When Bad Data Hits the Wrong Note: The Data Understanding & Quality Crisis Facing the Music Industry

One of the core issues that the Music Industry is struggling to solve is a fundamentally simple one: music rights are wholly dependent on being able to correctly identify what was used, who created it and who owns the relevant rights. However, with the volume of data, its sources and formats continually growing and evolving, much of the information available within the industry is inaccurate, incomplete, fragmented, or siloed. This results in a royalty system that is no longer fit-for-purpose in today’s technology driven age. Music is becoming ever increasingly global and that means digital. Therefore, improving data understanding and overall quality is not simply an administrative overhead, it is a foundational requirement for ensuring that artists, songwriters and rights holders receive the correct payment in an efficient and timely fashion.

As with all business sectors the modern music industry generates more data than ever before. Every stream, download, broadcast, performance, sale and licence creates valuable information that in theory allow artists, songwriters, publishers and rights holders to be accurately identified and correctly recompensed. Yet the industry is still plagued by incomplete, inconsistent and inaccurate data, housed within a fragmented and siloed ecosystem involving: artists; labels; publishers; collecting societies; streaming platforms; distributors; broadcasters; and technology companies. A lack of data understanding and poor-quality is endemic and is arguably the prime factor in royalties being delayed, miscalculated, misdirected or lost within the system that is meant to manage them.

At the heart of the problem is the correct identification of the music and the people who own the rights to it. A single recording can involve multiple parties, including: performers; record labels; producers; composers; lyricists; and publishers. The same song may also have several versions, such as: an album recording; radio edit; remix; live performance; or acoustic version. It is entirely possible for each of these versions to have different ownership and licensing agreements. If the data associated to a particular piece of music is poorly understood, incomplete or inconsistent, it becomes very difficult to accurately determine who should receive payment.

A major source of this difficulty is an industry wide lack of accurate and consistent metadata which can include essential information for royalty payments such as: the song title; artist name; songwriter; publisher; label; recording identifier; and ownership percentages. When this information is missing, duplicated, has terminological variants, or is captured within multiple – often incompatible – applications the ability to match a usage to the correct rights holder becomes considerably more complex. For example, an artist might be credited under a stage name on one platform and a legal name elsewhere, or a song title may appear with different spellings. What could appear as a minor administrative discrepancy regularly prevents systems from accurately matching usage to the appropriate rights record.

The fragmented, and often siloed structure of the global music industry – a single database containing a definitive record of every musical work, recording, creator and rights owner does not exist – significantly amplifies these discrepancies. Data is distributed across thousands of organisations and individuals operating in different territories and, very often, using different platforms. Streaming services may hold usage data, record labels may control master-right information, publishers may manage composition rights, while collecting societies administer royalties in particular countries.

Music is now a global entity and international consumption complicates the situation even further. A song created in one country can be streamed in hundreds of territories, with royalties potentially passing through several organisations before reaching the right person. Different countries may have different licensing rules, collection societies, and reporting standards. Consequently, data that is sufficient for identifying a rights holder in one market may be inadequate in another. Every additional intermediary creates another opportunity for information to become incomplete, duplicated or incorrectly transformed, and used.

When a platform or collecting society cannot confidently identify the appropriate rights holder, money may be placed into an unmatched or unidentified pool rather than distributed immediately. Although such funds may eventually be claimed, the process can be lengthy, uncertain, and open to abuse. In many cases, rights holders may not even know that money is waiting to be claimed because they lack visibility into where their works have been used.

This disproportionately affects independent, and upcoming, artists as well as the smaller rights holders. Major organisations, that look to dominate the sector, may have dedicated rights-management teams capable of correcting metadata, registering works and pursuing missing royalties across multiple territories. However, the complexities of today’s data ecosystems make this an increasingly difficult and costly challenge. It is the smaller, independent creators – many of whom are the future of the industry – who often lack the resources or expertise to perform these tasks. As a result, the more fragmented and complex the royalty ecosystem becomes, the greater the risk that smaller participants, who are just beginning their music journey, are financially disadvantaged by poor data.

The complex issues surrounding the correct royalty payments issues are amplified further by the sheer volume of music that is now being released. The ability to digitally distribute media has dramatically lowered the barriers to entering the market, resulting in hundreds of thousands of new recordings and works being annually registered and globally consumed. The scale of this activity makes manual data correction at the speed and scale required virtually impossible. However, the success of automated royalty processing is predicated on the accuracy and validity of the underlying information. Poor quality data produces poor quality outcomes at significant scale.

The likelihood or errors, and poor insights, is further compounded as poorly understood data moves through the music supply chain. For example, if a distributor receives information from an artist and it is inaccurately recorded within a system, at some point it may find itself incorporated into reporting platforms used by another organisation. This creates a cascading error that contaminates multiple data repositories which not only negatively impacts royalty payments for writers and performers it also carries significant implications – both financial and reputational – for the infected organisations.

The complex issues created by a lack of data understanding and poor-quality generate a vicious cycle that the music industry must extract itself from. Inaccurate metadata makes royalties harder to allocate; unallocated royalties create disputes and administrative costs; disputes require organisations to investigate and reconcile records; fragmented systems make those investigations difficult; and it is the grass roots of the industry that suffer most.

Improving royalty outcomes so that payments flow efficiently to creators instead of becoming trapped in complex administrative processes requires a greater data understanding across the entire music industry. Data understanding initially stems from knowing where the data comes from, how it is compiled and structured, how it flows across the industry, and how it is used (and by whom). A consolidated ‘single pane of glass’ view can: connect fragmented information; enhance data quality; improve standardisation and interoperability; expose inconsistencies, improve collaboration, transparency and accountability; and enable everyone within the music industry to make better decisions quicker.

As the music industry becomes increasingly data-driven, improved data quality delivered through better data understanding is the foundation for long-term, sustainable progress, failing to address these issues now runs the very real risk of placing the industry on a path of continual decline.