How music industry silos are leading to unpaid royalties.

The last year in the music industry has seen an increasing amount of activity regarding the operational side of rights management. More recently we have seen:

SACEM, PRS and ASCAP announce a joint blockchain project Figure: SACEM, PRS and ASCAP announce a joint blockchain project

There seems to be a pattern of major collection societies around the world merging efforts to improve data accuracy. While some folks in the industry have criticized these initiatives by indicating it as a “power move”, we feel differently. The main questions are:

Before we go further into it, I want to add some context to this via a quick Music INDUSTRY 101.

A Music Industry refresher

The music industry is often mistakenly referred to as the recording industry or vice-versa, while in actuality the music industry is multifaceted with sub-industries like recording, publishing, synch, licensing, etc. The roots for this evolutionary consequence can be traced to centuries ago, back to what I call the “Great Divide of the Copyrights”, which founded the distinction between the rights of the composition and the rights of the recording.

This distinction has remained within the music industry, recognized by every imaginable service provider (the glamorous record labels to the bookkeeping publishers). Even the collection societies or PROs (Performing Rights Organizations) are established to serve these specific verticals. This siloed design of the music industry branches out its influence even at the operational workforce or the ‘leaf ends’, in many instances forcing an entire industry to operate disparately.

Song Copyright Symbol, Recording Copyright Symbol Figure L-R: Song Copyright Symbol, Recording Copyright Symbol

This all leads us back to why there is a data problem in the modern music industry.

Speaking from a software development and data design perspective, for a system to be considered “well-designed”, relationships need to be established and standards need to be maintained. Relationships across data is important. Heck, there is an entirely new technology industry that is based on data relationships; the IOT or Internet of Things.

The problem lies in the Music Industry’s data model

Currently there is absolutely no recognized standard or relationship between the data identifiers ISRC (International Standard Recording Code used to identify a recording), and ISWC (International Standard Musical Work Code used to identify a composition).

When a song gets remixed, covered, radio edited, etc, there is a new ISRC assigned to each new version with the assumption that an ISWC for the original composer will easily be associated with the new ISRC. While ISWC is supposed to be immutable, this is not a forced standard and we see duplicate ISWCs for the same song on some occasions.

All of these new data points can sprout up without any hard relationship to the original. So how will the collection societies effectively identify which song is associated with which record? Generally, the obvious solution is to match over names across the databases.

There’s always room for mistakes, but the state of the system now is that a simple case of a spelling error in the artist’s name can result in misidentifying the correct rights holders. And as this data moves between service providers, there’s a compounding effect that results in loss of artist ownership and royalty, as well as a cost impact absorbed by service providers to lookup, identify conflicts, and clean up the data sets (although usually the conflicts aren’t identified).

Because composition and recording rights have remained separated, there is a fundamental missing link between these two incredibly valuable rights information sets. And as we’ve seen with the digitisation of consumption and streaming, where high plays don’t necessarily translate to high royalties, the recording and publishing industries are struggling to get the most value out of their rights data. This missing relationship has brought the music industry to a highly fragile state—something that that major organizations like the PROs are starting to realize and take action on.

What we need is a global identifier — CompoRecord ID (CRID)

One new identifier is all we need to solve most of the data problems we see in matching the rightholders’ data. This new identifier will enforce the missing relationship in data. At Paperchain we recognized this problem at the very beginning of our efforts to help solve this. In our proprietary data validation platform Pura, we came up with an identifier called a “CompoRecord ID” or CRID, which encapsulates both the composition right holders’ information and the recording ownership. This allows us to seamlessly traverse through our data and fetch the right owners’ information.

Following is an image from our PURA platform’s prototype, showing both the composition and recording rights’ ownership information side-by-side and linked by a CRID.

paperchain pura links composition and recording IDs through the CRID

The problem Is grave

Every year, hundreds of millions of dollars go unpaid or misallocated and only to go into “revenue black boxes” (escrow accounts held until the correct rights owners are identified) due to these compounding errors.

That’s why we’ve seen the movements that we have by some of the major PROs. Digital sales and streaming in particular is generating more data than most organizations can deal with, and it’s revealing a problem that has always existed but now can’t be ignored.

At some level there needs to be a consolidation of composition and recording identifiers from the various verticals of the music industry or at least a way to better connect these databases and build working relationships between the data sets. This is certainly easier said than done, but not impossible. Most importantly however, this unification is absolutely necessary and it needs to be done at scale before artists lose out for good.