Verified
A number from a source that knows the truth: your distributor statement, a CMO payout (PRS, GEMA, ASCAP, APRA, TONO, Gramo), or a directly-read contract clause.
Methodology
Recoup is a contract reader and an earnings projector. This page says, in plain English, exactly how we know what we know — and where the gaps are. The confidence chips throughout the app link here.
Rates verified 2026-Q2 · Refreshed quarterly · 35 markets · 6 streaming platforms
Recoup focuses on what your record contract actually says, and what your music is reasonably earning right now from streaming and radio.
In scope
Out of scope
Every number in Recoup carries one of three labels. The label tells you how directly we measured it.
A number from a source that knows the truth: your distributor statement, a CMO payout (PRS, GEMA, ASCAP, APRA, TONO, Gramo), or a directly-read contract clause.
A number sourced from a reliable third-party tracker (Soundcharts, public broadcaster APIs, scraped public charts) and multiplied by published country rates. Within a few percent of what statements will show, but not directly verified.
An estimate built from published country averages — used when no direct or third-party data exists for that market or income stream. Useful for a ballpark, not for a tax return.
Internally we track seven provenance tiers. Each tier maps to one of the three user-facing labels above.
| Tier | What it is | User label |
|---|---|---|
| oauth | Connected to your account — distributor login, Apple for Artists | verified |
| native | Read from a first-party broadcaster API (NRK PSAPI, SR open data, BBC RMS, ABC AU) | verified |
| statement | Parsed from a CMO or distributor statement you uploaded (TONO, Gramo, DistroKid, TuneCore, CD Baby) | verified |
| wrapi | Third-party tracker API (Soundcharts — covers all major DSPs and 200+ countries) | tracked |
| scraped | Public charts and discovery feeds parsed without an official API | tracked |
| derived | Calculated from another tracked input plus a published rate (radio plays × country radio rate) | tracked |
| modeled | Country average applied where no per-song signal exists yet | modeled |
The rollup rule for any aggregated number is “worst label wins” — if one ingredient is modeled, the total reads as modeled.
Recoup ships per-country rates for 35 markets across 6 streaming platforms (Spotify, Apple Music, Tidal, YouTube Music, Amazon, Deezer). Each market carries a confidence tag indicating how tightly its rate is triangulated against public sources.
Per-country rates cross-checked against multiple published sources (Spotify Loud & Clear, Chartlex, IFPI, MBW). Recoup's estimates here are the closest to what statements will show.
One solid published source per market. Country-specific publishing-share splits applied; rates within the typical industry range but not as tightly triangulated as the high-confidence set.
Interpolated from regional averages or single older sources. Useful for a directional estimate; treat the number as a wide band, not a precise prediction.
Streaming data is “tracked” across every market here (Soundcharts). Radio is modeled at V1; native radio connectors land in a follow-up.
A single “per-stream rate” doesn't exist. What you actually earn from one play depends on five separate factors, each modeled explicitly:
Spotify, Apple, Tidal, YouTube Music, Amazon, and Deezer all pay different rates per stream. Apple typically pays ~1.6× Spotify; Tidal HiFi ~3×; YouTube Music ~0.4×. Each platform has its own column in our rate table.
The same platform pays differently per listener country. Spotify in Norway pays ~6× what it pays in India. We sum contributions per country rather than averaging across them.
Spotify pools subscription revenue and pays out pro-rata. Deezer uses ACPS (user-centric, paid only when subscribers actually played you). Our rates are net of typical pool deductions for each platform.
Roughly 15% of every stream's gross goes to writers (CISAC/BIEM convention). That 15% slice splits between performance (radio-play side) and mechanical (sheet-music side) — and the split is country-specific: Norway 70/30, Germany 33/67, UK 50/50, Netherlands 75/25. Recoup's projection respects this per-country split instead of using one global ratio.
Some contracts further reduce mechanical income — most notably the US Controlled Composition Clause (major-label US deals cut mech by 25–50%). When the contract reader detects this clause, the mechanical line in your projection drops accordingly. Performance and master are unaffected.
Worked example — Norway vs Germany
Same song, same artist, 1,000 Spotify streams. Watch how the writer slice flips depending on the listener country:
Norway (70/30 perf/mech split) Gross from 1,000 streams ≈ $4.30 → Writer slice (15%) ≈ $0.65 → Performance (70%) ≈ $0.45 → Mechanical (30%) ≈ $0.20 Germany (33/67 perf/mech split) Gross from 1,000 streams ≈ $4.20 → Writer slice (15%) ≈ $0.63 → Performance (33%) ≈ $0.21 → Mechanical (67%) ≈ $0.42
Same gross, opposite payout shape. A naive global ratio would over-pay German performance income by ~2× and under-pay German mechanical income by ~3×. The 5-layer model fixes that.
Honest scope. Some line items below are landing in a follow-up; others are permanently out of scope because they belong to your manual practice, not a recurring streaming projection:
The per-country rate table and the publishing-share splits are triangulated from these public sources. No keys or credentials appear here.