Quick tips
- Write three candidates, never just one.
- Run each one for two weeks before you choose.
- Pick the one that compounds with what you do.
There are two tabs open. One is a language course you have wanted for years, the other is a woodworking class ten minutes from your house, and both of them are real. You have the money, you have found roughly five hours a week, and you are ready to put a serious stretch of your life into being genuinely good at something.
The only thing in the way is that you cannot pick, and while you are not picking, the five hours go nowhere. This is a good problem and it deserves a procedure rather than a coin toss. Choosing well is not about finding a single true calling. It is about picking the bet with the best compounding, running a short live test before you commit anything you would miss, and then closing the question so your attention stops leaking into the tab you did not choose.
How do I pick one skill when three of them interest me?
Write down three candidates, run each one for two weeks at the hours you actually have, and commit to the one you kept opening without having to negotiate with yourself. Two weeks is long enough for the novelty to wear off and short enough that you have sunk nothing you mind losing.
Three is the right number for a reason. One candidate gives you nothing to compare against, so you cannot tell enthusiasm from fit, and you will keep wondering about the road you did not take for a year. Five or six turns the choosing itself into the hobby, which is how this decision gets avoided for a decade.
While you run the trials, log the wrong thing on purpose. Do not log enjoyment, because everything is enjoyable in week one. Log whether you opened it on the day you were tired, whether you thought about it while doing something else, and how long it took to start once you sat down. Those three tell you far more about the next two years than how good the first session felt.
What makes a skill worth two years rather than two months?
Three tests, and the skill worth two years passes all three: it compounds with something you already do, you can get real feedback on it inside a month, and its core will outlive its current tools.
Compounding is the one people skip. A skill that sits next to your existing work pays twice over one that sits on its own, because every hour you spend on it makes an hour of the other thing better and you get chances to use it without arranging anything. If you write for a living, learning to draw diagrams compounds. Learning the guitar does not, which is fine as long as you are choosing it knowing that.
Feedback is the second test, and it decides how fast hours convert into skill. If nobody can tell you within a month whether you are getting better, you will spend two years practicing a mistake with perfect discipline. Look for a domain with a scoreboard: a recording you can listen back to, a joint that either closes or does not, a person who will read the page and tell you where they got bored.
The third test is durability, and it has a number on it. David Deming and Kadeem Noray, working with US career data, found that the initially high economic return to applied technical degrees falls by more than half in the first decade of working life, because the specific technical content keeps changing under the people who learned it. That is not an argument against learning tools. It is an argument for choosing a skill whose core sits underneath the tools.
How do I test a skill in two weeks without wasting the time?
Pick one narrow real output and ship it inside the fortnight. Not a course and not a chapter of theory: one small thing that exists at the end and that somebody else can look at.
Two minutes of recorded playing. One stool. Ten minutes of conversation with a speaker who will correct you. One page of writing about something you actually know. The output matters because it moves you past the part of every new skill that feels like reading, and it is the reading part that fools people into thinking they have chosen well.
Keep a three line log each night: minutes, what you did, and whether you had to talk yourself into starting. At the end of two weeks the log decides, not your enthusiasm on the last evening. And run the trials one at a time rather than in parallel, because side by side comparison rewards whichever one you happened to have more energy for that week.
Does it matter whether I have any natural talent for it?
Less than the domain matters. A meta-analysis by Brooke Macnamara, David Hambrick and Frederick Oswald found that deliberate practice accounted for roughly 26 percent of the differences in performance in games, 21 percent in music and 18 percent in sports, and much less in education and in professional work.
Read that as a fact about domains rather than a verdict on you. Where the task is stable and the feedback is immediate, hours convert into skill at a high rate, and those are exactly the places where somebody starting from nothing can move fast. Where the feedback is slow and the environment keeps shifting, hours still count, but they need better design and more outside eyes to pay off. So do not choose on talent, choose on feedback speed.
What if I choose wrong?
You lose a few months, not the two years, and far less than that once you count what transfers. Most of what you build in the first months of any serious skill is general: how to run a practice session, how to sit with being bad at something in front of a competent person, how to break a big thing into a small thing you can do on a Tuesday.
The expensive mistake is not choosing wrong. It is re-choosing every eleven weeks. Angela Duckworth and colleagues studied what predicted people finishing genuinely hard things, across West Point cadets, undergraduates and national spelling bee competitors, and what predicted it was grit, which they defined as perseverance and passion for long-term goals and which did not track IQ. Read that against your own decision and the useful part is what it does not say. Nobody in that research was distinguished by picking flawlessly.
Which means a merely good choice, held, beats a perfect choice that you keep reopening. Set a review date instead of an escape hatch. Two years from the start, look at it honestly and decide again. Between now and then the question is closed, and closing it is most of what you are buying with this whole procedure.
Commit out loud, then stop shopping
Once the trials are done, make the decision expensive to reverse in the cheapest way available. Say it out loud to one person who will ask you about it. Write the skill on something you see. Put the hours in the calendar with the skill's real name on them rather than a vague block, and close the other tabs.
This is not a smaller life. Choosing one skill is how you concentrate force, and in ten years the difference between genuinely good at one thing and pleasantly acquainted with three is not a small difference, it is a different life. The two candidates you did not pick are not gone. They are queued, and you will come to them with a practice habit already built, which is a better way to arrive than you would have managed this year anyway.
So pick the one you kept opening, give it the hours you already found, and let it get hard. Being unsure was never the problem. Being unallocated was.
Sources
- Psychological Science, Deliberate practice and performance in music, games, sports, education, and professions: a meta-analysis
- Journal of Personality and Social Psychology, Grit: perseverance and passion for long-term goals
- National Bureau of Economic Research, STEM Careers and the Changing Skill Requirements of Work