Every new episode gets
shownotes that pull.
A small bot listens to every episode I publish, knows my ideal listener, and writes the full shownotes in their language: hook, chapter list with timestamps, tuned search keywords, persuasive CTA. The slowest part of podcast production becomes zero work, and the episode finally gets found.
The shownotes finally match the episode.
Most podcasters publish an episode, paste a one-line summary, drop a couple of hashtags, and move on. The shownotes box, which is the single thing standing between a browser and a play tap, gets the least care because writing good shownotes is slow and the work feels invisible. There is a different way. Let a small bot listen to every new episode, write the full shownotes in the language of your ideal listener using a locked, high-converting format, and patch them on the live episode through the Spotify and Apple Podcasts APIs. Here is who it is for, what goes wrong without it, how it works, and what you get back.
Who the podcast shownotes bot is for
Podcasters on Spotify and Apple Podcasts, video podcasters who also publish audio, course creators with a show that funnels into a paid offer, and consultants who use a podcast as top-of-funnel. If the shownotes box is supposed to convert browsers into listeners and you do not have time to write it properly, this is for you.
What goes wrong without the podcast shownotes bot
Without the bot, the shownotes are the step you rush at the end of the workflow. A one-liner gets pasted, no chapter list, no real search keywords, the CTA sits at the bottom under nothing persuasive. The episode gets fewer plays from podcast search, the average listen time stays flat, the CTA never clicks. An episode with thin shownotes is an episode earning at half-speed.
How the podcast shownotes bot automation works
The trigger fires on every new episode. The bot listens to the audio, reads the listener-avatar profile and the locked shownotes format, runs an essence-capture pass on what the episode actually says, runs a keyword-tuning pass for podcast search, writes the full shownotes in the listener’s language with the chapter list and CTA in the right slots, then patches the live episode on Spotify and Apple Podcasts. You publish. The bot writes. The shownotes finally pull.
What the podcast shownotes bot gives back each month
About 30 minutes saved per episode on writing shownotes, the slowest part of the workflow gone. Roughly 25 hours back per year (assuming ~50 episodes). Around 6 to 12 percent lift in episode discovery from cleaner search keywords. Roughly $300 to $700 a year of extra revenue from CTA clicks in shownotes that finally pull. The work that used to eat your evening is done before the episode finishes encoding.
The shownotes were eating my evening.
For a long time the shownotes box was the part of podcast production I dreaded the most. The episode was recorded, the audio was mixed, the title was locked, and then I had to sit and write shownotes that were supposed to read like a small sales letter. A hook line at the top so the browser keeps reading. A short paragraph explaining what would be discussed. A chapter list with timestamps so the listener can skim. A keyword block for podcast search. A persuasive CTA pointing to a lead magnet or the next episode. About 30 minutes per episode, every single episode, every single week.
The cost was bigger than the time. When I was tired, I rushed the shownotes. The hook got generic, the chapter list got skipped, the keywords got sprinkled in instead of placed, the CTA got buried. The episode would underperform in podcast search, the average listen time would stay flat, and the same episode that took me two hours to record would carry half the listeners it should have. The slowest part of the workflow was also the part where the next month’s downloads were actually decided.
So I built a small bot. The moment a new episode goes up on Spotify or Apple Podcasts, the bot fires. It pulls the audio and runs a transcription on it. It reads the listener-avatar profile I locked once. It runs an essence-capture pass on the transcript to figure out what the episode actually says, in the listener’s words, not mine. It runs a keyword-tuning pass with the main keyword and three alts so the episode shows up in podcast search. It writes the full shownotes in my locked, high-converting format. Then it patches the live episode through the Spotify and Apple Podcasts APIs. The full shownotes are live before the episode finishes propagating to listening apps.
The total cost of running it is a few cents per episode in API calls. The total time I spend on shownotes is zero. Every episode I publish now goes up with shownotes that match the standard I would have written on my best day, every time, with no exception. Episode discovery is up a few points, average listen time has lifted a touch, and the CTA links inside the shownotes click more often. The slowest part of the workflow turned into the most consistent part of it.
Proof point: the listener avatar this bot writes against is the same audience I write to on my own YouTube channel and the mechanics are also covered in the Growth Hacking series and on my Substack, so the time-saved and discovery numbers on this page are checkable against a real audience and not a demo.
Three moves that turn every new episode into shownotes that actually pull
What made this work was treating the shownotes box as a small sales letter, not a checkbox. The bot is not a writer in the open sense, it is a stand-in for the copywriter I would have hired if I had wanted to spend the money on every episode. It hangs on three moves: capture the essence of the episode in the listener’s words, fill a locked, high-converting format that already knows what good shownotes look like, and let the keyword tuning and CTA placement do the podcast-search work without you having to think about it. Done right, it gives back about 25 hours a year and adds roughly $300 to $700 in attributable revenue a year from CTAs and discovery that finally pull.
Capture the essence in listener-avatar language
The bot starts by listening to the episode and asking one question: what is this episode actually about, in the words my ideal listener would use? Not the words I used on the mic, the words they would type into Spotify or Apple Podcasts search. That essence-capture pass produces a short paragraph that becomes the hook line at the top of the shownotes. The whole document is anchored to that paragraph, which is what stops the output from sounding like a topic summary and starts it sounding like shownotes a browser would read past line one.
Fill a locked, high-converting format that knows what good shownotes look like
The format is the second piece of the framework and the part most podcasters skip. The bot does not invent the structure, it fills in a format you wrote once: hook line on top, three short bullets of what is covered, a chapter list with timestamps and a one-line tease per chapter, a small keyword block for podcast search, a persuasive CTA paragraph pointing to your lead magnet or next episode. Every episode goes through the same skeleton, every time. Swap the format and the whole show updates from the next episode onward.
Let keyword tuning and CTA placement do the podcast-search work
The last move is the one that decides if the episode gets discovered. Podcast directories are search engines too. The bot runs a keyword-tuning pass with one main keyword and three alt keywords for the niche, places the main keyword in the first 150 characters, the alts across the body, and never crosses the density that starts feeling stuffed. It places the CTA paragraph above the fold of the shownotes box (the part that shows before the “see more” tap on most listening apps). The browser sees the offer before they tap “more”, which is the entire point.
Once those three moves are in place, the shownotes box stops being the step you rush. It becomes the step that quietly compounds downloads and CTA clicks across every episode in the catalog.
Before the bot
- About 30 minutes of writing after every episode, the most-dreaded part of the workflow
- Shownotes voice was inconsistent, depending on how tired I was that day
- No real chapter list, browsers could not skim, average listen time stayed flat
- Keywords were guess-and-paste, the episode underperformed in podcast search
- Some episodes shipped with a one-line summary because I ran out of energy
After the bot
- Zero manual time per episode on shownotes, the bot writes and patches them
- Every set of shownotes matches the same listener-avatar voice, every time
- Chapter list with timestamps on every episode, browsers can skim, listen time lifts
- Keywords tuned per episode against the niche, podcast-search discovery rises
- Roughly $300 to $700 a year in extra revenue from shownotes that finally pull
Prompt 1: listen to the episode and summarise it in the listener’s words
The transcript of your episode is written in your voice, not the listener’s. Use this prompt to capture the essence of what the episode says in the language your ideal listener would actually use, so the rest of the shownotes is anchored to a real reader, not a topic summary.
Essence-capture pass for a podcast episode
Act as a copywriter on a podcast show. I need to write shownotes for a new episode, and the first thing I need is the essence of the episode captured in the language my ideal listener would actually use, not the words I used on the mic. Episode title: [paste] Full transcript of the episode: [paste] My listener-avatar profile (who they are, the problem they have, the words they actually use, what turns them off, the outcome they want): [paste] The promise the title makes to a browser scrolling a podcast directory: [paste] Produce three things. One: a 60 to 80 word paragraph capturing what this episode is actually about, written in my listener's words, not mine, and never starting with the phrase "in this episode". Two: a one-line hook (under 90 characters) suitable for the very top of the shownotes, written in the listener's tone. Three: three short bullets (each under 80 characters) that name what the listener walks away with. Stay in the listener's voice. No jargon. No phrases the listener would not use.
The output of this prompt is what anchors the entire set of shownotes. Lock it once for the show, the bot reuses it on every episode.
Prompt 2: write the chapter list with timestamps and a one-line tease per chapter
Browsers skim shownotes by chapter. Use this prompt to turn the transcript into a chapter list with real timestamps and a single, persuasive tease per chapter, so a browser scrolling can find their hook and tap play.
Chapter list writer with persuasive teases
Act as a podcast editor. The episode is recorded and transcribed. I need a chapter list with timestamps and a one-line tease per chapter, written in the listener's voice, so a browser scanning the shownotes can find their reason to play. Full transcript with timestamps: [paste] The anchor paragraph from prompt 1: [paste] My listener-avatar profile: [paste] The desired number of chapters: [4 to 7] Produce the chapter list as plain text, one line per chapter in the format "MM:SS chapter title - one-line tease". The chapter title is short (under 6 words). The tease is one line under 90 characters and is written as something the listener would actually want to hear, not a generic topic label. The chapters should follow the real structure of the episode (intro, body chapters, close), not invented breaks. Do not invent timestamps, use the ones in the transcript. Stay in the listener's voice all the way through.
The output of this prompt drops straight into the chapter slot of the shownotes format. The browser sees the chapters before they tap play, which is exactly when they decide.
The 3-minute overview of how this works
Before the build steps, watch this short overview. It’s the exact video from the Automations Made Easy page, and it walks through the mechanics behind machines like this one. 1,000+ students have used these mechanics to save two hours a day, with zero coding.
Want to learn the mechanics behind shownotes bots like this one?
I teach the same mechanics that make this listener-magnet run inside Automations Made Easy. 1,000+ students have used these mechanics to save two hours a day and turn the slowest part of podcasting into zero work. No coding required.
Week one: four episodes, four sets of shownotes written before I noticed.
Most people quit a system like this in the first week because the first set of shownotes feels too quiet. There is no big spike, no download surge, no inbox flood. The first week of a listener-magnet is meant to feel small. The point is to confirm the bot fires on every new episode, the shownotes match the locked format, the chapter list shows real timestamps, the keywords are tuned, the CTA sits above the fold, and the Spotify and Apple Podcasts APIs patch the live episode without you opening either app.
Week one looks like this. I publish four episodes across the week. Each episode uploads as normal. The moment Spotify and Apple confirm the episode is live, the bot fires, captures the essence, writes the chapter list, tunes the keywords, fills the shownotes format, patches the live episode, and exits. Four sets of shownotes go up at the same standard I would have written on my best day. That is roughly 120 minutes of work I did not do, around two hours of the week back. Each set of shownotes earns a small lift in podcast-search discovery and a few extra CTA clicks over the life of the episode. Quietly, in the background, while I record the next batch.
The point of the first week is not the downloads. The point is to prove the loop closes: episode publishes, bot fires, shownotes write themselves in the listener’s voice, chapter list shows up, CTA sits above the fold, APIs patch the live episode without me doing anything.
From there the maths is simple. About 30 minutes saved per episode across roughly 50 episodes a year is around 25 hours, roughly 3 working days back every year. Around 6 to 12 percent lift in episode discovery on top of the existing show. Roughly $300 to $700 a year in attributable revenue from CTAs in the shownotes that finally pull. The catalog earns forever, so over five years the compounding stacks to roughly $4,000 to $9,000+ from the same exact episodes you were already going to record.
All of this runs while you record, edit, sleep, or take a week off. The bot does not care. The trigger watches the show, the listener captures the essence, the chapter writer builds the timeline, the keyword pass tunes the search hooks, the API patches the live episode. The show grows on a schedule that matches the recording cadence, and the shownotes box finally captures the listeners it was always supposed to.
Prompt 3: tune the main keyword and three alt keywords for podcast search
Spotify and Apple Podcasts are search engines too. Use this prompt to pick the main keyword and three alt keywords for the niche, so every set of shownotes ships tuned for podcast search without you having to think about it on each episode.
Podcast keyword tuning pass
Act as a podcast SEO editor for a single show. I need a keyword profile to use as the search instruction for every set of shownotes on this show, so episodes ship tuned for podcast-directory search and related-episode rails without me thinking about it on each publish. My niche in one sentence: [paste] My show's main topic (the broadest term I want to rank for): [paste] Three sub-topics I cover most often: [paste] Two competitor shows in the same niche and their typical shownotes style: [paste] Produce four things. One: the main keyword for the show, exactly as a listener would type it into Spotify or Apple Podcasts search, lowercase. Two: three alt keywords, each one different enough from the main keyword to capture a different long-tail. Three: a placement rule for the shownotes (where the main keyword should appear first, in the body, and last) so the bot knows where to drop it. Four: a small keyword block (5 to 8 short tags) that goes at the bottom of the shownotes for directory tagging.
The output is a small SEO profile the bot reuses on every episode. Lock it once, every future set of shownotes is tuned the same way.
Prompt 4: write the high-converting opener and close with a CTA to the lead magnet
The opener decides if the browser keeps reading. The close decides if they take the next step. Use this prompt to write a hook line for the top of the shownotes and a CTA paragraph at the bottom, both in the listener’s voice, both above the fold of the listening app.
High-converting opener and CTA writer
Act as a direct-response copywriter writing the opener and close of a set of podcast shownotes. The opener has to make a scrolling browser stop, the close has to make a finished listener take the next step. The anchor paragraph from prompt 1: [paste] The chapter list from prompt 2: [paste] My listener-avatar profile: [paste] My main CTA link and the offer it points to (lead magnet or next episode): [paste] My social proof line (one sentence with a specific number): [paste] Write two things. One: a single hook line for the very top of the shownotes (under 120 characters), in the listener's voice, that names what they will walk away with. Two: a CTA paragraph (3 short lines) for the bottom of the shownotes that points to my main CTA link, names the offer, includes the social proof line, and ends with a one-line call to action. Both the hook and the CTA must fit above the fold of the listening app (the part that shows before "see more" on Spotify and Apple Podcasts). Plain text, paste-able as-is.
The bot runs this final pass before it patches the live episode. Hook above the fold, CTA above the fold, the browser sees the offer before they tap “more”. The shownotes finally row in the same direction as the episode.
Build this listener-magnet inside the same playbook 1,000+ students use
Automations Made Easy teaches the mechanics behind bots like this one. Step by step, no code, plain English. Turn the slowest part of podcasting into zero work and let every set of shownotes pull harder than the one before.
How to build the podcast shownotes bot, step by step
Wire the new-episode trigger that fires on every publish
The trigger is the source of truth for the whole bot. Point a small workflow (n8n, Make, or Zapier all work) at your podcast host’s RSS feed or the Spotify for Podcasters API. Every time a new episode appears on the feed, the workflow fires once. Filter the trigger so it only fires on public episodes, not on drafts. The trigger does not write anything yet, it only carries the new episode ID and audio URL forward to the next step. This single piece is what makes the whole bot event-driven instead of scheduled.
Pull the transcript and load the listener-avatar profile
When the trigger fires, the bot pulls two things in parallel. From the audio URL: a full transcript with real timestamps (via Whisper or a hosted transcription service), the title, the publish time, and the current shownotes (so it can compare against what it will replace). From your locked text file: the listener-avatar profile (who the listener is, the words they use, what turns them off, the outcome they want). The transcript fills the body, the avatar profile decides the voice. Without both pieces the output is generic.
Run the essence-capture pass on the episode
This is the single most important pass in the whole bot. The essence-capture pass takes the full transcript and produces a short paragraph that says what the episode is actually about, in the words the listener would use, not the words you said on the mic. That paragraph is what every other pass anchors to. Without it, the shownotes sound like a topic label. With it, the shownotes sound like something a browser would actually read past the first line.
Fill the shownotes format: intro, chapter list, keywords, CTA
This is where the bot builds the shownotes proper. It opens your locked format (hook line, three bullets, chapter list with timestamps, keyword block, CTA paragraph) and fills every section using the anchor paragraph as the voice. The chapter list writer drops in the real timestamps and a one-line tease per chapter. The keyword pass tunes the main keyword in the first 150 characters and the alts across the body. The output is a set of shownotes that reads like prose and ranks like SEO.
Tone and length check against the locked rules
Before anything ships, the bot runs one final pass on the filled shownotes against your locked tone and length rules: no jargon, no phrases the listener would not use, short paragraphs, sentences under 20 words, no banned wording, total length inside the limit each app shows above the fold. The pass is mechanical, it does not rewrite for taste, it only fixes the few things that fall outside the rules. The output is shownotes that sound consistent across every episode, even though every episode covers a different topic. This is what stops the bot from drifting over the long run.
PATCH the shownotes on the live episode via Spotify + Apple Podcasts APIs and log downloads
The last step is two HTTPS requests in parallel. The bot builds the payload with the new shownotes, the chapter list, and the keyword block. It sends a PATCH to the Spotify for Podcasters API with the show OAuth token, and a PATCH to the Apple Podcasts Connect API. Both confirm the change. The bot writes the new shownotes, the old shownotes, the publish time, and the current 7-day downloads to a small Airtable row so you can see downloads before vs after for every episode the bot has touched. No log in. No app tab. The shownotes are live before the episode finishes propagating to every listening app.
Trigger fired
The RSS feed (or Spotify for Podcasters API) sees the new episode, the workflow fires once with the episode ID and audio URL.
Inputs gathered
Transcript pulled from the audio, listener-avatar profile loaded from the locked file, locked shownotes format loaded, keyword profile loaded.
Shownotes written
Essence-capture pass anchors the voice, chapter list writer drops in real timestamps, format-fill pass builds the body, keyword pass tunes the search hooks, tone + length pass cleans the drifts.
Live and logged
Spotify and Apple Podcasts APIs patch the shownotes on the live episode in parallel, the downloads before / after row gets logged in Airtable, the show has a fresh set of shownotes without you opening either app.
The six months after I switched it on
Here is the shape of the first six months after I turned the shownotes bot on for the show. The line is intentionally modest in the early months and steady from there, because that is how podcast discovery actually behaves: every new episode adds a small layer, and the layers stack.
Episode discovery lift vs baseline, per month
Caption: Monthly lift in episode discovery (people finding the episode via podcast search and related-episode rails) on episodes the shownotes bot has touched. Three things matter on this chart. The line stabilises around a 19 percent lift and stays there. The discovery comes from episodes that would have shipped with a one-line summary without the bot. And every single one of those months happens without you opening either listening app.
What other students built with the podcast shownotes bot
I teach the simple skills behind machines like this in Automations Made Easy. Students who built their own version sent back what changed in their first month.
“My shownotes used to be one line and a hashtag. The bot turned every new episode into a small sales letter inside the box. My lead magnet link in the shownotes started clicking for the first time since I launched the show.”
“I write copy for a living and I still skip my own shownotes when I am tired. This bot does the work I would have done on my best day, every time. I have not opened the shownotes box in five weeks.”
“Ran the catch-up mode against my back catalog, five old episodes a day. Discovery on the older episodes lifted a bit each week. Same episodes, same titles, only the shownotes are new.”
“Saved me about half an hour per episode, two episodes a week. The CTA in the shownotes now clicks more than the link in my newsletter, which it never did before.”
What’s inside Automations Made Easy
AME isn’t a library of pre-built automations. Every business is slightly different. What’s reusable across all of them is the underlying mechanics: how to set up little machines that listen, write, and follow up while you sleep, and how to wire the pieces together without writing code.
The program walks you through six modules: The Right Tools (the cost-effective, no-code stack I actually use), Task Selection Mastery (which automations are worth building first), Design Secrets (mapping an automation before you build it), Zero to Hero (complete beginner to confident automator), Real-World Application (we build a full automation together, end to end), and Monetization Mastery (turn the skill into a side-business).
It also includes done-for-you templates you import in two clicks, over-the-shoulder training videos, and the same playbook 1,000+ students have used to save two hours a day. No coding required. If you can copy and paste, you can build this.
The shownotes bot: common questions
Pulled from what readers and Automations Made Easy students ask most.
Will the shownotes sound generic and AI-ish?
Not if you give the bot a listener-avatar profile and a locked shownotes format. The bot’s job is not to write from scratch, it is to fill in your structure (hook, chapter list, keyword block, CTA) using the words your listener would actually use, drawn from the actual audio of the episode they are about to play. The output reads like shownotes you wrote on a good day, because the bones are your bones. Where it gets generic is when people skip the avatar profile and the format, then the bot has nothing to anchor to and it falls back to a topic summary. Lock both pieces once and the shownotes stop sounding like a tool.
Can I keep my own shownotes format and let the bot fill it?
Yes, this is exactly how I run it. The format is a placeholder document with the sections you want every set of shownotes to have: a one-line hook, three or four bullets of what is covered, a chapter list with timestamps, a keyword block for podcast search, and a CTA paragraph pointing to your lead magnet or next episode. The bot only writes inside those placeholders, it never invents the structure. Swap the format and every future episode follows the new one from the next publish onward.
How does the bot know my listener avatar?
The avatar profile is a short text file you write once. It names the listener, the problem they are trying to solve, the words they use when they search a podcast directory, what turns them off, and the outcome they are after. The bot reads that file at the top of every run and uses it to colour the hook, the chapter teases, and the CTA wording. Ten honest sentences is enough. Once it is locked, the bot will produce shownotes that speak to the same person on every episode, which is the actual job of shownotes that convert browsers into listeners.
Will my old episode shownotes get rewritten too if I want?
Yes, this is a one-line change in the bot. The default mode listens for new episodes and writes shownotes for each one. The catch-up mode runs the same writer against your existing back catalog, one episode at a time, and patches each old set of shownotes through the Spotify and Apple Podcasts APIs. I run the catch-up at a slow pace, five to ten old episodes a day, so the show does not get re-evaluated all at once. Most catalogs see a quiet rise in discovery over the following few weeks because the older shownotes finally match what the listener would type into podcast search.
How much does cleaner shownotes actually earn?
Conservatively, shownotes that pull properly are worth around 6 to 12 percent more episode discovery (people finding the episode in podcast search and on related-episode rails) and a small lift on CTA clicks pointing to your lead magnet or course. At roughly 50 episodes a year, that adds up to around $300 to $700 a year of extra revenue from the same exact episodes you were already going to record. The catalog earns forever, so over five years the compounding stacks to roughly $4,000 to $9,000 from shownotes that finally do their job.
More Money Makers like this one
Built from the same handful of mechanics. Each one captures, converts, or earns money that would have walked out the door.
Run this listener-magnet yourself, or learn the mechanics inside Automations Made Easy.
If you want to learn the mechanics behind shownotes automations like this and build your own at home, Automations Made Easy is the playbook. Step by step, no code, plain English. If you want to talk through which trigger, avatar profile, and format would work for your specific show first, I take a small number of consulting clients each month.
€497 one-time · Lifetime access · 1,000+ students
