How to Read a Music Royalty Statement — Every Line, Every Source
A royalty statement is not an invoice; it is a transaction log. Once you know what one line
contains — and why the same song's money arrives through four separate pipes that never
describe it the same way — every statement becomes readable, whoever sent it.
What one line actually contains
Strip away the branding and every royalty statement is a table, and every row of that table
describes one parcel of money: this recording (or this song), used this way, in this place,
during this period, earned this amount. The columns vary wildly by source, but they are
naming the same handful of facts:
| Fact | What to look for |
| Period | Two dates matter and they are rarely the same: when the music was actually used (a sale month, a usage period, a performance quarter) and when the source reported it to you. Distributors typically post a store's report two to three months after the sale month; a performing-rights organisation can pay six to nine months after the broadcast. |
| Store / DSP | Where the money was earned — Spotify, Apple Music, "YouTube (Ads)", SiriusXM. On society statements this becomes a broadcast medium or licensee instead. |
| Territory | The country of the sale or performance. Usually a two-letter code, though some sources use three letters, and society statements may spell the country out. |
| Units | Streams, downloads, spins or plays. They can be fractional (long-track adjustments), zero (some reporting rows), or negative (returns and corrections — see below). |
| Gross, fees, net | The column that says "earnings" means different things on different statements. Some sources report only your net share, with their commission already deducted. Others show the store's gross payout, the fee taken, and the net remitted as separate columns. Before you sum anything, establish which of the three you are looking at. |
| Identifiers | An ISRC identifies a recording; an ISWC or a society work ID identifies the underlying song. Which one a statement carries tells you which right is being paid — and which statements can be joined to each other. |
The same song, four different pipes
One recording of one song generates money along several independent pipes, and each pipe sends
its own statement, on its own calendar, in its own vocabulary:
- Your distributor (DistroKid, TuneCore, CD Baby, Symphonic…) pays recording
revenue — the master's share of streams and downloads. Lines are recording-level, keyed by
ISRC, usually monthly, usually one row per store, country and month.
- Your PRO (ASCAP, BMI and their counterparts) pays performance royalties for
the composition — radio, television, venues, and the performance slice of streaming. Statements
are work-level, quarterly, and speak in credits, shares and classification codes rather than
streams.
- The MLC pays US mechanical royalties for the composition on downloads and
interactive streams. Monthly tab-separated files, at a very fine grain: one row per recording,
per service, per territory, per usage period, with your payable share of the work applied.
- SoundExchange pays digital-radio neighbouring rights for the recording —
satellite radio and webcasting under the US statutory licence, split 50% to the rights owner,
45% to the featured artist and 5% to session performers via the funds.
This is why the statements never look alike. They are not four views of one ledger; they are
four different rights, at different grains, administered by different organisations. A
distributor line carries an ISRC and a store name; a PRO line carries a work ID and a share
percentage; neither is wrong, and neither can be summed with the other without normalising
them into one schema first.
The traps
Four things routinely mislead people who read statements by eye — or break spreadsheets built
on top of them:
- Negative rows are real data. Returns, corrections and reversals appear as negative
units or negative earnings; Spotify Discovery Mode commissions arrive as negative amounts on
otherwise ordinary rows; some distributors mark voided transactions with negative counts.
Deleting them because they "look wrong" inflates your income — they are part of the
statement's own total.
- Foreign-currency columns sit beside USD totals. A statement can show the store's
payout in its local currency, the exchange rate, and the converted amount side by side. Sum
the wrong column and you have double-converted — or added euros to dollars. Find the column
the source itself treats as the payable total, and sum only that.
- Work-level and recording-level rows do not mix. Publishing sources may deliver both
a per-transaction detail file and a per-song summary of the same money. Combine the two grains
in one total and you have counted everything twice.
- YouTube is many revenue types wearing one logo. Ad-supported plays, Premium
subscription shares, Content ID claims on other people's uploads, Shorts and Art Tracks can
each arrive as separate rows or separate store labels — sometimes with extra columns (video
ID, commission rate, policy type) that are blank for every other store. "How much did YouTube
pay me?" is only answerable after grouping all of them.
Sanity checks worth doing
- Do the rows sum to the printed total? If a statement carries its own totals, the
lines you extracted should reproduce them to the cent. If they do not, something was lost,
double-counted or mis-parsed. Most machine exports carry no printed totals at all — that is
normal, and it is why an independent tie-out matters when one is available.
- Are per-stream rates plausible? Divide net by units per store. Premium interactive
streams mostly land in the low tenths of a cent; ad-supported streams and YouTube views run
far lower. A per-unit rate that is off by an order of magnitude usually means a mis-mapped
column — gross read as net, or money read as units.
- Are any periods missing? List the statement periods you hold for each source and
look for gaps. A missing quarter is either money you have not collected or a file you have
not downloaded — and if you are ever valuing
a catalog, a buyer's analysts will find the gap even if you did not.
Or let the machine do the reading
The free converter does everything on this page mechanically: it detects each statement's
format, normalizes every line into one documented
schema, checks that extracted rows sum to each statement's own printed totals to the cent,
and hands you a six-sheet Excel earnings tape. Parsing runs entirely in your browser — your
statements are never uploaded.
Try the free converter
Supported formats
Column-by-column dictionaries for every format the converter reads natively:
Read next: the normalized earnings-tape format ·
preparing an earnings tape for catalog diligence ·
back to the converter.