When the Mechanical Licensing Collective (MLC) in the United States began operating, it uncovered more than $400 million in historically unmatched royalties. That money wasn't lost in the literal sense: it had been generated, platforms had paid it out, but no one could prove who actually owned each recording.

How does an industry end up with $400 million in unclaimed royalties? The answer isn't fraud or individual negligence. It's a royalty system built for a pre-streaming catalog that now has to process millions of new releases every month using the same architecture from twenty years ago.

Where does the leak actually start?

Money leaks out at four concrete points in the chain, and all four share the same root cause:

• Ingestion: incomplete metadata from the moment the file enters the system.
• Registration: gaps in territorial rights registration.
• Matching: failures linking a recording to its true rights holder.
• Allocation: unattributed revenue that ends up in so called black box funds, held for one to three years before redistribution, almost always split proportionally across the entire market instead of going to whoever actually generated the stream.

The Music Business Association estimates this costs the industry $100 million a year, just from the lack of standardized metadata. And the root cause of all four leak points is exactly the same: incomplete data at the moment the release enters the system.

Why isn't an ISRC enough to collect your royalties?

A master and a composition travel different paths through the distribution pipeline. The master is identified by an ISRC, the composition by an ISWC. Two codes, two databases, two parties collecting different things: the ISRC moves master royalties (streaming, sync), the ISWC moves composition royalties (mechanicals, performance).

When the link between the two is never established, or breaks somewhere in the pipeline, the system has no way of knowing that ISRC corresponds to that ISWC. The stream still gets counted. The composition royalty simply never reaches anyone and ends up padding the black box.

What does a platform require before accepting your release?

DSPs don't demand perfect metadata out of bureaucracy: they demand it because without it, there's no way to pay anyone correctly. At minimum, every release needs:

Correct formatting: artist, album title, track number, and ISRC for each track.
Complete credits: songwriters, producers, featured artists.
Consistent genre and mood tagging.
Accurate release dates and territorial restrictions.

Without this, the release gets delayed or rejected. And if it slips through incomplete, the bill comes later: duplicate tracks, exclusion from editorial playlists, lost sync opportunities, fewer algorithmic recommendations.

How does a label avoid showing up in the next "unclaimed" report?

Prevention isn't chasing the money after it's lost. It's locking down the catalog before it ships:

Collect metadata from day one of production, not the last week before release.
Follow established standards (RIAA, DDEX guidelines) instead of label-specific criteria.
Centralize metadata management in a single system, not scattered spreadsheets.
Verify every ISRC and every ISWC before submission, not after the DSP rejects it.
Automate validation instead of reviewing release by release manually.

This is what a well-built White Label infrastructure does: it links ISRC and ISWC from ingestion, validates credits automatically, and keeps your catalog off the next "unclaimed royalties" list.

‍