The Feynman Technique Explained: Learn Anything by Teaching It

Richard Feynman did not publish a method called the Feynman Technique. The name was attached later to something he did habitually and talked about often: when he did not understand a thing, he would try to explain it in the simplest terms possible, and pay close attention to the exact place his explanation collapsed.

That is the whole idea, and its power is that it is almost impossible to fake. You can reread a chapter while understanding very little. You cannot narrate a clear causal account of it while understanding very little, because the gaps announce themselves out loud.

The four steps

1. Write the concept at the top of a blank page

One concept, narrow enough to explain in a few minutes. Not thermodynamics but why heat flows from hot to cold. Not the French Revolution but why the price of bread mattered politically in 1789. Vague scope is the most common reason the technique produces nothing useful.

2. Explain it in plain words, from memory, as if to a beginner

Close the source. This part is non-negotiable, because with the book open you are transcribing, not retrieving. Write or speak a full explanation for someone intelligent who knows nothing about the topic. Ban jargon unless you define it in the same breath. Use concrete examples and, where it fits, an analogy.

Speaking aloud is often better than writing, because pauses and filler words mark the weak joints precisely. Recording yourself for two minutes and listening back is uncomfortable and extremely informative.

3. Hunt the gaps, deliberately

This is the step people skip, and skipping it turns the whole method into a pleasant monologue. Go back through your explanation looking for specific failure signatures.

• Places you reached for a technical term because you could not say the thing plainly.
• Sentences that assert what happens but never explain why it happens.
• Steps joined by and then rather than by a cause.
• Points where you said something like it just works that way.
• Anything you could not give a single concrete example of.

Each of those is a coordinate. Go back to the source material and read only what those coordinates point at. Targeted rereading after a failed attempt is far more efficient than rereading everything upfront.

4. Simplify and tell it again

Rewrite the explanation, shorter and cleaner, with the gaps repaired. Then repeat the loop until you can get through it without stalling. The finished artefact is a short plain-language account you could give to a curious twelve-year-old.

Why it works

Three well-supported findings do the heavy lifting here, and it is worth being precise about them rather than claiming the technique is uniquely powerful.

First, retrieval. Pulling information out of memory strengthens it more than putting it in again, which is the finding behind the testing effect and the reason self-quizzing beats rereading. We cover the evidence and the practicalities in our guide to retrieval practice.

Second, generation and elaboration. Producing an answer or an explanation yourself, rather than reading one, tends to improve later recall, and explaining how and why something works helps you build the causal structure that isolated facts lack. The Feynman Technique is essentially a generation task with a demanding audience.

Third, calibration. Fluency is a liar. Text you have read three times feels known because it is easy to process, not because you could reproduce it, and that mismatch is the core of the illusion of knowing. An explanation attempt is a cheap, brutal calibration check: either the sentences come out or they do not.

What the research does not say is that teaching guarantees mastery, or that this method beats every alternative for every subject. Expecting to teach seems to help partly by changing how you study, and the effect sizes in that literature are modest. Treat the technique as a reliable, low-cost diagnostic rather than a miracle.

Common mistakes

One more subtle trap: simplifying into inaccuracy. Plain language is the goal, but an analogy that quietly misrepresents the mechanism is worse than a slightly technical sentence that is true. When you notice your simplification has changed the meaning, say so and keep the harder version.

A four-step routine you can run in minutes

1. 1Minute 0 to 1: write the concept and who your imaginary beginner is.
2. 2Minute 1 to 4: explain it aloud or on paper from memory, no source, no jargon.
3. 3Minute 4 to 6: mark every gap using the failure signatures above.
4. 4Minute 6 to 10: reread only the sections your gaps point at, then retell the explanation once, shorter.
5. 5Later: schedule one retell at a day, then a week, then a month.

That last line is the part that turns a good session into durable knowledge. A single successful explanation shows you understand something now, not that you will in April. Spacing the retells out is what keeps it, for the reasons laid out in spaced repetition explained.

Where the raw material comes from

The technique needs something to explain, and short self-contained ideas are far easier to practise on than whole chapters. A single mechanism, a single event, a single principle: something you can hold in one explanation.

Disclosure: MindSnap is our app. We mention it here for one honest reason, which is that its lessons are already cut to that size. A two-minute story-driven lesson gives you exactly one idea, which makes it a convenient prompt to close the app and explain what you just read to nobody in particular. The library covers five flagship story-driven collections plus unlimited topics you can request, so you can practise on whatever you are actually curious about. Any well-scoped source works just as well.

Why explaining feels harder than it should

Because it is the first honest measurement you have taken. Reading is smooth and gives you continuous positive feedback. Explaining removes the scaffolding and shows you the shape of what you actually hold, which is usually smaller and patchier than expected.

That discomfort is the signal, not a sign you chose a hard topic. Feynman's real contribution was not a productivity hack, it was the willingness to keep going back to the point where the explanation broke, instead of accepting the pleasant feeling of having read something.