What’s the point of having good metadata and rights information for licensing? It should be to actually know who owns a song and who owns recordings of the song so you can get permission and pay for usage or block it.
However, as we have seen with the disasters of Google Books and YouTube ContentID, there are people out there who use metadata to shield themselves from liability regardless of how inaccurate it is. This is something to keep in mind when dealing with companies that want to collect your rights data and aggregate it into a database that they can then sell. (An example might be a company like Loudr.) As we saw with Rightsflow, the Copyright Office unknown copyright owner list includes a license request from Rightsflow for Gangnam Style–a rather bizarre result.
There’s an easy fix for this problem, though. Make sure you have a written data confidentiality agreement that allows you to terminate any rights in the case of an assignment, sale or other transfer of the licensee’s company or assets. Unless you like the idea of supporting spunky startup that is secretly hell bent on getting acquired by a Google or the highest bidder before they flip the keys to the company to the first bum on the street (aka an assignment for the benefit of creditors, orderly windup or the Founder’s Disappearing Act).