Don’t try to “find your passion.” Instead, take an optimal learning approach to your career.

Published in

College to Career

6 min readAug 21, 2016

If you are a college student or recent grad confused about your career path, understanding optimal learning can help frame your thinking around how tackle that confusion and begin charting your own course.

“Find your passion” is bad career advice, especially for students.

So many career resources advise people to “find and pursue their passion.” Some example articles of this sort:

I do agree that there is some benefit to this line of thinking — it encourages people to reflect and identify their internal drivers. But based on my own experience, I don’t think “finding your passion” is a constructive mindset to have because…

…it’s too tall of an order, especially for young people with little real world experience. It can paralyze people from taking the first plunge with an open mind by pressuring them to theoretically over-analyze their options in order to identify the “perfect” one.
…passion isn’t static. “Finding your passion” makes it sound like there’s a concrete passion that you should be able to locate. In my view, passion gets cultivated over time as you learn more about yourself and the world. It’s important to keep an open mind rather than be narrowly focused on searching for a “right” answer.
…it’s not actionable. What does “finding your passion” even mean? It’s too abstract.

By the way, I’m not the first one who thinks this way:

**It’s better to take an optimal learning approach to career development.**

In statistics and computational mathematics, there is a concept called optimal learning. It’s about “how to collect information as efficiently as possible, primarily for settings where collecting information is time consuming and expensive.” In a nutshell, optimal learning optimizes the tradeoff between the exploration of the unknowns (expending resources today to gain new information that can potentially enable better decisions in the future) and the exploitation of the better knowns (cashing in on options that have higher expected benefits).

There is an elegant solution concept in optimal learning, called the knowledge gradient. It says to choose an option with the highest expected marginal value of a measurement in terms of the value of the information gained by the measurement. That is, the knowledge gradient does not identify the choice simply based on the expected value; rather, it identifies the choice which will do the most to help eventually identify the best choice, based on the expected value as well as the uncertainty around the different possible choices.

Illustration (source):

If we have five alternatives (as shown below) with different levels of uncertainty about each alternative, we want to evaluate the alternative that offers the greatest chance of improving the final solution. So alternative 2 may be much more attractive to evaluate than alternatives 3 and 4.

(For more info on optimal learning, check out this introductory paper by Professor Warren Powell at Princeton).

I have found optimal learning to be highly applicable to career navigation for the following reasons:

There is an overwhelming number of different career options that a typical college student can consider
There are many dimensions to think about when deciding on a career path
Every attempt at collecting information (i.e., trying out a job) is costly, so you want to be thoughtful about maximizing the efficiency of your search

So how would you go about applying optimal learning to your own career? Here are the main steps:

1. Establish your “priors” (i.e., your initial sense of what you like and don’t like)

Below are some dimensions that I think students should think about.

Problem space. What problems energize you? (K-12 education? Economic development in emerging markets? Poverty alleviation? Climate change?)
Structure vs. flexibility. Are you someone who likes hierarchy and defined ways of working? Or are you someone who likes flexible engagements? Do you value routine and stability, or excitement that comes from unpredictable changes? Thinking about your daily habits (or lack thereof) and about how you manage your time might give you a clue about your preferences.
Independence vs. collaboration. Do you love thinking deeply about problems by yourself and working into the wee hours of the night at your own pace? Or do you get energized by working alongside teammates?
Geography/mobility. Do you have a particular geographical preference? What about with regards to travel?
Key extrinsic drivers. What extrinsic factors are important to you? Money? Prestige? Influence?

Note that across almost all of these dimensions, you may find yourself simply guessing what you might prefer. You might find that you are particularly excited about one or two dimensions and have no distinct views over the others. If you are like me when I was back in college, your guesses are likely pretty far from what you will actually end up preferring. But that’s totally OK and normal. What is important is staying openminded to trying things out.

Don’t get stressed out and spend months overthinking this step. Embrace the uncertainty and enjoy the start of an exciting, life-long journey of self discovery.

2. To start, look for a job (internship or full time) that best aligns with your priors and gives you great exit opportunities.

Once you have your priors set, look for a job that seems to best match your preferences and gets you excited. For some of you, you might be able to use your summer internship opportunity to try out a real job. For others, you might have spent summers pursuing non-professional endeavors and the full-time gig out of college will be your first real job.

Some of you might ask — Doesn’t optimal learning tell you to seek uncertainty? To apply the framework, shouldn’t I look for a job that I know absolutely nothing about? In my opinion, you will learn a ton in your first job no matter what because the uncertainties are actually very high regardless of what job you take. So at this point, I would focus on finding something that excites you rather than worrying too much about explicitly seeking uncertainty.

Also, it’s really important that you look for a job that will set you up for strong exit opportunities so that when you do (inevitably) decide to explore different opportunities, you can do so with ease. This means you should prioritize jobs in which you will learn a lot, find strong mentors who will vouch for you, and elevate your professional brand so that you can land future job interviews.

3. On an ongoing basis, evaluate your satisfaction across the different job dimensions, actively learn about your preferences, and articulate key uncertainties you still have about your career path.

While you are working, take time regularly to examine the various dimensions of your job and think about what excites you and what frustrates you. It’s important to be specific. For example, rather than simply saying, “I don’t like this job because I don’t like my boss,” it’s important to understand exactly what it is about your boss and the surrounding team structure that’s causing you frustration. Also, it’s important to separate one-off nuisances (which are unavoidable in any job) from fundamental, structural sources of issues.

As you learn about your preferences across the different dimensions, start translating your learnings into questions/uncertainties you want to answer about your career direction.

4. Try new roles that can address your uncertainties

Once your learning/excitement for your current role has flattened and you don’t think staying on the current trajectory can boost your growth, don’t be afraid to switch to a new role (either within or outside of your current company) with a high knowledge gradient — i.e., a role in which you believe you can better push the boundaries of your understanding around your career path.

Optimal learning likely sounds pretty abstract at this point. If so, check out this story where I talk about how I’ve applied this concept to my own career — from Wall Street to McKinsey to Tanzania to Colombia.