

If you find this guide helpful, preordering the book (especially from a physical retailer!) is a great way to support us.

Preorder the book