The MLC’s Next Phase: Identification to Allocation

Since The MLC’s inception in 2018 through to the start of its start of distributions in 2021 and beyond it has been closely watched- by the market at large, by key stake holding publishers and by compatriot CMOs the world over. Mostly this has been positive- in particular their approach to data transparency and availability as required by the Music Modernization Act that birthed the society.

Photo by Raffaele Parente

Starting a large scale CMO from scratch is no mean feat- particularly from a data and operations standpoint. Despite its detractors, The MLC has generally done a fair job in getting value out the door to rights holders and simplifying the entire mechanical licensing landscape in the US- in my opinion successfully fulfilling its intended purpose.

However, from next year there is to be a material change in how monies are distributed, a change that will affect all rights holders, and a change that other CMOs are more than familiar with, yet The MLC has intentionally left until its operations and data coverage have matured. Pro rata distributions: https://pages.themlc.com/looking-one-year-ahead-market-share-distributions.

The early phase of the MLC was about building infrastructure- ingesting vast amounts of usage data from DSPs, constructing a centralised database of works and recordings, and giving the industry visibility into how those datasets align.

As we move into the next phase, it’s likely the conversation will begin shifting.

“Is it matched?” to “Is it correct?”

Under a pure usage-based system, the question is relatively simple: Is this recording correctly linked to this work, and are my shares registered?

Under a pro-rata system, the question becomes more nuanced, and more consequential: Is my data complete relative to the rest of the market?

When distributions are made proportionally, a rights holder’s share of the pie is no longer just a function of your own data, but a function of everyone else’s too. If your catalog is under-matched or misattributed you’re not just missing revenue in isolation, but losing share.

Across MLC datasets (and, more broadly, across CMO data globally), there are several recurring issues to address where value accumulates:

1. Duplicate work registrations. Multiple registrations for the same work, often caused by inconsistent metadata in combination with parallel submissions from different parties will quietly split or block income. Even when everything looks “correct” at a surface level, duplicates can dilute earnings in ways that are difficult to detect without systematic analysis.

2. Unmatched recordings This is the most visible category, and oft talked about. Conveniently with the MLC it’s also the one most directly addressed by tooling they provide (see below). If a recording isn’t linked to a work, the associated royalties have nowhere to flow- which is 100% critical for a society that only licenses digital mechanical usage. In a identification-based system, that’s a missed payment. In a pro-rata system it can equate to a reduction in relative share.

3. Incorrect matches Less obvious and harder to value (but equally important) are cases where recordings are already matched, but to the wrong works. Here, the data may look “complete”, but that does not mean it is necessarily “correct”. In aggregate these misallocations can distort distributions in ways that are difficult to unwind. This will also have the highest impact on pro-rata distributions. If someone else is already being paid your "identified" money, they will also receive an increased "unidentified" allocation.

4. Unclaimed shares. Even where works are correctly registered and recordings correctly matched, unclaimed shares can remain. Whether due to missing co-writer data, unresolved conflicts, or gaps in representation. These gaps don’t just delay income- they stop, or worse, redistribute it (we've regularly surfaced instances of %’s not just unclaimed, but claimed by other incorrect rights holders).

The Tools Exist- will others follow suit?

One of the most under-appreciated aspects of the MLC is that it hasn’t just surfaced data, but has also provided mechanisms to act on it. See our previous exploration of how both transparency and access need to work in tandem to enfranchise global data cleanup.

If we focus on unmatched recordings there are, broadly speaking, two complementary workflows available at The MLC- tools that I would strongly encourage other CMOs to follow suit in implementing:

1. Bulk Matching

For larger catalogs, the primary lever is the bulk submission process, designed for high-volume, lower-value matching, where the objective is to systematically improve coverage across 10s of thousands of recordings.

This process taps into an extended ecosystem of matching technologies and providers, including platforms like Blokur, Salt, and SX Works. The result is a supplemental matching network, where scale is achieved through automation and cross-referencing across multiple data sources. As we’ve also explored previously, incremental improvements compound at scale, and necessitate technology solutions to implement.

2. Targeted Resolution

Alongside this sits the matching tool inside the MLC Portal: where higher-value, lower-volume issues can be focussed on. These are cases where human oversight is required and where the economic impact materially justifies manual intervention. If bulk matching is about coverage, the portal is about prioritisation

This combination of data visibility and actionable tooling is relatively unique.

Historically, many collection societies have provided neither, or at best one without the other. Either data without the ability to act on it, or processes without transparency into where value actually sits or can be retrieved. The MLC, by contrast, has leaned into both and the hope is that others will look on and implement similar.

Looking One Year Ahead

The move to market-share distributions is, in many ways, a natural evolution. Every CMO (on paper) wants to get value out of its account and into that of rights holders, and beyond to creators themselves. Analogy distributions are a sad necessity in pursuit of this- when data simply isn’t available to accurately attribute value. Would we rather see value accrue and not be distributed? Should we (of course) work to minimise what must be paid by analogy rather than validated usage -> rights holder match?

Kudos must go to MLC for at least holding off pro-rata payments initially. However from the next year the incentives are going to change. NOW is the time to clean up your data, NOW is the time to ensure your catalog is correctly represented in their data set, so you and your creators only see benefit from this operational evolution.

Identification exposes everything upstream- every duplicate, missing link and inconsistency. However allocation exposes these issue not in isolation, but relative to the market.

The game will quickly move from “Is my data clean enough” to “Is my data cleaner than my competitors”? The time to act is now.