Have you ever wanted to learn something, but weren’t sure where to start? Maybe you want to learn a language, programming or business. Maybe you want the confidence to tackle supposedly “hard” subjects like math, finance or physics. Today I’m going to show you how.
I’m going to describe the process I’ve used to condense a lot of learning into a short period of time. This is the same process I used  to learn MIT’s 4-year computer science curriculum in twelve months, teach myself languages, business and intellectual subjects like physics and psychology.
This article is going to be a bit longer (~3500 words), so you may want to bookmark it for later.
I’m going to focus on the strategy for learning, meaning how you choose to break down a nebulous goal like “learn to speak French” or “understand personal finance” into something concrete and actionable. As much as possible, I’ll try to provide links to specific low-level tactics I use, such as the Feynman technique, visual mnemonics or active recall as well.
This strategy is just one possibility. If you’ve found success with another, by all means, go ahead! I only want to share the method I’ve been honing for years across a variety of different subjects.
The Steps in 2-Minutes
If you’re short on reading time, I’ll summarize the steps for you:
- Take your learning goal, and craft it into a compelling, obsession-worthy mission.
- Find material to learn from, structure it into a flexible curriculum.
- Define feedback mechanisms to constantly direct your future learning efforts and ensure high-intensity, active recall.
- Test and enforce a schedule that is sustainable over the entire lifetime of the project.
- Develop a long-term retention strategy (formal or informal).
There’s a few points that may be different from what you’re used to:
The first is that the learning goal is oriented around a obsessive mission. Many people trying to learn something adopt a haphazard, casual approach. In general, I’ve found this wastes a lot more time and produces lesser results.
The second is that the strategy is defined by high-feedback practice. In a classroom setting, students can be forgiven for neglecting this step because it is already partially provided in the form of assignments and quizzes. When teaching yourself something, it is very easy to slip into learning tasks that are devoid of feedback and so it takes months to realize you’re off course.
Finally, the process is driven by mentally intensive, active learning methods. Although this can be uncomfortable at first, the speed of that results come makes it worthwhile. You can spend months on a slower strategy, get discouraged and give up which could be fixed by going through some initial discomfort but seeing results quickly.
Now, onto the steps…
Step One: Craft an Obsession
Almost anything can be achieved with the right motivation. The motivation you bring to a project forms the foundation for all your efforts. If that foundation is unstable, you don’t have a chance at success even if you use all the “correct” learning techniques.
My approach has been to choose short-term obsessive missions for learning new things. The word obsession usually has a negative connotation, being paired with “dangerous” or “unhealthy”. But obsession can also be a positive force. By structuring your project around a compelling mission, you focus your enthusiasm for the subject (or the rewards it can bring in your life) onto a single target.
The MIT Challenge was a good example of this. I took the vague goal I had of wanting a computer science education, and crafted it so that it would become very interesting to me. Had I instead started off with the aim of “learning a lot about computer science” I doubt I could have accomplished nearly as much in ten years, let alone one.
Your missions don’t need to be as ambitious or all-consuming, however. Even a project that only takes a couple hours a week can still be compelling.
Here are a few ingredients I’ve found helpful for taking a vague goal and crafting a mission you can get excited about:
- Give it a name. Naming your project helps you define it. A name helps you identify the boundaries of what you’re trying to accomplish with this particular mission, and which you aren’t. Having a name also helps you think about the project as a unified whole instead of a random collection of loosely related learning tasks.
- Pick a specific objective. Narrow your ambitions onto something concrete. Instead of just trying to learn a language, have a goal of speaking only in the target language for an entire day, for example.
- Constrain the scope. Instead of just defining what you’d like to accomplish, also define which things are outside of the scope. This doesn’t mean you have to avoid learning anything outside of those constraints, but it helps you prioritize the vague desire many autodidacts have to “learn everything” onto something attainable in a project.
- Hit the challenge sweet spot. The ideal amount of challenge is that it should be hard enough that you aren’t sure whether you’ll be successful, but not so hard that you give up. If you’ve put off learning something because it scares you, try lowering the challenge. If you’ve given up because you’ve been bored before, try increasing the challenge.
Building a compelling mission isn’t too difficult, once you try. The majority of the time people skip this step, in my mind, is because they either don’t realize it’s important, or they falsely convince themselves that there’s no way learning about *insert subject* could be compelling.
Step Two: Build a Flexible Curriculum
The next step is to gather material. The problem is rarely that there isn’t any material available, but that the material can be hard to find or that good material can be drowning in irrelevant or lousy content.
I’ve found it important to choose material from a wider net than others may cast. This way you can shift between resources to meet your goals. Here are some points to look for when trying to find material:
- Open courses and MOOCs (edX , Coursera , Udemy , MIT’s OCW )
- Textbooks (look in the comments here  for good ones)
- Pre-packaged courses
- Book lists (Personal MBA , Good Reads , etc.)
- Videos (YouTube, Khan Academy , Patrick JMT , etc.)
- Journal articles
- Private instruction/tutoring
- Educational software
Depending on the size of your project, you may want to spend a few hours looking for different options. I must note, that with the exception of some MOOCs and pre-packaged courses, you’ll almost always need to draw from multiple sources.
Another piece of advice—don’t let a lack of complete courses bother you. I did two-thirds of the MIT Challenge just using suggested textbooks and minimal guidance from MIT’s OCW, and in most cases the deficiency was negligible. The difficulty is almost always from the subject, not a lack of resources.
Once you’ve identified material, you need to develop a flexible curriculum around it. By flexible, I mean that, unlike school, the curriculum is something that you can modify and adjust depending on your progress.
When I’d go through a class during the MIT Challenge, I’d often have a few resources to choose from: videos, textbook, external tutorials and articles. My curriculum would be to pursue one resource, but use the feedback I was getting to adjust it. After watching videos, for example, I could use the textbook or articles to fill the missing gaps.
This is even more true when your goal isn’t to learn a particular set of knowledge, but to acquire a useful skill. When I’m learning a new programming language, I often go through several different resources, switching whenever my feedback indicates my weaknesses are more easily fixed using a different resource.
The final key with a curriculum is to not get overwhelmed. The purpose of picking out material isn’t to try to cover all of it. Instead, it should be to give you a starting point for structuring your learning efforts. Even if a different resource turns out to be slightly more efficient later, you can adjust.
I’ve found it useful to do most of this step prior to starting my project. For me, gathering material is distracting from the task of actually learning from it. This is why investing a day or two into researching, bookmarking, downloading and purchasing all of the material you might use in advance is so helpful.
Side note: If you’re not sure about a paid resource, check if it has a free trial/money-back offer. Most have free trials, so you can do a pilot with it before committing your money. For those that don’t, used and library options can significantly reduce the cost. I’d often get textbooks for under ten dollars during the MIT Challenge, so cost is rarely the limiting factor.
Step Three: Define Feedback Mechanisms
Feedback is essential to learning. The first reason is because it helps you guide your progress. If you’re failing practice problems or can’t code a simple program, you know you need to adjust your learning methods.
The second reason is that thinking about feedback mechanisms tends to promote efficient learning methods. One feedback tool you might use is practice problems, which has demonstrated effectiveness in increasing long-term retention .
How do you incorporate feedback?
The two most straightforward ways are by producing something or practicing something. Although not guaranteed to provide feedback, if you’re doing either of these as a significant amount of your learning time, you’ll probably be getting feedback.
Combining learning a programming language, for example, with a set of mini-projects where you actually write valid code ensures that you’re getting feedback. Learning about design while building models or illustrations gives you a chance to observe whether the lessons are creating improvement.
Practicing speaking a language with native speakers ensures that all your learning efforts with SRS, audio courses or phrase books is actually helping you speak. Practice problems for math or physics ensure your conceptual understanding is growing.
Which feedback mechanism you use will depend on what resources you have and what the subject is. Even if you can’t pick a perfectly suited feedback mechanism, you can incorporate smaller feedback drills to ensure you’re not completely without feedback. These smaller mechanisms can include: self-quizzing on learned material, writing Feynmans  without reference material or using software like Anki .
The best feedback goes directly toward your project’s mission. If your mission is to perform a skill or speak authoritatively about a topic, then practicing that skill or writing about the subject are ideal feedback mechanisms. If your goal is to have a particular set of knowledge, self-testing and explaining the knowledge to yourself are good mechanisms.
Step Four: Enforce a Schedule
Many self-learners can successfully reach this point in their project, but fail on the next one: actually doing all the work. It’s one thing to tell yourself you’re going to learn about biology or history. It’s another thing to actually execute the curriculum you’ve devised and accomplish the mission.
The first half is in preparation. Without a compelling mission, it’s easy to get bored and quit. Without a curriculum, it’s easy to get lost and give up. Without feedback mechanisms in place, it’s easy to not learn anything at all.
The second half is in establishing a schedule that allows you to follow through with the reading, watching and practicing you need to do. Here are a couple frameworks I’ve found helpful for successfully implementing such a system:
1. The “Every Day” Plan
The first strategy is to do a little bit of work every day. I did this with a friend on a project to learn languages (which I’ll hopefully be sharing more on in the summer). Because of conflicting schedules and the desire to stay at the same pace, we decided to do an hour lesson, in the morning, every day.
In the past, I’ve done similar approaches to book-reading projects. When I want to cover a large swatch of information on a particular domain, I would get several books and devote 30-60 minutes reading them at the same time each day.
The process is simple:
- Define a certain time period, every day, when you’ll do your work. It doesn’t need to be a long time period to be effective.
- Commit to following this time period, without exception, for at least three weeks. The number is arbitrary, but I’ve found that enforcing the habit strictly in the beginning is essential.
The advantage of this strategy is that the effort quickly becomes a habit. This is the approach to use if your project is not going to be full-time and it will require some self-discipline to execute. The other strategies I’ll mention can also be effective, but they have greater risks that you’ll drop the ball when your motivation wanes.
2. The Obsessive Burst
This strategy is one I’ve used on projects which interested me deeply, and were short (in the span of a few weeks). The idea is simply to work on the project during most of your off-hours until it is completed.
This method only works if you’re genuinely motivated enough to pull it off or there is a compelling external reason for such devotion. If you’re rolling your eyes at this possibility, do yourself a favor and opt for strategy #1 instead.
The advantage of this method is that it utilizes your initial motivation. Some projects that could be finished quickly, I opted for this approach because I knew that my motivation would wane after a few weeks and I wanted to see results quickly. The disadvantages are obvious, but in some instances they don’t matter.
3. The Precommitted Schedule
A final strategy I’ll mention is simply to precommit to a certain goal, or certain hours. If you’re making use of tutoring or outside help, simply committing to your tutor to have finished an amount by each lesson will give you motivation. Opting for a structured MOOC or course plan can also be helpful, since they provide you with constraints you’re required to follow.
Another alternative is to set up short-term exams which you need to pass along the way. This could be useful in studying for a larger self-study exam (SAT, MCAT, GMAT, LSAT, CFA, etc.). Basically, you could break down practice exams into segments and resolve to be able to ace a particular segment by the end of the week, giving you the motivation to learn that section without procrastinating.
Anti-Strategies (or Plans that Rarely Succeed)
In contrast with the above three mentioned strategies, I’ve also found some approaches that tend to work poorly. This doesn’t mean they never succeed, but rather that they require disproportionately more motivation or self-discipline to execute. These include:
- Working on your project whenever you feel like.
- Not establishing particular scheduled hours or deadlines.
- Planning to begin a learning task later, without providing a compelling reason why it should be delayed.
In the end, you know yourself and your motivation. If getting stuff done isn’t a problem for you—don’t worry about this step. If it is, I’d recommend using strategy #1 in most cases. It’s a good default go-to approach when you’re not sure which one to apply.
Step Five: Long-Term Retention
This final step is an optional one. For many learning projects, I pursue this step informally because I know my lifestyle and goals will allow me to circle back to the knowledge I acquired previously at some point.
For those who are worried that such an informal approach may lead to losing a lot of the knowledge acquired, taking additional steps can be useful. Adding a strategy for long-term practice and retention can make sure that you don’t forget things years later.
Learning for Long-Term Retention
My first weapon against the long-term decay of memory is to learn it better, the first time around. I’ve found that learning with the goal of understanding promotes the best long-term retention compared to memorized facts.
Consider learning physics. Most students spend a great deal of time memorizing formulas and the situations where they apply. Smart students spend time trying to build the intuitive principles for what the formulas are saying and why they work.
Sometimes learning to understand isn’t a short-term goal. Learning how to solve a particular problem with an equation takes a lot less time than trying to build an intuition around how it works, but years later the equations will be forgotten and the intuition will remain.
This is why I recommend metaphors, visualization, diagrams and the Feynman technique when learning. They promote the process of decoding an abstract idea into an intuition that you can keep with you much longer than memorized trivia.
This doesn’t mean understandings are immune to forgetful minds, but simply that they persist longer.
Here are some other mechanisms you can use to ensure long-term retention:
1. The Orbit Strategy
Think of how the moon orbits the Earth, returning to the same relative position each month. This strategy works similarly—after completing a project, set a notice on your calendar a few months or years into the future. Once the time comes, do a mini project to reactivate those skills.
I intend to do this broadly with the programming and computer science knowledge I acquired during the MIT Challenge. By doing a mini project every 6-12 months, I hope to sustain my skills even when I’m at a stage in my life where they aren’t a main part of my career.
I recently executed this successfully with French. Even though it had been over two years since I lived in France (and spoke French infrequently) I made the goal of going back to Paris for a month and speaking exclusively en français. I was surprised that I was even able to improve my French from where I had left it after that burst.
If the goal is only sustaining, not improving, then the period of the orbits doesn’t need to be fixed. Increasing the spacing between each burst can probably sustain the same level up to a point. An exception would be very high levels of skill (which decay more quickly) and where the skill itself changes rapidly (such as programming).
2. Scheduled Practice
Another strategy is to schedule practice or recall regularly, in small doses. I know that Benny Lewis , who speaks around ten languages fluently, uses this approach to maintain his ability. By speaking the languages every week or so, he can continue to sustain and improve his abilities over the long-term.
3. Formal Systems (SRS)
If these strategies are too informal for you, then you can opt for implementing an even more structured review using a spaced repetition system such as Anki . This would be particularly useful if you needed to retain a large corpus of factual information you aren’t using frequently. I suspect medical and law students, for example, would benefit from having the factual details of their courses inputted into Anki, which they would then get reminders of long after the class was taken, so the knowledge doesn’t fade.
Implementing the 5 Steps
This is just a framework for planning and executing a self-education project. As such, you may have a lot more questions about handling the specifics. Here are a few articles I’ve written on the details of learning efficiently:
- How to understand hard ideas .
- How to take notes . (PDF)
- How to read faster .
- How to read a textbook .
- How to use metaphors/visualizations to study for an exam .
- How to quickly learn a free online course .
- How to get work done .
- How to focus on what you’re learning .
- How to access ‘hidden’ courses  from MIT.
- How to learn really hard subjects .
- How to translate your self-education into an actual degree .
I cover all of this comprehensively in my course , but those above free resources should be a good starting point if you’re not ready to invest in it.
The Benefits of Learning Well
Self-education can seem like a luxury at times. Or it can look like an exercise in intellectual wastefulness—something that doesn’t materially improve your life. I’ve found the opposite is true: learning more gives you an enormous advantage in almost any area of life you choose to apply it toward.
The people I know with the best careers, relationships and lives are the ones who learn continuously. I always strive to have a learning project at all times, and following these steps have been essential to make them successful, instead of something I idly start and never finish.
Image courtesy of Hash Milhan .