The Feynman Technique: What Feynman Actually Said Beats the Four Steps Sold in His Name

You have been handed a business you do not understand and a board date in three weeks. The Feynman technique is a way to learn it fast: pick the subject, explain it in plain words to someone new to it, find where the explanation breaks, go back to the source, and simplify until it holds.

That is the version every guide teaches. Our newsroom read 16 of them, including seven of the ten pages Google ranked highest for the term, and counted what they tell you and what they cite. Then we checked the primary record for what Feynman actually said.

What We Found

Of 16 published Feynman Technique guides we read on 8 November 2026, 14 give four steps and 10 mention a child, yet only one names a book by or about Feynman as its source.

The single source is James Gleick’s biography, named by the Todoist guide and linked to a shop listing rather than a page. Four guides say outright that Feynman developed the method. Eight guides come from universities, and six of those eight bring a child into it, from a third grader at Kansas State to “an 8-10 year old” at Western University in Canada.

The primary record contradicts the child. The story the technique rests on is about a freshman, a first-year university physics student. A child cannot tell you that your explanation of a margin bridge is wrong. A sharp new hire can.

What Is the Feynman Technique?

The Feynman technique is a study method named after the physicist Richard Feynman. You choose a subject, explain it in simple language as if teaching someone new to it, note where you stall or reach for jargon, return to the source to fill those gaps, and repeat until the explanation is short and clear.

The guides agree on the shape and disagree on the details. Farnam Street’s guide, which Bucknell credits as a source, asks you to teach it to a 12-year-old and then to “test and archive”. Bucknell’s Teaching and Learning Center marks the simplify step as optional. Western University stretches the method to six steps and North Greenville University to five. The other 14, Farnam Street and Bucknell included, settle on four.

The idea under every version is that explaining forces you to recall, and recall shows the gaps that rereading hides. Our guide on how to learn faster covers the research on retrieval, and the piece on how to remember things quickly covers what keeps it in your head afterwards.

Did Richard Feynman Create the Feynman Technique?

We found no evidence that he did. None of the primary sources we read, his own books, his Caltech speeches or his colleagues’ accounts, describes a numbered method under his name. The earliest dated use of the name we found is a blog post by Scott Young on 1 September 2011, 23 years after Feynman died.

Young’s post from that day announces a video called “The Feynman Technique” as the first video of a free seven-day learning bootcamp. LearningWiki says the technique in its modern form “seems to have been created/popularized by Scott Young around August 2011”, though it gives no citation. Farnam Street’s guide carries a publication date of April 2012 in its page data.

Some guides hedge honestly. Tes says his method “has since been formalised into what is known as the Feynman technique”, and a discussion thread in Google’s top results that day questions whether Feynman created it at all. The fair reading is that other people assembled the steps from things Feynman did and said, and put his name on the result.

Numbered methods travel well, as the Dave Ramsey baby steps show, though Ramsey named his own.

What Feynman Actually Said

Six passages in the primary record carry the ideas the guides use, and together they make a sharper method than the four steps.

The freshman test comes from David Goodstein, a Caltech colleague, who asked Feynman to explain why spin one-half particles obey Fermi-Dirac statistics. Feynman promised a freshman lecture on it. The account, printed in Physics Today in February 1989 and reprinted in Caltech magazine’s Feynman at 100, ends: “I couldn’t do it. I couldn’t reduce it to the freshman level. That means we really don’t understand it.” Goodstein calls that reformulation “the standard by which he measured whether something was really understood”.

Two things in that story rarely reach the guides; of the 16, only Science ABC retells it. The audience was a freshman, an intelligent adult without the specialist vocabulary. And the test failed. Feynman stopped and concluded that the field did not understand its own result.

The notebook comes from Gleick’s 1992 biography, Genius. Facing his oral examinations as a Princeton graduate student, Feynman drove to MIT, “opened a fresh notebook” and wrote on the title page “NOTEBOOK OF THINGS I DON’T KNOW ABOUT”, in the passage quoted by Cal Newport. He “worked for weeks at disassembling each branch of physics”. Newport, whose method we covered in Cal Newport time blocking, reads the notebook as a way to keep going on hard material.

The gap-finding step has its best source in Surely You’re Joking, Mr. Feynman! (1985), where he describes teaching physics students in Brazil. In the excerpt reproduced here, he finds that “the students had memorized everything” but “they didn’t know what anything meant”. They could recite definitions and could not apply them.

The jargon test is his father’s. In What Do You Care What Other People Think? (1988), his father names a bird in several languages and tells him that knowing the names tells you nothing about the bird. Feynman’s line, quoted by Today in Science History, is “I learned very early the difference between knowing the name of something and knowing something.”

The last two are about checking yourself. His 1974 Caltech commencement address, Cargo Cult Science, says “The first principle is that you must not fool yourself”. And according to Alan Alda’s Caltech commencement speech, the top of his last blackboard read “What I cannot create, I do not understand.”

Primary source	What it supports	What it does not support	Link
Goodstein on the freshman lecture, Physics Today, 1989	Explaining simply as a test of understanding	A child as the audience; the test always succeeding	Caltech magazine
Gleick, Genius (1992), the notebook	Writing down what you do not know, then rebuilding each subject	A four-step sequence	Cal Newport, quoting Gleick
Surely You’re Joking, Mr. Feynman! (1985), Brazil	Recitation without understanding as the failure to find	Any named method	Excerpt
What Do You Care What Other People Think? (1988), the bird	Cutting jargon; names are not knowledge	A step order	Today in Science History
Cargo Cult Science, Caltech, 1974	Testing your own understanding hard	A study technique	Caltech
His last blackboard, 1988, per Alan Alda	Rebuilding a thing yourself as proof you understand it	Explaining as the final step	Caltech
Scott Young, 1 September 2011	The name “The Feynman Technique” in use	Feynman using the name	scotthyoung.com

Feynman Technique Guides Compared: 16 Pages, Counted

The table holds the dataset behind the headline. “Child” means the page brings a child into the explanation step, as the listener or as a comparison. “Names a source” means it names a book by or about Feynman as its source; a photo credit or an unnamed video does not count. “Says he developed it” means the page states that Feynman created or developed the method.

Guide	Type	Steps	Child	Names a source	Says he developed it	Link
Farnam Street	Blog	4	Yes, a 12-year-old	No	No	fs.blog
Todoist	Software company	4	No	Yes, Gleick	Yes	todoist.com
University of York	University	4	Yes	No	Yes	york.ac.uk
Adobe Acrobat	Software company	4	No	No	No	adobe.com
Kansas State University	University	4	Yes, a third grader	No	No	k-state.edu
Bucknell University	University	4	No	No	No	bucknell.edu
University of Colorado Boulder	University	4	Yes, a six-year-old	No	Yes	colorado.edu
Tes Magazine	Education publisher	4	No	No	No	tes.com
Innovation Training	Training company	4	No	No	Yes	innovationtraining.org
University of Limerick	University	4	Yes	No	No	ul.ie
Western University	University	6	Yes, 8 to 10	No	No	uwo.ca
Cal State San Bernardino	University	4	No	No	No	csusb.edu
Ali Abdaal	Creator blog	4	Yes	No	No	aliabdaal.com
Science ABC	Explainer site	4	Yes, as a comparison	No	No	scienceabc.com
Goodnotes	Software company	4	Yes	No	No	goodnotes.com
North Greenville University	University	5	Yes, a sixth grader	No	No	ngu.edu
Oxford Learning (dropped)	Tutoring company	Page failed to load				oxfordlearning.com
Noji (dropped)	App blog	Page returned not found				noji.io
Saint Mary’s University of Minnesota (dropped)	University	PDF address could not be located				smumn.edu

The University of Colorado page carries the six-year-old line and attributes it to Albert Einstein, and Bucknell opens with an Einstein quote too. A method named after one physicist is introduced with a line credited to another.

How Do You Use the Feynman Technique at Work?

Use the Feynman technique at work by changing the audience and the finish line. Explain the subject to a capable colleague from another function, the freshman in Goodstein’s story, and count it learned only when you can rebuild the number or the argument yourself, from the source, without notes.

First, open the notebook. Write the subject at the top and, underneath, every question you cannot answer yet. Feynman’s title page was a list of what he did not know, which is a more honest starting point than a list of what you do.

Second, prepare the freshman version. Pick a real listener: a smart analyst from finance if the subject is operational, a smart engineer if it is commercial. Explain it to them out loud in ten minutes.

Third, run the bird test on your own words. Every term you used that the listener had to take on trust is a name, not knowledge. Go back to the source for each: the ledger, the contract, the person on the floor who does the work.

Fourth, rebuild it. Feynman’s blackboard line sets a higher bar than explaining. If you cannot reconstruct the number from its inputs, you do not understand it yet. Then ask someone to try to break your version, because you are the easiest person to fool.

Fitting it into a week is the hard part. A first pass needs about an hour of quiet; the morning slot people build with the Robin Sharma 5 AM Club works if you already keep it, and the two-minute rule from our Atomic Habits summary works if you do not: write one question in the notebook each day.

Learning a New Market in a Fortnight

Suppose you are sent to judge an entry into a market you have never sold into. The notebook page fills fast: who buys, who pays, how long the sale takes, what a customer is worth over five years, who the incumbent is and why customers stay with them.

The freshman version is ten minutes with your head of finance on how money moves in that market. Where you stall, you have borrowed someone else’s summary. “The channel is consolidating” is a bird name. Who bought whom, at what price, and what prices did afterwards is knowledge.

The rebuild is a market size built from customer counts and prices, set beside the consultant’s figure. If the two differ by a factor of three, the gap is your next notebook page.

A New P&L Line

A line on the management accounts that you did not create is an easy place to be fooled. You can read its trend for a year without being able to say what moves it.

Explain the line to a colleague outside finance: what creates an entry, what reverses one, which judgement calls sit behind the figure, and who makes them. If you cannot answer the last two, ask the controller to walk you through one month’s entries. That is the Brazil test applied to accounts, where the label tells you nothing about the business.

Then rebuild one month from the ledger. After that, a variance on the line becomes a question you can ask well.

Explaining a Decision to a Board

A board is a room full of freshmen in Goodstein’s sense: intelligent, short of time, and outside your vocabulary. The freshman version of a decision is the decision, the reason, the cost, what would change your mind, and what you are not sure about, in that order.

Simon Sinek’s ordering, covered in our piece on the Simon Sinek golden circle, puts the reason first, which suits a board paper. The Feynman addition is the last item. Goodstein’s story ends with an admission, and a paper that names what management does not yet understand is easier to trust.

The same discipline works on a smaller page. Our breakdown of the Tim Ferriss cold email template is a freshman test for a stranger’s inbox: one read, one clear ask. Our decision making matrix helps when the reason has several parts and the board needs to see how they were weighed.

Where the Feynman Technique Fails

The child framing is the first problem. Simplifying for a child rewards dropping the parts that are hard to say, and in business those are usually the risks. A freshman listener can ask why; a child audience lets you skip it.

The second problem is that fluent explaining feels like understanding. Feynman’s students in Brazil could recite. A clean explanation of a market you have never sold into is still a guess, which is why the rebuild matters more than the talk.

The third is cost. By Gleick’s account, Feynman spent weeks on his notebook before one examination. A manager who runs the full method on every subject will run it on none, so keep it for the two or three things a decision turns on.

Our Position: Keep the Steps, Drop the Child, Add the Rebuild

The four steps are a reasonable study routine, and nothing in the primary record contradicts them. The guides go wrong in two places: they swap Goodstein’s freshman for a child, and they stop at explaining when Feynman’s own standard was creating.

The attribution question matters less than the content, but it is a useful warning. A method with a famous name gets repeated without anyone checking the source, and 15 of the 16 guides we read name no book by or about Feynman. Our pieces on the Let Them theory summary, the Mel Robbins books, the Tony Robbins books and the Mark Wahlberg morning routine cover other named methods from the same shelf.

Use the Feynman technique, then. Run it on the market, the line or the decision that matters this quarter, explain it to the smartest person outside your function, and do not call it learned until you can rebuild it.

How We Did This

On 8 November 2026 we read 19 published guides to the Feynman technique: the US top 10 Google results for the term that day (oxfordlearning.com, noji.io, tes.com, a smumn.edu PDF, fs.blog, subjectguides.york.ac.uk, todoist.com, colorado.edu, adobe.com and innovationtraining.org, plus a discussion thread questioning whether Feynman created it), university study guides and other guides from the first pages of a web search. Three were dropped: Oxford Learning failed to load, Noji returned not found, and we could not locate the smumn.edu PDF. A USC copy of the Todoist guide was excluded as a duplicate, and Kansas State’s web and PDF versions counted once. For each of the 16 left we recorded the number of steps, whether a child appears in the explanation step, whether the page names a book by or about Feynman as its source, and whether it says he developed the method, with a supporting quote for every value. Forum threads and videos were excluded. We then read seven primary and near-primary sources on what Feynman said and did. The full dataset, with the quote and source for every row, is available on request.

Frequently Asked Questions

What are the 4 steps of the Feynman technique?

Choose a subject, explain it in plain language as if teaching someone new to it, find the gaps where you stall or use jargon and return to the source to fill them, then simplify and repeat. Fourteen of the 16 guides we read use four steps; Western University uses six and North Greenville University five.

Did Einstein or Feynman say “if you can’t explain it simply, you don’t understand it”?

We found no primary source for either. Bucknell’s guide opens with “If you can’t explain it simply, you don’t understand it well enough” and credits Einstein, and the University of Colorado guide credits a six-year-old version to him; neither cites a source. The closest verified line is Feynman’s, from David Goodstein’s account: “I couldn’t reduce it to the freshman level. That means we really don’t understand it.”

Is the Feynman technique good for adults learning at work?

Yes, with two changes. Explain to a capable colleague from another function rather than an imagined child, so the listener can challenge you, and treat the work as finished only when you can rebuild the number or argument yourself from the source.

Feynman’s own test ended with him saying the field did not understand its result, and the version worth copying keeps that ending.