Samurai Memory
Log in Get started free
Features How it works Pricing Get extension FAQ Log in
Samurai Memory / Guides

How to actually remember what you learn from podcasts (what the research says)

By Nicolò Borghi · Updated October 21, 2026 · 10 min read

Remember what you learn from podcasts: line drawing of headphones and a notebook.
The short answer

Listening isn't the problem: on average, people understand what they hear about as well as what they read. The problem is what happens after. Most of what you hear fades within days unless you pull it back out of your memory. Four habits have strong research behind them: write down what you remember right after an episode and check it, review it again a day and a week later, explain it to someone without notes, and turn one idea into a specific plan. Re-listening and highlighting feel productive but work much less well.

Why you forget most of a podcast

You finish a two-hour interview feeling sharper. A week later you remember the guest's name and one story. That's normal, and there are three reasons for it.

1. Memory fades fastest at the start

In 1885 Hermann Ebbinghaus measured his own forgetting and found that most of the loss happens in the first hours and days, then slows down. A 2015 replication found the same curve. Both studies used nonsense syllables, though, and a 2024 review of 256 papers found that meaningful material, like stories and course content, often fades more slowly. So the shape holds, but be wary of precise numbers.

You may have seen "we forget 70% of what we learn within 24 hours" or "we remember 10% of what we read and 90% of what we teach". Neither has a real study behind it. The second one, the "learning pyramid", has been traced back and has no data at all.

2. Your mind wanders more when you listen

In a 2013 study, 211 students took in passages from the same book by listening, reading aloud or reading silently, and were asked at random moments whether they were paying attention. Their minds were wandering at 51% of the checks while listening, against 36% while reading silently, and they remembered least from listening. In studies of recorded lectures, attention also drops as the video goes on: minds wander more in the second half, and memory follows.

3. You can't easily go back

A 2022 review of 46 studies (4,687 people) found no meaningful difference overall between listening and reading. Reading did a bit better in two cases: when readers set their own pace, and on questions that needed deeper understanding. A reader slows down and rereads a hard paragraph; a listener lets it go past. Listening isn't worse. You just have to make up for those missing moments of effort.

Re-listening feels like learning. Recalling is learning.

The most replicated finding in learning research is the testing effect: trying to recall something strengthens memory far more than going over it again. In a classic 2006 experiment, students either read a passage four times or read it once and then wrote down everything they remembered, three times, with no feedback. Here is what they remembered:

Tested 5 minutes later

Read it 4 times83%
Read once, recalled 3 times71%

Tested 1 week later

Read it 4 times40%
Read once, recalled 3 times61%
Share of the passage's ideas recalled. Roediger & Karpicke (2006), Experiment 2, 180 students.

Rereading wins if the test is in five minutes. After a week, the recall group remembers half again as much. And the people who reread four times felt the most confident they'd remember. That's the trap with re-listening to an episode or skimming your highlights: the content sounds familiar, and familiar feels like known. A large 2013 review of study techniques rated rereading and highlighting as "low utility", and self-testing and spaced review as the only two "high utility" ones.

Five habits that make podcasts stick

None of these studies used podcasts: they used texts and recorded lectures. But the principles are among the most robust in psychology, and they translate directly.

1. Do a 5-minute brain dump right after the episode

Close the app. On a blank page or a note on your phone, write everything you remember: the main argument, the stories, the numbers, what you disagreed with. Don't look anything up while you write. In a 2011 study published in Science, students who did this kind of free recall remembered about 50% more a week later (67% vs 45%) than students who spent the same time making a concept map with the text in front of them. Most of them expected the opposite.

Then check what you wrote against the source. This step matters more than it looks: a 2014 analysis of 61 studies found the testing effect is much stronger with feedback, and close to zero when people recalled little and never checked. Skim the transcript or a summary, fix what you got wrong, and add the one or two points you missed.

2. Pause every 15–20 minutes and recall three points

Long episodes are where minds wander most. In a 2013 study, people watched a 21-minute recorded lecture in four parts. One group answered a few quick questions after each part. Their minds wandered at 19% of the checks, against about 40% for the others, they took more notes, and they scored 90% on the final test, against 68% for those who just watched. On a long podcast, pause at a natural break and say or type the three main points so far. It takes thirty seconds.

3. Come back to it the next day, and again a week or two later

Spreading out reviews beats cramming them into one session: that's the spacing effect, confirmed by a review of 317 experiments. The ideal gap grows with how long you want to remember. In one study of more than 1,350 people, to remember something a week later the best gap was a day or two; to remember it a year later, it was a few weeks. For an idea you want to keep, that means: brain dump tonight, a two-minute recall tomorrow, another one in one to two weeks. Each time, recall first, then check. Sleeping between sessions helps too: sleep after learning is when the brain consolidates new memories.

4. Explain it to someone, without your notes

Telling a colleague or a friend "I heard something interesting on a podcast" is a recall exercise in disguise. A 2018 study compared people who taught material from memory, people who taught it reading from a script, and people who simply wrote down what they remembered. A week later, teaching without notes and writing from memory worked equally well, and teaching from a script didn't help. The benefit comes from the effort of recalling, not from talking. Notes on your phone defeat the purpose.

5. Turn one idea into an if-then plan

Business podcasts are full of advice you agree with and never apply. The fix with the best evidence is an implementation intention: a plan in the form "when X happens, I'll do Y". "When I open Monday's pipeline review, I'll ask each rep for the one deal they'd drop" beats "I should qualify deals harder". A 2006 analysis of 94 studies found these if-then plans had a medium-to-large effect on actually reaching goals. Pick one idea per episode, not ten.

The routine at a glance

  1. While listeningEvery 15–20 minutes, pause and recall 3 points.
  2. Right after5-minute brain dump from memory. Then check it against the transcript or summary.
  3. Next dayRecall it again in 2 minutes, or explain it to someone without notes.
  4. 1–2 weeks laterRecall once more, check, and write one if-then plan.

Speed, multitasking and notes: what the evidence says

Is it OK to listen at 1.5x or 2x?

Up to about 1.5x, the cost is very small for most people. A 2025 review of 24 studies found that the loss grows with speed: very small up to 1.5x, moderate to large at 2x and above. One well-known study of 231 students found no loss at 2x, but with short 13–15 minute videos; at 2.5x scores fell clearly. Older listeners lose more at high speeds. A practical rule: speed up the parts you know, slow down for the ideas you want to keep.

Can you learn while driving or working out?

Splitting your attention while you take something in costs memory, a finding repeated in lab studies since the 1990s. No study has measured how much people remember from podcasts while driving or at the gym, so here's the honest version: an easy walk or a familiar, quiet drive probably costs little; heavy traffic, a hard workout or answering email costs a lot, and the audio is what loses. Listen to the episodes you want to learn from when your head is free, and keep the rest for the commute. (For safety, studies found that listening to the radio or an audiobook didn't impair driving the way a phone conversation did.)

Should you take notes while listening?

How you take notes matters less than people think. The famous finding that handwritten notes beat laptop notes did not hold up: a 2021 replication with twice as many participants found no difference. What does matter is reviewing notes later, as a self-test (cover them, recall, then check) rather than a reread. If writing notes makes you stop listening, skip them during the episode and do the brain dump at the end.

Where Samurai Memory fits

All of the above works with a notebook and YouTube's own transcript (here's how to open it and summarize it with ChatGPT or Claude). The step people skip is checking: after a brain dump, you need the source to compare against, and scrubbing a two-hour video to find one sentence is slow. I built Samurai Memory partly for this. One click under a YouTube video saves its transcript to your library, so that:

5 videos a month are free, with every feature. No credit card, sign in with Google.

Try it on your next episode →

Samurai works with podcasts published on YouTube that have captions. One warning: reading the summary instead of recalling is just rereading. Recall first.

FAQ

Do you actually learn from podcasts?

Yes, but passive listening leaks. Research finds listening and reading produce similar understanding on average, but minds wander more while listening, and most of what you hear fades within days unless you recall it. Writing down what you remember after an episode, then checking it, makes a large difference.

Is listening to a podcast as good as reading?

On average, almost. A 2022 review of 46 studies found no meaningful overall difference between listening and reading. Reading did slightly better when readers set their own pace and on questions needing deeper understanding, because readers can slow down and go back.

How do I remember things from podcasts?

Right after the episode, spend five minutes writing everything you remember without looking, then check it against the transcript or a summary. Recall it again the next day and one to two weeks later, explain it to someone without notes, and turn one idea into a specific if-then plan.

What playback speed is best for learning?

Up to about 1.5x costs little for most people. At 2x and faster, most studies find test scores drop, more so for older listeners. Speed up familiar parts and slow down for ideas you want to keep.

Can I learn from podcasts while driving or exercising?

Some, but less. Dividing attention while taking information in reduces memory. Easy activities like a relaxed walk probably cost little; demanding ones like heavy traffic or hard workouts cost a lot. Save the episodes you most want to learn from for when your mind is free.

Should I take notes while listening to a podcast?

Optional. Whether notes are handwritten or typed makes little difference; reviewing them later as a self-test matters more. If note-taking makes you lose the thread, skip it during the episode and write a brain dump from memory at the end.

Is it true we forget 70% of what we learn in 24 hours?

There's no study behind that number. Ebbinghaus showed in 1885 that forgetting is fastest in the first hours and days, and this was replicated in 2015, but with nonsense syllables. Meaningful material often fades more slowly. The pattern is real; the precise percentage is not.

Sources

"I used to finish a two-hour episode feeling smarter and, a week later, remember one sentence. What changed it wasn't listening to more, it was spending five minutes after each episode writing down what I remembered, then checking it against the transcript."

Nicolò Borghi · Made by AlwaysBeta