Many of the biggest music platforms pool everyone's subscription fees, then split them by each artist's share of all streams, rather than paying only the artists each subscriber plays. Here's an illustrative estimate of what that could mean for the artists you listen to.
Pick whichever is easier. You can switch later.
Once your file is read, your top artists appear here for you to check. No file yet? .
Why are some results a range? ⓘWe don't know how big your “other” artists are, so their shared-pool amount is a range. The low end assumes every one is tiny (about 40 streams a month, under the payout minimum). The high end assumes every one is as big as one of the platform's ~80 biggest artists (about 240,000,000 streams a month), capped so your fee can't be paid out twice. For most listeners the truth is much nearer the low end. You can edit both under Global market assumptions; the sources are in the Methodology. Uploading your history replaces the artist count with your actual count.
Greyed-out rows are examples: click one and type to replace it with your own artist. Leave a row blank to skip it. Not sure how big an artist is? ⓘRough size is how many streams the artist gets on the platform in a month (we scale it up for your lifetime view). Pick the closest round number. Or switch to Listeners and type the “monthly listeners” number from the artist's profile page (on Spotify, at open.spotify.com); we'll estimate a range, using 2.4–20 streams per listener.
For a second opinion, kworb.net/spotify is an unofficial tracker that estimates stream counts from Spotify's public charts. Platforms don't publish exact stream counts for each artist, and small or independent artists often have no public number at all. That's fine here: the goal is to show how the payout methods work, not to produce a royalty statement. The sizes come from published figures (see the Methodology).
What you pay each month, and how much you listen in total. The defaults are a reasonable starting point.
How a fee is split is a business choice, not a law of nature. Most listeners have never been told how theirs is divided, so the first step is the easiest one: show them yours.
These aren't open yet. Open one to see what's planned.
All quantities refer to one listener over a period of M months. Subscriptions are paid monthly, so the pot of money available to artists scales with the period:
Your net fee is divided only among the artists you streamed, in proportion to your own listening. Artist i receives:
By construction Σi Ui = F: every dollar you pay reaches an artist you actually played, and no one else's listening affects your money.
All subscribers' fees are pooled. Each rights-holder is paid its share of total platform streams, provided it clears the payout threshold. T is a number of streams per year and Gi is streams per month, so the test is 12 Gi ≥ T, about 83 streams a month with the default T = 1,000 (1[·] is 1 if the condition holds and 0 otherwise):
The implied average payout per stream is r = R/G. The part of your fee that ends up with artist i is the same pool share applied to your fee:
Note that si does not appear in (4). A single listener is a negligible share of G, so the pool is effectively exogenous: how much you play an artist does not change what your fee sends them. Your money flows to artists in proportion to how popular they are with everyone, and Σi Pi ≤ F.
For each artist we report the proportional gap between the two models. The aggregate gap, shown at the top of your results and on the share image as the “top-N gap,” covers only the listed artists for which a global stream count was entered:
If Pi is exactly zero because the artist is below the threshold, the calculator shows “Nothing”. If you never streamed the artist, Ui = 0 and the gap is undefined (“n/a”). Because both models use the same F, the dollar figures are directly comparable.
Your listed artists are usually a minority of your listening. Dividing by your total streams S (not by the listed streams) keeps their shares honest, and the remainder is modelled as one extra line. If the listed streams ever exceed the total, S is set to the listed streams so no share exceeds 100%.
Under user-centric payment the tail simply gets its share of your listening:
Assumption: a range, not a point. Under the pro-rata pool the answer depends on who those other artists are, which your history does not reveal. So we bound it. The minimum treats all K of them as tiny, with gmin platform-wide streams each (default 40 per month, the lowest size tier — below the roughly 83 a month implied by a 1,000-streams-a-year payout minimum, so they earn nothing). The maximum treats all of them as being as big as one of the platform's ~80 biggest artists, with gmax each (default 240 million per month, the “Top 80” reference point: the size of one of the platform's ~80 biggest artists; see the evidence below), capped so your fee cannot be paid out more than once:
K comes from the “other artists” quick estimate when you type your data in, and is counted exactly from your history (distinct artists minus those listed) when you upload it. The rest of your fee under the pro-rata pool goes to artists you never streamed, which is also a range because it is whatever the tail does not take:
The “Where your fee goes” card sums to F at every point in that range: user-centric = listed + tail (nothing to artists you never streamed); pro-rata = listed + tail + never streamed. Because (9) and (10) are linear in K and g (until the cap binds), doubling either doubles the corresponding end of the tail range; the amounts for your listed artists are unaffected. The share of your fee reaching artists you actually stream is therefore also a range, from (Σi Pi + Ptailmin)/F to (Σi Pi + Ptailmax)/F.
The greyed-out starter rows are fictional listening habits, not real artists. They open in “Monthly listeners” mode, which gives ranges; switch the toggle to “Rough size” to follow this single-value example. With p = $5.50, τ = 30% and M = 1, F = $3.85, and the rows have 24, 5, 3, 8 and 1 streams out of S = 413. The first (Gi = 100 million streams a month): Ui = 3.85 × 24/413 ≈ $0.224, but Pi = 3.85 × 100×10⁶/266×10⁹ ≈ $0.0014, a gap of about −99%. The last (a 1-billion-a-month hit, played once): Ui ≈ $0.009 but Pi = 3.85 × 10⁹/266×10⁹ ≈ $0.014, a gap of about +55%: the biggest hits collect more under the pool than your own listening would send them. The lo-fi row (Gi = 100) clears the payout minimum (1,200 streams a year), but its pool share of your fee is about a millionth of a cent, so it shows as “<$0.001”. The five rows cover 41 of 413 streams, so L = 372 and Utail = 3.85 × 372/413 ≈ $3.47. With K = 200 other artists (the typical subscriber), Ptailmin = $0 (tiny artists fall under the threshold) and Ptailmax = 200 × 3.85 × 240×10⁶/266×10⁹ ≈ $0.69. So under the pro-rata pool between about 0.4% and 18% of the fee reaches artists you actually stream, and the remaining $3.14–$3.83 goes to artists you never streamed.
The “most recent month” view uses M = 1 and your streams from the last 30 days of your history. The “lifetime” view uses all your streams and sets M to the number of months your history spans, so F is your monthly net fee multiplied by that number. Platform-side inputs (Gi, G and the pool R) are entered per month. The shares Gi/G are ratios on the same monthly basis, so they don't change with the period, and your pro-rata dollars Pi grow with F exactly as the user-centric ones do. Pool-wide figures, such as an artist's total modelled payout from the pool, are multiplied by M: Πi × M over the whole period.
Three dropdowns fill in numbers people usually don't know offhand. They only pre-fill the inputs above, and every number stays editable.
Your streams from your listening habit. Streams are estimated from hours listened per day h, an average track length ℓ = 3.5 minutes, and d = 30 days:
An artist's global size (Gi). The “Rough size” dropdown offers round numbers of streams per month platform-wide, each ten times the last: 100, 1 thousand, 10 thousand, and so on up to 1 billion. They are deliberately coarse: the point is the order of magnitude, not a precise count. The table in the evidence section below gives real reference points for what those sizes mean (for example, only about 80 artists are above roughly 240 million a month). Alternatively, switch the toggle to “Listeners” and enter the artist's monthly listeners; only the selected mode is used. The long-tail range uses about 240 million a month as its high-end default (gmax, the size of the platform's ~80 biggest artists) and 40 a month as its low end (gmin, under the payout minimum).
Other artists you play (K). Tiers of about 10, 40, 100, 200 and 400 other artists. The default is 200, because the platform reports that its average subscriber listens to more than 200 different artists a month (see the evidence below). Uploading your history replaces this with a count taken from your data: the number of distinct artists in the period minus the ones listed.
Few people know an artist's platform-wide streams, but the platform shows every artist's monthly listeners on their profile page: the number of unique people who played them in the last 28 days (definition from the platform). That counts people, not streams, so it cannot stand in for Gi directly. We convert it with a ratio of streams per listener, ρ, and because that ratio varies a great deal between artists we use a range, not a single value:
Equation (4) is then applied at each end, so Pi becomes a range: Pimin = 1[12 Gimin ≥ T] · F · Gimin/G, and likewise for max. Everything downstream (the gap in (5), the top-N gap in (6), the “Where your fee goes” card) carries both ends, and ranges are combined like with like (all lows together, all highs together), so the fee still sums to F at each end. The toggle above the artist list decides which is used: a rough size or the listener count, never both.
Where the defaults come from (ρmin = 2.4, ρmax = 20 streams per listener per month):
Reading the result. The ratio range is a factor of about eight, so an artist entered by listeners gets a pro-rata amount spanning a factor of about eight. That is deliberate: it is as precise as the evidence allows. Choosing a rough size instead gives a single value. The low end rests on this model's payout rate, so a higher true rate would lower it (toward 1, the minimum, since a monthly listener has by definition played at least once); the high end uses reported stream counts directly and does not depend on it.
These reference points, used for the long-tail range and for judging what a rough size means, follow the platform's published royalty brackets rather than a guess. Each bracket is “at least” a threshold, so the typical member of a tier is somewhat larger than the number shown. Royalty amounts are converted to streams with this model's own implied payout rate, r = R/G = $917 million ÷ 266 billion ≈ $0.00345 per stream: streams per month = royalties per year ÷ r ÷ 12. The royalty figures are for royalties generated on the platform (recording and publishing), which is also what R measures.
| Tier | Used | Evidence |
|---|---|---|
| The top artist | 2.2B / mo | The most-streamed artist of 2024 had 26.6 billion streams that year (Associated Press). 26.6B ÷ 12 ≈ 2.2B. For comparison, the 2025 leader had 19.8 billion (≈ 1.65B a month) (Rolling Stone). |
| Top 80 | 240M / mo | In 2025, the 80+ top artists each generated over $10 million in royalties on the platform (Spotify, Loud & Clear 2026; also Music Ally). $10M ÷ $0.00345 ≈ 2.9B streams a year ≈ 240M a month. |
| Top 1,500 | 24M / mo | More than 1,500 artists generated over $1 million (same sources) → ≈ 24M a month. Spotify notes that capturing 1% of streams from 1% of listeners is enough to earn $1 million a year. |
| Top 13,800 | 2.4M / mo | More than 13,800 artists generated at least $100,000 (same sources) → ≈ 2.4M a month. |
| Top 81,000 | 240K / mo | Over 81,000 artists earned at least $10,000 (Spotify, Loud & Clear FAQ) → ≈ 240K a month. Cross-check: the 100,000th-ranked artist generated just over $7,300 in 2025 (Spotify, Loud & Clear 2026) ≈ 176K a month. |
| The smallest | 40 / mo | 88% of 253 million tracks received 1,000 or fewer streams in 2025 (Luminate via Hypebot). Under the platform's royalty policy a track needs at least 1,000 streams in the past year to earn recording royalties (Spotify FAQ), about 83 a month; 40 a month is below it. Such tracks are about 0.5% of all streams. |
Caveats on the conversion. The platform says it does not pay a fixed rate per stream (Spotify FAQ), so r is our own pool-divided-by-streams figure, and the stream counts above rest on it: a true rate twice as high would halve every tier. The ladder is geometric (each tier ten times the last), which matches the reported brackets only roughly. The smallest reference point uses an annual minimum that, in the real rules, applies per track.
Platform-wide inputs and their checks.
Draft wording: to be reviewed by legal counsel before launch.
This is an independent project. It is not affiliated with, endorsed by or sponsored by Spotify, Apple, Tidal, Qobuz, Instagram, Meta, X, Bluesky, Bandcamp, Patreon, Luminate or any other company or person named on this page. Those names are trademarks of their owners and are used only to identify them, for example to explain how to download your own data or to cite a source. The greyed-out example rows are fictional.