The gap Substack never closed#
Substack solved distribution for independent writers. It did not solve my problem.
Newsletter open rates on Substack commonly run somewhere between 30 and 50 percent, which is excellent by email standards. Read-through is a different number entirely, and it is much lower. Long-form written content generally sees only 20 to 30 percent of readers reach the end.
So the typical issue you ship reaches an inbox, gets opened by roughly a third to a half of your list, and gets finished by a fraction of those. The people who did not finish are not uninterested. They opened it. They meant to come back. Then the day happened.
Audio addresses that specific failure, and only that one. It does not find you new subscribers, which matters for how you read the promise in this article's title. What it does is convert people who already subscribed but never actually consume what you send.
What Substack gives you for free#
Before considering any external tool, know what is already included, because it is more than most writers realise and it is genuinely good.
Substack includes full podcast hosting with automatic RSS distribution to Apple Podcasts and Spotify. It is free to host and distribute, and writers running paid subscriptions keep 90 percent of revenue minus card fees.
The feature list is deep:
- Auto-generated podcast RSS, compatible with any podcast player
- Private subscriber-only feeds, where each paid subscriber gets a unique private RSS URL
- Per-section feeds, so different series can have separate RSS
- Podcast import from another host via RSS, Apple, or Spotify link
- Analytics by episode, player app, and country
- Transcript upload and an inline transcript editor
- Human voiceover upload, if you want to record it yourself
- AI voiceover: automatic audio narration of written posts, with six built-in voices
- Syndicate voiceover to RSS, so the generated audio goes into your podcast feed
There are also monetisation mechanics worth knowing. You can publish an episode as paid-only and attach a free preview version for non-paying subscribers. You can ship a full ad-free episode to paid subscribers and a sponsored version to everyone else. The early-access setting posts an episode to paying subscribers first via private feed, then releases it publicly later.
If your goal is simply "my subscribers should be able to listen to this," Substack's native voiceover does that at zero cost and zero additional workflow. Any article that skips past this to sell you something is not being straight with you.
Where the native voiceover stops working#
The honest limitation is not the quality of the voices. It is what the feature is doing.
Substack's AI voiceover narrates your post. It reads your written words aloud, in order, in one voice. That is text-to-speech, and text-to-speech inherits every structural property of the thing it is reading.
Written essays are built for eyes. They use paragraph breaks, subheadings, pull quotes, bullet lists, italics, and links to signal structure. A reader skims ahead, glances back, and navigates visually. None of that survives narration. Read aloud, a well-structured essay becomes an undifferentiated stream where the listener cannot see where they are, cannot skim, and cannot flip back when they lose the thread.
The second problem is that written prose assumes a reader who can re-read. Spoken information is transient: once a sentence has passed, it is gone unless the listener holds it in working memory. Sentences that work beautifully on the page, long ones with subordinate clauses and parenthetical asides, are difficult to follow when they arrive once, at someone else's pace, while the listener is driving.
This is why a narrated essay often gets abandoned at minute three even though the written version was good. The content is fine. The format conversion did not happen. Something was read aloud, but nothing was rewritten for the ear.
Where the native feature genuinely is the right answer: short posts, personal letters, anything already written conversationally, and accessibility, giving subscribers who prefer or need audio a way to consume what you publish. For those cases, it is free, instant, and sufficient. Use it.
Where it struggles: long analytical essays, anything technical, anything with data or structure that the reader was expected to navigate visually, and anything you want to work as a standalone podcast that stands up next to other podcasts in someone's feed.
Does audio actually double your audience?#
Being precise about the claim, because the honest version is more useful than the marketing version.
Audio does not double your subscriber count. Nothing about adding an audio version makes new people discover you. Discovery comes from search, referrals, recommendations, and the work covered in our guide to growing a podcast audience without ads.
What audio can roughly double is the effective reach of each issue. If 25 percent of openers finish your written piece, and an audio version gets consumed by a meaningful share of those who would not have finished, the number of people who actually received your argument goes up substantially without a single new subscriber. Podcast episodes typically see far higher completion than long written pieces, commonly cited around 70 to 80 percent, because listening is a mode people default to finishing.
The mechanism is that audio reaches a block of time that written content structurally cannot: commutes, workouts, dog walks, dishes, folding laundry. That is real, uncontested attention. Your subscribers are not choosing your essay over a competitor's essay during those hours. They are choosing your essay over nothing, because reading is impossible.
There is a second effect that matters more for paid newsletters, covered next.
Audio as a paid tier#
This is the strategic reason to care, and it is underused.
Audio is a genuinely iterable product. The same content in two delivery modes, where one of them is uniquely valuable to a busy professional, gives you a paid wedge that does not require writing anything additional.
Common configurations:
- Text free, audio paid. Everyone gets the essay. Paying subscribers get the listenable version. You are not paywalling ideas, which many writers dislike doing, you are paywalling convenience.
- Both free, early access paid. Paid subscribers get the episode days before it opens publicly, using Substack's early-access setting.
- Free version sponsored, paid version clean. Substack supports shipping an ad-free episode to paying subscribers and a sponsored one to everyone else.
- Free preview, full episode paid. Attach a short free version so non-paying subscribers hear enough to want the rest.
The reason this works better than most upsells is that it does not ask anyone to value your writing more highly. It asks whether they would rather listen than read, which is a question about their day rather than about you.These same configurations work off Substack too, with the platform-by-platform detail in how to turn your newsletter into a podcast.
One platform detail to know: private RSS feeds work across podcast players, and Substack now supports linking a Substack account with Spotify so paying subscribers can access paid episodes there too.
Which option fits you#
Three realistic paths, and the right one depends on what you are optimising for.
Record it yourself. Best when your voice, personality, and delivery are part of why people subscribe. Nothing substitutes for a person who is genuinely good company. Cost is time: reading a 2,000-word essay well takes multiple takes, and editing takes longer than people expect. If you have never recorded anything, our guide on starting a podcast with no equipment covers doing it with a phone and a closet.
Use Substack's built-in voiceover. Best when you want audio to exist with zero additional workflow, your posts are short or conversational, and accessibility is the main goal. Free, instant, and already integrated with your feed.
Convert to a restructured conversational episode. Best when your essays are long or analytical, when you want the audio to work as a real podcast rather than a narration, and when you are treating audio as a paid tier that has to feel worth paying for.
The failure mode to avoid is picking option three when option two would have done, or picking option two and then being surprised that a narrated 3,000-word analytical essay does not hold listeners.
Where Sprep fits#
We build a document-to-podcast tool, so treat this as an interested party talking.
The distinction from Substack's native voiceover is the one described above, and it is the whole argument: Sprep does not read your essay aloud. It restructures it into a two-host conversation, where one voice asks the questions your reader would have asked and the other delivers your argument. That format exists because it solves the transience problem: questions create natural checkpoints, and a listener who drifts can rejoin at the next one.
Every word of the generated script is editable before any audio renders, which is what makes the episode publishable under your own name rather than something that sounds vaguely like you. You can fix phrasing, correct pronunciation of names and acronyms, and add a spoken call to action pointing back to the issue. An approved script can also be generated in over 70 languages, which matters if your list spans countries.
For the step-by-step version of this workflow, from published Substack issue to distributed episode, we have a separate playbook: turn your newsletter into a Spotify show in 15 minutes.
Where it is the wrong tool. If your subscribers are there for your actual voice, record yourself. If your posts are short and conversational, Substack's free feature is genuinely sufficient, and you should use it rather than adding a step. And if a piece was unclear in writing, conversion will produce a clear recording of unclear thinking. The format change is faithful to what you give it.
This is one application of a broader idea covered in our guide to content repurposing: one asset can serve several different moments of your audience's attention, provided each version is designed for a different moment rather than being the same thing cut smaller.
FAQ#
Does Substack have a podcast feature? Yes, and it is comprehensive. Substack includes free podcast hosting, auto-generated RSS distribution to Apple Podcasts and Spotify, private subscriber-only feeds, per-section feeds, podcast import from other hosts, analytics by episode and player, and transcript upload. Writers running paid subscriptions keep 90 percent of revenue minus card fees.
Can Substack turn my posts into audio automatically? Yes. Substack includes an AI voiceover feature that generates audio narration of written posts using six built-in voices, and that audio can be syndicated into your podcast RSS feed. It is free and requires no additional workflow, which makes it the sensible default for short or conversational posts.
What is the difference between Substack's AI voiceover and a document-to-podcast tool? Substack's voiceover narrates your existing text aloud in a single voice. A document-to-podcast tool restructures the material into a conversational format designed for listening, with different pacing and structure than the written original. Narration inherits the structure of an essay written for eyes, which is why long analytical pieces often lose listeners when simply read aloud.
Will adding a podcast actually grow my Substack subscriber count? Not directly. Audio does not create discovery on its own. What it does is increase how much of your existing list actually consumes each issue, since podcast completion rates run far higher than long-form reading completion. Growth in subscriber numbers still comes from search, referrals, recommendations, and cross-promotion.
Can I make the audio version paid and keep the text free? Yes, and it is one of the more effective configurations. Substack supports paid-only episodes, free preview versions of paid episodes, ad-supported free versions alongside ad-free paid versions, and early access where paying subscribers receive an episode before it opens publicly.
How do paid subscribers listen to private Substack podcasts? Each subscriber receives a private RSS feed unique to them, which they can add to the podcast player of their choice, and the link is included in their welcome email as well as on the episode page. Substack also supports linking a Substack account with Spotify so paid episodes are accessible there.
Do I need a separate podcast host if I publish on Substack? Usually not. Substack's built-in hosting generates the RSS feed and handles distribution to major platforms, including private feeds for paid subscribers. External hosting mainly makes sense if you want your show to exist independently of your newsletter or need features Substack does not offer.
How long should a Substack audio edition be? Match it to the essay rather than to a target. A 2,000-word piece typically becomes ten to fifteen minutes, which fits a commute. What matters more than length is whether the audio was restructured for listening, since a long narrated essay loses people faster than a long conversational one.
Should I record my Substack posts myself or generate the audio? Record yourself if your voice and personality are part of why people subscribe, since nothing substitutes for that. Generate the audio if the writing is the product, if recording consistently is what would cause you to stop, or if you want audio available for every issue rather than occasionally.
Does publishing the same content as text and audio hurt SEO? No. Text and audio are different media rather than competing pages, so there is no duplicate content issue. Duplication concerns only arise when substantially identical written content sits at multiple URLs you control, which is resolved with canonical tags.
Try it on one issue#
If your essays are long enough that narration would not hold listeners, the free plan turns one piece you have already published into a full conversational episode of up to 15 minutes, with the script editor included so you can fix every word before any audio is generated.
See it in action