There is no question that the rise of streaming services and other digital music companies have revitalized a flagging music industry, providing a new and growing revenue stream for musicians around the globe. But what happens when a service like Spotify, Apple Music or Amazon can't identify the copyright owner of a song that's played on its service -- even if that song has been legally licensed?
For several years, digital music services have been allowed to host and stream copyrighted music even if the metadata attached to a particular song is missing or incomplete, through filings with the U.S. Copyright Office. This has been a major thorn in the transition to a digital music economy; if metadata is not there, services and companies, even with the best intentions, cannot determine who to pay. In those situations, the services will put the accrued royalties into an escrow account and file a Notice Of Intent with the Copyright Office, usually in bulky, individually-downloadable Excel documents that include thousands of song titles apiece.
The Copyright Office's web site shows 1,080 entries on its site dating back to April 2016, many of which include multiple individual spreadsheet filings per entry; between Dec. 1 and Dec. 5, for example, Spotify has filed 35 documents (seven per day); Google 10 documents; Amazon four documents; and iHeart two. All told, according to SoundExchange's recently-formed SXWorks organization, there are now 57 million Notices of Intent filed with the Copyright Office, all of which represent royalties that are being held in escrow until a publisher can search through the data, identify the works it owns, and apply to collect that money.