All Posts

In the past month alone we’ve seen three announcements that are as fascinating individually as they are illustrative collectively. Primary Wave in talks to acquire Kobalt. BMG reportedly exploring a deal for Concord. The now-concluded acquisition of Downtown by Universal Music Group Each headline by itself is significant, but taken together they suggest something more structural is currently at play. A re-majorisation of the music business.

There was a moment, not that long ago, when music rights acquisitions felt almost mechanical- streaming growth was steady, capital was abundant. The underwriting conversation revolved around a small set of variables: historical cashflow, growth assumptions, discount rate, and multiple. If you believed in the long-term durability and expansion of streaming, the rest followed.

On Spotify’s latest earnings call, there was a few moments, including the quote “Technology is seldom disruptive on its own. Significant disruption happens when new technologies enable new asymmetric business models.” The real question isn’t whether AI is powerful, but whether it enables new business models entirely.

Music rights are nuanced, and always have been. Creative collaboration, shared ownership, territorial differences, long tails of value, and cultural context are all part of what makes music operationally distinct from many other industries. There is, however, an important distinction to be made between nuance and friction.

“We are a defacto data company. We don’t work in the music industry-  we work in the data industry.” This quote from Caroline Champarnaud of SACEM speaks deeply to the operational reality of a modern collecting society- though it resonates far beyond that context, into the very fabric of music publishing and copyright administration in the modern age. The more I’ve thought about it since, the more it feels like an accurate description of the publishing ecosystem.

Last week I was at Music Ally Connect in London- and as you can well imagine AI took it’s usual place as the topic du jour. Despite being well trodden ground there was some fascinating elements- both trends from the day at large and specific comments that gave insight to how the Music industry is currently both thinking about and preparing for (/guarding against) AI. Mostly these broadly fell into 3 camps- intense negativity, bullish positivity and cautious ambivalence.

Top songs are what attract capital. Music, and entertainment at large, is a hits driven business. This may be true, as most of the value magnetises to the ‘head’ of catalogs, but it doesn’t mean that everything else has no value. In fact- the truth is far from it. Large catalogs can offered history, predictability, and a level of institutional comfort that made underwriting easier- let alone the base level of recognizability and ‘icon’ status that marketing departments dream

Competitive advantage in music is often framed around ownership, but data provides an equally powerful advantage. Serious operators across the music ecosystem are all looking at broadly the same raw data inputs (keeping in mind last week’s look at how data is no longer a scarce resource)- with cloud infrastructure and AI flattening the technical barriers to entry. Data is no longer the differentiator it once was.

For years, conversations around music data have revolved around access- to global copyright databases, better royalty data, deeper DSP reporting (+ the terms of DSP licenses). Data in this framing is treated as a scarce resource- something hoarded, guarded, and used to justify competitive advantage. The great news is the tide has shifted in recent years, meaning this framing no longer holds.

Every industry has its myths- narratives that get repeated so often they start to feel like facts. Music may be one of the most culturally powerful forces in society, but it’s also one of the most misunderstood commercial ecosystems. Misconceptions are often perpetuated by those without years of internal industry experience- not only colouring how the business is perceived (particularly operationally), but also by implication how we think about innovation and investment.

Music is one of the most culturally powerful forces on earth. It anchors memories, shapes identities, moves crowds, ignites movements, defines decades, and travels further and faster than almost any other creative expression. It also underpins nearly every other entertainment and social experience we care about. And yet… commercially and systemically, it is profoundly undervalued.

For years music publishers have learned to live within a strange paradigm. Chatter around the industry is endlessly about innovation, data, efficiency and the collaboration required to realise gains for all. Yet for all the discussion, the data sets relied upon for collecting and attributing royalties continue to operate with significant limitations- none of which need to be spelled out here. Publishers, despite best efforts, often only discover issues after value has already

There’s a recurring dichotomy in how music publishing ‘value’ is discussed. The media conversation clusters around the headline acquisition of iconic catalogs and blockbuster songwriters- the “big deals” naturally command attention. They are tangible, easy to explain, and fit the narrative that music IP behaves like a financial asset. However there is also a real engine of value outside those deals that make the music headlines. There is still material revenue to be found ac

AI is leverage like humanity has never seen before- but it is amplifying both signal and noise. AI isn’t creating wisdom- it doesn’t make us smarter, or more insightful, or even necessarily more effective. It is simply scaling what already exists.

Every vertical in the entertainment industries has always rested on two foundational pillars: content and distribution. One produces the art- the stories, songs, and IP that audiences love, the other makes sure those works can actually reach them. The symbiosis and tension between the two is as old as the media business itself, but in practice it’s far from a zero-sum game. Content without distribution is invisible. Distribution without content is empty.

The music industry has always existed as a vehicle between art and commerce. In the last decade, a third dimension has pervasively entered the mix, and has been delicately impacting the status quo- institutional finance. Catalogs that were once considered illiquid creative assets are now treated as financial instruments. Music rights are being packaged, leveraged, and traded with the same language and logic once reserved for bonds, real estate and pension funds.

For decades, sub-publishers formed the backbone of the global music publishing network- local partners handling collections, licensing, and promotion in markets where a publisher lacked direct presence. They were the trusted representatives on the ground, fluent in the local language, culture, and society politics and nuance. But in today’s hyper-connected, digitally-driven industry, consumption has become overwhelmingly global.

The music industry runs on identifiers. Without them, the complex ecosystem of recordings, compositions, rights, and royalties would quickly collapse into chaos (... more so than it already is). The two most critical identifiers are ISRC and ISWC- the codes that uniquely define distinct audio recordings and song compositions respectively. Their existence looks simple on the surface, but they represent both different philosophies and underpin key workflows and challenges in at

From continuous large scale investments and strategic manoeuvres to tales of narrowing margins and constant restructuring, the music publishing industry seems caught in a tension between optimism and anxiety. On the one hand, Goldman Sachs’ Music in the Air 2025 and Bolero’s subsequent analysis paint a bullish medium and long-term picture. Meanwhile, publishing revenues alone- according to Goldman Sachs- are projected to rise more modestly.

In 2017 Google’s AlphaZero defeated the longstanding championship winning Stockfish 8 program at the game of Chess. AlphaZero only learned the game through machine learning, by playing itself. To conquer the world’s most powerful chess engine it only took 4 hours of training. 4 hours vs. the entire accumulation of human knowledge about the game. In Harari’s words- “Because AlphaZero learned nothing from a human, many of its winning moves and strategies seemed unconventional