Faceless podcasts: what Spotify actually pays for

The short version

Read in September 2026 from Spotify's Partner Program page, monetization options, Partner Program monetization policies, payouts and Platform Rules.

Two of the three routes need video

Spotify describes the Partner Program as letting you "earn from multiple revenue streams depending on the types of content you publish and who streams it." That last clause is doing more work than it looks. Here is what each stream needs.

So a faceless audio podcast is not banned from anything. It simply cannot reach two of the three streams — and they are the two that any show with a camera gets by default. Every roundup promising "multiple revenue streams" is describing a video show.

One condition in that table deserves reading twice. Spotify states: "In order for a video episode to qualify for Premium video revenue, you need to insert at least one ad break in the episode." Premium members are the audience paying specifically not to hear ads; to be paid for reaching them, you must insert an ad break anyway. A show that never does earns nothing from Premium viewing, however much of it there is.

One road with two tolls

On 7 January 2026 Spotify cut the entry bar hard: from 12 episodes, 10,000 hours and 2,000 listeners down to 3 episodes, 2,000 consumption hours and 1,000 audience count, both measured over the last 30 days. That is the change the coverage picked up, and it is real.

What the coverage misses is the shape. YouTube gives you two roads and lets either carry you — watch hours or Shorts views. Spotify's requirements are a pure AND: five hundred devoted listeners racking up 2,000 hours do not qualify, and neither do three thousand casual ones who sample an episode.

Run the two numbers against each other and the three-episode minimum falls apart. 2,000 hours across 1,000 people is two hours per audience member per month. At the three-episode minimum, your entire catalogue would have to be two hours long — 40-minute episodes — and every one of your thousand listeners would have to finish every one of them. That does not happen. Assume a 50% average completion rate, which is our assumption rather than a Spotify figure, and you need four hours of catalogue: eight 30-minute episodes, not three.

Our arithmetic on Spotify's two published figures. The direction of the error matters more than the exact completion rate: any value below 100% pushes the required catalogue up, never down. Spotify separately notes that "a stream is counted when a listener or viewer on Spotify streams your episode or video for 60 seconds or more" — so a 60-second listen builds your audience count while contributing almost nothing to your consumption hours. The two tolls are paid in different currencies.

Treat "3 episodes" as paperwork and ignore it when planning. The number that governs your release schedule is two hours of consumption per listener, and the only lever you control is how much finished audio exists to be consumed.

The $10 that points both ways

Spotify uses the same figure for two opposite purposes, on two different clocks, and only one of them appears in the guides.

The friendly one is the payout floor. Spotify pays "your current balance after deducting any fees and taxes, as long as that balance amounts to $10 or more", and if it does not, "we'll continue to accumulate your revenue until it reaches $10 or more." Balances roll over. That is better treatment than TikTok's $50 threshold, whose terms say a below-threshold balance may be deemed expired, and it is a tenth of what AdSense requires.

The unfriendly one is a survival test, buried under a heading about supporting active shows:

Read the conjunction. Publishing is not enough on its own — you must also have earned $10 inside the window, which is about $1.67 a month. And because it is the same $10 that governs payment, the two rules describe the same show from opposite ends: a show that has never been paid is precisely the show this clause removes. The rollover that looks generous is a promise to keep counting money that the retention rule may stop you earning first.

What Spotify does not say anywhere we could find is what becomes of an accumulating sub-$10 balance when a show is removed under this rule. The gap is unresolved, and it is the one question worth asking support before planning around the rollover.

The threshold to aim at is 100, not 1,000

Every podcast monetisation guide points at the Partner Program. Spotify runs a second route with a bar a tenth as high, and for a small show it is reachable roughly a year earlier.

A 10x lower audience bar, no hours requirement, and double the revenue share. The catch is what you are selling: "subscriber-only episodes that your fans can access for a monthly fee" means asking an audience to pay for access to you — a harder sell for a show with a synthetic narrator and no host to be attached to. It is the same pattern this site found on Instagram, where the reachable routes are the parasocial ones.

That does not make it the wrong target. It makes it an honest one: at 100 listeners, 30 subscribers at $5 clears more than the ad share will pay at ten times the audience, and until 19 October 2026 Spotify takes none of it beyond a stated 5.5% payment processing fee. If your show cannot sell that, a 50% share of an ad pool is unlikely to rescue it.

Correction, September 2026 — this route is repriced on 20 October 2026. Spotify’s fee schedule states that Podcast Subscriptions “is evolving into Memberships on October 20th, 2026”, that current terms “will remain in effect until October 19th, 2026”, and that the replacement is a “Platform fee: 10% (includes payment processing fees, foreign exchange fees, and payout fees), deducted from net earnings”. The 5.5% figure above, and the “100% of your earnings” wording in the table, describe the arrangement only until that date. The old structure was 5.5% plus foreign exchange fees of 0.8–1.6% where they applied, so a creator who never converted currency sees the larger jump — 5.5% to 10%, about 82% more. Existing subscribers who migrate keep the old terms “until the subscriber cancels”, which makes a subscriber signed up before the date permanently worth more than one signed up after it. Full breakdown in our currency conversion fees page.

Added September 2026 — the obvious response to that correction is probably wrong. Going from 5.5% to 10% reads like a reason to take the show to Patreon instead, and on the published rates it is not. Spotify says its 10% already covers processing, foreign exchange and payout fees; Patreon’s pricing page states the same 10% figure and then adds that it is charged “plus payment processing, currency conversion, and payout fees”, none of which it publishes anywhere we can read. Buy Me a Coffee advertises 5% and its own fee page lists four further charges that take a $5 recurring international membership to 15.9%. The flat 10% is still the cheapest published number of the three — what you actually give up by staying is the exportable email list Patreon includes. We compare all three in Patreon vs Buy Me a Coffee vs Spotify fees.

What Spotify restricts is consent, not AI

The genre states as fact that Spotify requires you to disclose AI use in your show description. We could not find that rule. It is not in the community guidelines, not in the Partner Program monetization policies, and not in Spotify's own May 2026 announcement about AI and podcasts. What that announcement says is narrower and only about consent:

That is the whole of it. A stock synthetic voice belonging to nobody real is not described; a clone of someone you lack permission from is described twice over, since the Platform Rules separately restrict "posing as another person, brand, or organization in a misleading manner." The contrast with YouTube is worth knowing if you publish to both: YouTube requires disclosure of realistic synthetic content and exempts cloning your own voice, while Spotify publishes no disclosure duty at all and restricts cloning someone else's. A third variant exists too: Amazon KDP requires a declaration of AI-generated text, images or translations on every publish and every republish, but makes it privately to Amazon rather than to readers. Three platforms, three regimes, no portable answer. Disclosing anyway costs nothing — just do not repeat, as settled fact, a rule Spotify has not written.

Three clauses that do bite are rarely mentioned. The monetization policies make ineligible "content with a disproportionate host-read ad or sponsorship load" — aimed squarely at the "scale sponsor reads" model this page used to recommend — and "content with audio that consists primarily of non-speech noises or sounds", which rules out the ambient and sleep-sound shows pitched as the easiest faceless format. The community guidelines then define spam to include "content made solely to improve SEO or promote a product or service." The load-bearing word is solely: a show that recommends things is fine; one that exists only as a funnel is what the clause describes.

What the narration bill has to earn

Spotify's payout floor is the lowest of any programme on this site, and it is still less than half of one month's narration. ElevenLabs Creator is $22 a month for 121,000 credits at "1 credit per character", with a 50%-off first month. Spotify needs to owe you $10 before it sends anything. You must clear that floor 2.2 times over just to cover the voice.

Set it against the retention rule and the picture gets starker: staying in the Partner Program requires about $1.67 a month, so a show can satisfy Spotify that it is worth keeping while losing roughly thirteen times that on narration alone. The programme's floor is not a business test and should not be mistaken for one.

Capacity is not the constraint. At roughly 900 characters a minute of speech — our estimate, not a vendor figure — 121,000 credits cover about 134 minutes a month, so four 30-minute episodes use about 89% of the entry plan: a weekly show on the $22 tier, with the eight-episode catalogue the entry gate implies taking about 1.8 months of credits. You will run short of listeners long before you run short of characters, so upgrading the voice plan is almost never the fix. If the line needs to be cheaper, our alternatives priced per finished hour put Fish Audio near $3.30 against roughly $9.82 on ElevenLabs Creator, and what AI narration costs per hour converts every published meter into the same unit.

Hosting on Spotify for Creators is required for the Partner Program. Note also that the commercial licence you need for any monetised show starts at ElevenLabs' $6 Starter tier — the free plan excludes it.

What we could not verify

• Any per-stream or RPM figure. Spotify publishes none for podcasts, and describes Premium video revenue only as "calculated in accordance with our proprietary payment formula, which may be adjusted by us from time to time." No dollar-per-thousand figure appears anywhere above. The previous version of this page promised readers they could "scale sponsor views" and quoted no source for any of it.
• What "audience count" and "consumption hours" formally mean. Spotify uses both as eligibility thresholds and defines neither. It defines only a stream, at 60 seconds. Whether audience count is unique listeners over 30 days is strongly implied — the monetization options page calls it "unique listeners" — but the two pages use different words for the same gate.
• What happens to a sub-$10 balance when a show is removed for earning under $10 in six months. Unaddressed on every page we read. Flagged above as the question to ask support.
• The completion rate behind the catalogue arithmetic. The 50% figure is ours. Spotify publishes no completion benchmark. Only the direction is certain: lower completion means more catalogue needed, never less.
• Whether Spotify has any AI-disclosure rule at all. We searched the community guidelines, the monetization policies, the Platform Rules and the May 2026 announcement and found none. Reporting an absence is weaker than reporting a permission — Spotify could publish one tomorrow.
• How any of it is enforced. Everything here is published wording. We have not tested how the impersonation policy or the six-month rule are applied. A clause is not a prediction.

A September 2026 snapshot of pages that change without announcement — the entry thresholds moved in January 2026 and the eligible-market list has since grown to 19. The two entry numbers, the $10 rules and the video conditions are the load-bearing facts; a change to any of them moves the conclusion rather than a detail.

FAQ

Related reads: the blog index, voice cloning for long-form audio, TikTok's $50 rule, Pinterest's affiliate rules, and what affiliate programs require before they pay.