How to repurpose a podcast into a newsletter
Podcast newsletters open better than almost any other category, and most of them are still transcript dumps. Here is how to find the structure an episode already had and write from it.
7 minute read
A transcript records someone thinking, it is not a draft
The standard failure is treating the transcript as a rough first draft that needs tidying. It is not a draft at all. Spontaneous speech carries restarts, repairs, and fillers that are structural rather than careless, and they track effort. Bortfeld and colleagues, writing in Language and Speech, found disfluency rates rose when a speaker took the harder role in a task and when the topic was abstract rather than concrete, which they read as evidence that disfluency marks planning difficulty. That has an awkward implication for your edit. The passages where you or your guest were doing the most original thinking are the ones that look worst on the page. Judging your material by how cleanly the transcript reads will systematically discard your best segments.
You have about fifty seconds and two hundred words
Then there is what a reader will actually give you. Jakob Nielsen's eyetracking research at Nielsen Norman Group, covering 42 participants across 117 newsletters, found the average time allocated to a newsletter after opening it was 51 seconds, that participants fully read only 19 percent of newsletters, and that 35 percent of the time they skimmed a fragment or merely glanced at it. Set that beside Marc Brysbaert's meta-analysis of 190 studies and 18,573 participants, which puts adult silent reading of non-fiction at 238 words a minute, and the arithmetic is unforgiving: 51 seconds buys roughly 200 words of real reading. A nine-thousand-word transcript is not forty-five times the value. It is the same email with the useful part buried in it.
Find the structure the episode already had
Every decent episode has a shape, and it is almost never chronological. Somewhere in the hour there is a claim the rest of the conversation kept circling, a moment where one of you pushed back, and a point where something got settled or conspicuously did not. That is your structure. Write it as a single sentence first: what this episode established. Then pick the two or three exchanges that got you there, arranged in whatever order makes the argument rather than the order they were recorded. Working forward from minute zero produces show notes, which are a different and much less valuable product. If you cannot write that one sentence, you either have a thin episode or you have not listened back properly, and both are worth discovering before you draft.
What to quote and what to paraphrase
The rule that keeps these readable comes from journalism. Roy Peter Clark's guidance at Poynter is to begin from the assumption that you can write the idea better than the source said it out loud, and to reach for the direct quote only when that assumption fails. Applied to an episode, most of what your guest said should be paraphrased in your voice, tightly, with clear attribution. Quote when the wording itself is the value: an unusually sharp line, a figure stated precisely, something contentious enough that readers need to see it verbatim, or a moment where the exact words are the point. Two or three real quotes in an issue is plenty. A newsletter that is mostly quotation marks is a transcript with better typography, which is the thing you were avoiding.
Clean in three passes, in this order
Editing a transcript feels endless because people try to do three jobs in one read. Split them. First pass, correctness: proper nouns, company names, technical terms, numbers and units. Automatic transcription is reliably weakest on exactly these, and a mangled guest name is the single error every reader notices and nobody forgives. Second pass, structure: delete everything outside the two or three exchanges you chose, without softening or rewriting yet. Third pass, prose: strip fillers and false starts, join fragments into sentences, and rewrite anything that only worked because of tone of voice. Done in this order each pass finishes. Done together, you re-read the same paragraph with three purposes in mind and complete none of them.
Some of the episode has no written equivalent
Part of what made the conversation good will not convert, and you should stop trying to make it. Timing, the pause before a difficult answer, two people laughing at the same instant, the way a voice changes when a question lands badly. Written down these become stage directions, and stage directions in a newsletter read as padding. The honest move is to point at them rather than transcribe them: name what happens, say roughly where in the episode it sits, and link straight to the moment. This is also the strongest practical argument for shipping a clip alongside the issue, because fifteen seconds of the actual exchange does in one play what three paragraphs of description will not do at all.
Decide what the issue is for, then batch the whole thing
Before you write another one, settle what the newsletter is actually for. If its job is to make people listen, keep it short, argue one idea, and put the play link high. If its job is to stand alone for readers who will never press play, it has to work with no audio at all and carry the full argument. Both are legitimate. Mixing them produces the issue that half-explains something and then asks you to go and listen, which serves neither reader. Whichever you pick, one recording should be feeding the written issue, the clips, and the social posts in a single session rather than three. That is the pass FrameOS is built for: the full episode in, captioned vertical clips out in one sitting, so the writing and the clips ship the same day. 300 credits for 3 days · no card.
FAQ
Can I just send my podcast transcript as a newsletter?
It will not work. Nielsen Norman Group's eyetracking research found readers give a newsletter about 51 seconds after opening it and fully read only 19 percent of them, which at typical adult reading speeds is roughly 200 words. A transcript also preserves the restarts and fillers that make spoken thinking readable in audio and unreadable on a page. Use it as source material, not as a draft.
How long should a podcast newsletter be?
Shorter than instinct suggests. The realistic reading budget is a couple of hundred words, so build the issue around one argument and two or three supporting exchanges. If you want to serve readers who go deeper, put the extra material below a clear break rather than in front of the point, so the people skimming still leave with the thing you wanted them to have.
Should I quote my guest directly or paraphrase them?
Paraphrase by default, quote by exception. Poynter's Roy Peter Clark advises starting from the assumption that you can write the idea better than it was said aloud and quoting only when you cannot. Reserve direct quotes for lines where the phrasing itself carries the value, precise figures, and anything contentious enough that readers should see the exact words. Two or three per issue is usually enough.
What is the difference between show notes and a podcast newsletter?
Show notes follow the episode in order and exist so listeners can navigate it. A newsletter follows an argument and exists so a non-listener still gets something. If you find yourself summarising minute by minute, you are writing show notes. Both are useful, but only one of them is worth an email, and confusing the two is why most podcast newsletters underperform their audience.
Sources
- The State of Newsletters 2026 · beehiiv
- Email Newsletters: Surviving Inbox Congestion · Nielsen Norman Group
- How many words do we read per minute? A review and meta-analysis of reading rate · Journal of Memory and Language
- Disfluency rates in conversation: effects of age, relationship, topic, role, and gender · Language and Speech
- 15 tips for handling quotes · Poynter
Related pages
Keep reading
More Podcasts →How to livestream and clip it afterwards
Twitch deletes most streamers' past broadcasts after seven days and Facebook deletes live video after thirty. What you record locally, and how, decides what you can still post next week.
8 minute readHow to build a clip library you can actually re-use
Four hundred exported clips in a folder is not a library. A library is one row per clip, carrying the hook line, a short tag set, and a record of everywhere it has already run.
8 minute readHow to turn a blog post into a video
A 1,500-word article read aloud is roughly eight minutes of video. Here is how to work out which single idea in it earns the runtime, and why most lists do not survive the jump.
8 minute readTurn one long video into a week of posts
300 credits for 3 days · no card.
Start free