The Least Interesting Thing About a Catalog Acquisition Is the Catalog

The least interesting thing about a catalog acquisition is the catalog.

Say that in a deal room and somebody slides the streaming report across the table: monthly listeners, sync history, the one riff that keeps turning up in car ads. All real, and all the part of the asset that any two competent analysts will price within a few percent of each other. The gap between a deal that closes near the seller's number and one that dies in diligence nine months later is almost never the songs. It is whether anyone can prove, on paper, who owns them.

I score indie games and short films, which puts me at the far end of this pipe. When the rights under a piece of music are clean, I get a yes on a cue in two days. When they are not, I get a maybe that outlives picture lock, and I write around it. That view from the buying end is worth something to anyone on the acquiring end, because the property that makes a catalog licensable at speed is the same property that makes it worth owning.

What a catalog acquisition actually includes

It transfers a bundle of rights plus the administrative record that proves them: the copyrights in the compositions, frequently the copyrights in the recordings, the income streams attached to each, and the contracts underneath — songwriter and co-writer agreements, confirmed splits, producer and remixer deals, sample and interpolation licenses, session-player waivers, and the registrations sitting at performing rights organizations and collecting societies in every territory the money arrives from.

The shapes vary widely. Some deals take publishing only. Some take a royalty stream and no copyright at all. Some take the writer's share and leave the publisher's share where it sits; some take masters and leave neighboring-rights income with the featured performer. What stays constant is the thing actually being purchased: the ability to say yes to a use, collect for it, and warrant to the licensee that nobody else will surface later holding a claim to the same three bars.

One-stop is the entire ballgame

A few years ago I built a 78 BPM cue around a two-bar break — dusty, a hair behind the grid, the kind of thing that makes an edit breathe. The picture editor loved it. Clearance ran eleven weeks and ended in a no, because the master sat in one place, the composition in two others, and one of the three never answered an email. I replaced it in an afternoon with something I played myself. The client paid for the replacement, not the original. Nobody in that chain got paid for the better piece of music.

Multiply that by a supervision department's weekly volume and you have the reason one-stop control commands a premium that has nothing to do with taste. When one entity controls both the composition and the recording, a music supervisor sends one email and gets one signature. When it does not, the identical use needs two or five, each with its own quote, its own timeline, and its own veto.

Which is why the most instructive deals are often not the headline nine-figure ones. They are the quieter transactions in which a company that has administered an artist's songs for two decades also acquires the recorded rights — the shape of BMG's announced acquisition of Wolfmother frontman Andrew Stockdale's publishing and recorded rights. Read as a purchase of songs, it is a well-documented rock catalog with real chart and awards history. Read structurally, it converts a two-signature asset into a one-signature asset in a market where the second signature is what kills placements. The first reading gets you a press release. The second gets you the thesis.

Why a twenty-year relationship prices better than a spreadsheet

Trade coverage tends to treat the long-relationship framing in these announcements as sentiment — the affectionate quote, the demo CD somebody kept in a desk drawer. It reads as warmth because it is warmth, but it is also underwriting.

A company that has administered a writer for twenty years already holds the split confirmations, the society registrations, the correspondence trail from the year a co-writer's percentage changed. It knows which works were registered under two different ISWCs by mistake in 2011 and which territory has been paying into a black box ever since. That is diligence already performed and paid for. A buyer walking in cold has to reconstruct all of it from statements, and prices the uncertainty into a holdback, a longer indemnity tail, or a lower multiple.

The relationship also survives the closing, which matters more than the diligence does. Catalogs need confirmatory assignments signed years after the fact. They need a co-writer's estate to acknowledge a split. Under US law, grants can become subject to statutory termination decades downstream, which turns a cooperative writer into an asset and an estranged one into a standing risk — the specifics belong to counsel, but the commercial point does not. A seller who trusts the buyer answers the phone. A seller who feels outmaneuvered lets the letter sit.

The clause everyone reads twice now

The schedule of prior grants used to be a formality. It is now the first thing a careful buyer turns to, because the question underneath it has changed: has any part of this catalog already been licensed for machine-learning training, and can the seller warrant that it has not?

That question has a hard consequence. Whatever your view of AI training licensing — enthusiastic, litigious, wait-and-see — participating in it at all requires warrantable title. You cannot offer a dataset counterparty a body of work whose splits are contested, whose sideman waivers are missing, or whose masters carry a territorial carve-out nobody documented. Catalogs with clean paper get to choose whether to participate. Catalogs with murky paper have the choice made for them, and are worth less for it regardless of how the legal questions eventually resolve.

The adjacent asset is the artist themselves: name, voice, and likeness rights are governed separately from copyright, unevenly across jurisdictions, and are increasingly the subject of specific legislation. Whether they travel with a rights deal, and on what terms, is a drafting question that deserves its own negotiation rather than a line in the recitals.

We spend most of our time at City of Punk on the downstream end of this — the producers rendering a 90-second bed at 48kHz who need to know what license they actually hold. The two ends of the pipe have converged on the same anxiety. Everyone now wants provenance they can point at.

A diligence pass you can run in a week

None of this requires a forensic audit to start. It requires pulling six things and reading them honestly.

Three red flags worth stopping for: royalty statements that do not reconcile to society data in the two largest territories, any work where the seller cannot name the other side of a co-write, and a seller unwilling to sign a post-closing cooperation covenant. The last one is not a paperwork problem. It is a forecast.

What this piece did not answer

It did not tell you what to pay. Multiples move with rates, streaming growth assumptions, and how much dry powder is chasing the same twelve catalogs in a given quarter, and any number set down here would be wrong by the time you read it. It did not tell you how to value AI training rights, because that market has no settled comparables and the law underneath it is unresolved in most jurisdictions. And it did not tell you whether the wave of catalogs now entering statutory termination windows will reprice the whole asset class — that is the genuinely open question.

Where to look next: the assignment documents themselves rather than the summary schedule; society registrations pulled territory by territory rather than through the seller's aggregate statements; and the representations and warranties in the last three deals your counterparty announced, which tell you more about their real risk appetite than any press release will. Then read the acquisitive publishers' annual filings, where the strategy is stated in language nobody bothers to make charming.

The songs are what you fall for. The paperwork is what you buy.