How to Take Better Notes While Listening to Podcasts

If you’re reading this, you’ve probably tried to take notes on a podcast at some point and discovered the problem: it’s much harder than taking notes on a book.

A book sits still. You can underline a sentence with a pen. A podcast moves at the speed of speech, and by the time you’ve reached for your phone, the sentence is gone.

This post is a working system, what I call the marginalia method, for capturing the moments that matter without breaking your stride. I’ll walk through the three steps, then show you the tools that actually make it work.

I take a lot of podcast notes. I built Margin partly because the existing tools weren’t quite right. But the method below works with any tool, or no tool at all. The principles are what matter.

Why podcast notes fail (and why you forget)

Before the method, the diagnosis.

Most people fail at podcast note-taking for one of four reasons:

1. They wait until later. They tell themselves they’ll write it down after the walk. By the time they sit down, they’ve forgotten the specific sentence, and the specific sentence was the whole point.

2. They try to type. They unlock their phone, swipe to Notes, and try to type with one thumb while crossing a street. The note ends up being three words long and incomprehensible by morning.

3. They take too many notes. They mark every other minute. By the end of the episode they have thirty notes, none of which is the real note, because nothing is highlighted relative to anything else.

4. They have no system for going back. Even when they take a good note, it lives in a Notes app folder with no connection to the episode, the timestamp, or the moment. Six months later they can’t find anything.

The marginalia method fixes all four by mimicking the way readers have always marked books: a small, instant gesture, made in the margin, retrievable later.

The method, in three steps

Step 1: Capture in the moment, not later

The single most important principle is capture while the moment is happening. Not after. Not at the next stoplight. Not when you sit down at your desk. In the moment.

The reason this matters is that what you want to keep is not a paraphrase of what was said, it’s your reaction to what was said. The reaction is alive for about ten seconds. After that, you have a memory of the reaction, which is not the same thing. The fading is fast and irreversible.

If you can’t capture in the moment, capture within 30 seconds of the moment. After that, accept that you’ve lost it.

Practical translation: - Whatever tool you use, it has to be available without unlocking your phone. - The capture gesture has to be one motion. - If the capture takes more than 5 seconds, you’ll skip 80% of the time.

Step 2: Capture small, not complete

The instinct most people have when they start taking podcast notes is to write a complete thought: “The host argued that distribution is more important than the product itself, citing Microsoft’s history of winning markets through bundling.” This is too much. You’re not going to do it. You’ll do it for the first three notes and quit.

The right grain is the trigger phrase, the smallest possible reference back to the moment. Examples:

• “moat = distribution, not weights”
• “Howard Marks, second-level thinking”
• “the Edwin Land obsession thing, Founders, ~22 min”

If you’re capturing by voice (recommended), say one sentence. The full thought can come later when you review. The note’s job is to anchor your memory, not to replace it.

This is the same principle as marginalia in books. When you underline a sentence, you’re not summarizing, you’re bookmarking your attention. You’re saying here. This. The rest is reconstruction.

Step 3: Always include the moment

The single thing that turns notes from disposable text into a real library is anchoring them to the moment that triggered them.

This means two pieces of metadata:

1. The episode, what show, what episode, what date
2. The timestamp, where in the episode you were when you marked the note

With both, you can come back six months later, tap the note, and land back at that exact second of audio. Without them, you have a paragraph of text floating in space, no way to verify, no way to re-listen.

This is the biggest difference between a “note” and what readers do with books. When you underline page 73, the underline is on page 73. The position is the metadata. With audio, the position has to be manually attached, and almost no tool does this for you automatically.

A few that do (and which I’ll cover below): Margin (Spotify), Snipd (any podcast played inside Snipd), Pocket Casts bookmarks.

The system, in practice

Here is what the loop actually looks like, for me, on a typical weekday morning.

1. I leave the house. AirPods in. Podcast starts on Spotify.
2. I’m walking. About 14 minutes in, the host says something that makes me stop.
3. I press and hold the Margin button on my lock screen. Spotify pauses.
4. I say into the phone: “the moat = distribution thing. compare to how Microsoft did this in the 90s.” About six seconds.
5. I release. Spotify resumes. I keep walking.
6. The note is now saved with the episode name, the date, and a timestamp of 14:23. It’s in my library forever, transcribed on-device by Apple’s Speech framework, and I never opened an app.

This is the lowest-friction version of the method. If you don’t have Margin, the closest equivalents are:

• Snipd: tap “snip” while listening inside Snipd’s player. Adds the last 30 seconds of audio as a clip with transcript.
• Pocket Casts: tap “bookmark” while listening. Saves the timestamp; lets you add a title.
• Voice Memos + manual timestamp: record a voice memo, jot the show + minute in the memo title. Crude but works.

Each of these is more friction than Margin (which is why I built it). But each works.

What to do with the notes

The other question I get a lot is: what do you do with all the notes once you have them?

This is the part where most podcast note-taking falls apart, because nobody told you the back half of the workflow.

Here’s what I do:

Daily, nothing. I don’t process notes the day I take them. They live in Margin’s library, organized by episode, retrievable.

Weekly, scan. Once a week, on Sunday, I open the library and scroll through the week’s notes. Most of them I don’t touch. A few of them surface as worth deeper thought. I’ll re-listen to the moment for those.

Monthly, promote. Once a month or so, the small handful of notes that have continued to matter get promoted into a separate place, a Notion database for me, but it could be Obsidian, Apple Notes, anywhere. These are the ones I keep coming back to.

Quarterly, write. Every few months I write something, an essay, a Twitter thread, a memo to myself, that draws on the promoted notes. This is the step that actually makes the listening pay off in your thinking. If you never write from the notes, the listening was entertainment, not learning.

Most people don’t do steps 3 and 4. They take notes and the notes pile up. The piling up is fine, it’s better than not taking notes, but the real return is in the writing-from-notes step.

Five common mistakes (and the fixes)

Mistake 1: Trying to type instead of speak. Fix: voice is faster, lower-friction, and forces brevity. Type only when you’re at a desk.

Mistake 2: Notes with no episode context. Fix: any tool that auto-attaches the episode and timestamp. If you’re using a generic note app, at least write the show name + timestamp at the top of every note.

Mistake 3: Marking too much. Fix: aim for 3-5 notes per episode max. The discipline of choosing what to mark is part of the value.

Mistake 4: Never reviewing. Fix: a weekly 10-minute scan on Sunday. The notes you don’t revisit are notes you might as well not have taken.

Mistake 5: Treating notes as a finished product. Fix: notes are raw material. The output is whatever you make from them, an essay, a decision, a conversation. Don’t optimize the note-taking; optimize the using-them.

The deeper point about marginalia

The reason this works, the reason underlining a book and capturing a podcast note are the same thing, is that the act of marking is itself generative.

When you decide this sentence, not the other sentences, you’re making a small judgment about what’s worth your attention. That judgment is the beginning of thinking. It’s how marking books shaped writers like Coleridge and Susan Sontag. Their margins were where their books happened.

The same is true for podcasts. The point of marking the moment isn’t to retrieve the moment later (though that’s nice). The point is that the act of marking itself sharpens what you notice. You start listening for the things worth marking, which means you start listening more actively, which means more of the listening sticks.

This is the reason I made the press-and-hold gesture the entire product of Margin. The gesture is the whole point. Everything else, the library, the timestamps, the transcripts, exists in service of the gesture.

Get started

If you want the easy version: download Margin when it launches (you can get on the list at margin.fm).

If you want to start today, with whatever’s on your phone: open Voice Memos. The next time you hear something on a podcast that makes you stop, hit pause on the podcast, hit record on Voice Memos, say one sentence about why you cared, then resume the podcast.

That’s the whole method. The tools are downstream. The habit is what matters.

Selinay