Research and analysis

Executive Summary: Music streaming metadata report

Published 14 April 2026

Interview findings by sector

The report presents findings by sector, based on interviews and written responses.

Creators

The main findings from creators were that:

• multiple writers are often involved in creating a musical work - writer shares and metadata are not always agreed at the point of creation, causing delays in registration and allocation of identifiers
• unique identifiers Interested Party Information number (IPI), International Performer Number (IPN), International Standard Name Identifier (ISNI) are not used universally or are hard to access - there is support for wider adoption of ISNI numbers for session musicians
• errors and omissions are common, including incomplete registrations and mismatched data
• there is a lack of understanding and education about metadata and music rights - the system is seen as complex and reliant on jargon

Music publishers

The overall findings from music publishers were that:

• publishers often lack access to authoritative International Standard Recording Code (ISRC) and International Standard Musical Work Code (ISWC) links at the time of release - this means accurate writer credits are not always available when recordings are provided to Digital Service Providers (DSPs)
• access to IPI information is limited, with technical barriers for some users
• there is no process for publishers to check other publishers’ registrations for split copyright works
• publishers are not in the recording metadata supply chain, so they rely on non-authoritative sources for this data
• competing data silos across the industry lead to discrepancies, duplicate effort and extra work to resolve them

Record labels

The main findings from record labels were that:

• labels cannot provide authoritative works information beyond what artists supply - they often do not have access to writer identifiers and rely on names, which can cause errors
• the timeline for releasing new recordings is usually faster than the timeline for agreeing metadata about the underlying work
• labels are not the ‘source of truth’ for works information and are concerned about pressure to provide more data to DSPs
• there is no central registry for ISRCs, leading to duplication and confusion
• better integration of tools and databases is needed to improve performer information

Collective Management Organisations (Authors’ rights and neighbouring rights)

The overall findings from Collective Management Organisations (CMOs) were that:

• CMOs often receive information about recordings post release from DSP rather than rights holders - this delays CMOs providing complete works to recording links information and in turn delays payment
• split copyrights and poor historic data create further challenges
• there are problems matching ISRCs and ISWCs, with many works lacking one or both identifiers
• contributor data quality varies, and there is a lack of comprehensive performer information, especially for session musicians
• the lack of an official central ISRC registry/database contributes to problems in maintaining data quality

Distributors and aggregators

The core findings from distributors and aggregators were that:

• metadata is supplied by labels and artists, but writer information and ISWC are not mandatory
• duplicate ISRCs and errors are common, especially in large catalogues
• aggregators allow anyone to upload music, but data checks are minimal - this affects data integrity
• there is little incentive for DIY artists to provide complete metadata

DSPs

The main findings from DSPs were that:

• DSPs do not have access to ISWC and IPI databases, which would help improve data accuracy
• the quality of works information in files from labels and distributors is often poor
• there is reluctance among CMOs and publishers to share songwriter information with DSPs
• ISRC accuracy affects the publishing licensors’ matching process and user experience

Solutions firms and experts

The overall findings from solutions firms and experts were:

• many of the issues raised are repeated across sectors - data gaps and disputes cause delays, and DIY platforms often deliver poor quality data to DSPs

Main takeaways

The main takeaways across all sectors were that:

• the lack of authoritative (ISWC/ISRC) links is a major problem throughout the supply chain
• many creators and their representatives struggle with metadata requirements
• aggregator platform rules allow incomplete and inaccurate metadata to be supplied to DSPs
• ISRC and ISWC allocation processes are different, with ISRCs allocated locally and quickly but ISWCs centrally and more slowly
• there is a lack of commonly used and accessible performer identifiers
• the most authoritative sources of metadata are authors’ rights CMOs for musical works and record labels for sound recordings, but both have limitations
• the industry faces challenges with standards, including slow adoption, inconsistent implementation, and gaps in business rules

Volumetrics findings

The volumetrics are designed to provide a numerical basis to support the development of the Key Performance Indicators (KPIs) - the sectoral responses show that we secured quantitative data from the following:

• five music publishers
• six record labels
• seven CMOs (of which four handled authors’ rights and three dealt with neighbouring rights)
• five distributors and five DSPs

The volumetrics support some of the interview findings, offering figures to support where metadata is incomplete or inconsistent across the supply chain.

Publishers

The core volumetrics findings from publishers were that:

• works had between 1 and 4 recordings, with one case showing a ten‑fold gap between active and charting works
• the average number of writers ranged from 1.6 to around 4, with more writers on charting works
• full writer IPI data was captured for 75% to 100% of works where rights are held by one publisher but dropped to 30% to 40% where publishers held partial rights
• ISRC information was often missing, one publisher included it in only 10% of registrations

Labels

The main volumetrics findings from labels were that:

• streaming splits between frontline (FL) and catalogue (Cat) ranged from 53% (FL) vs 47% (Cat) to 20% (FL) vs 80% (Cat) across different firms
• most labels had no writer IPIs or identifiers in their systems - ISWC data was also rarely captured

CMOs (Authors’ rights)

The core volumetrics findings from authors’ rights CMOs were that:

• the average number of writers per work was 1.71
• one CMO recorded 87% of works with an ISWC
• most works arrived with complete controlled IPI numbers
• 66% of registrations came from publishers and 33% from writers

CMOs (Neighbouring rights)

The core volumetrics findings from neighbouring rights CMOs were that:

• ISRC revisions occurred on 17% of claims each year
• between 4% and 11% of recordings had more than one ISRC, including a dataset of around 5 million recordings
• complete performer identifiers (IPN/ISNI) ranged from 7% to 73%, with higher rates for active and high‑value recordings

Distributors

The core volumetrics findings from distributors were that:

• only 5% of recordings needed contributor updates after delivery to DSPs
• songwriter credits ranged from 33% to 100% of catalogue, but almost no distributor supplied writer IPIs or ISWCs

DSPs

The core volumetrics findings from DSPs were that:

• DSPs processed large quantities of recordings but received very limited metadata
• one DSP reported 47.8 million annual recordings, most from independent labels and aggregators
• less than 0.1% of recordings arrived mapped to an ISWC
• ISRC–ISWC matching ranged from 0% to 85%, depending on source
• 82% of recordings had no streaming activity - only 0.0016% exceeded 100 million plays
• almost all recordings arrived with an ISRC, although one DSP reported 85.4%
• writer names were included on 40% to almost all recordings, but creator IDs were extremely low (in one case 0.22%)
• for recordings with usage, 20% lacked writer information
• reported averages showed 2 writers per recording
• duplicate ISRCs ranged from 4.5% to 23% across DSPs

Report evidence gaps

While all efforts were made to be as comprehensive as possible in collecting data from across the industry, there are some gaps. The gaps are:

• creator representation - it was challenging to engage a representative number of music creators, especially younger self-releasing artists - this was partly addressed by working with creator representative organisations, but their views may not fully reflect the wider DIY community
• aggregator representation - a major digital aggregator did not participate, so the findings may not represent the full picture for self-releasing artists
• publisher and DSP gaps - the research did not include all relevant music publishing administration services or all major DSPs - some findings are based on competitor views and available data

Final recommendations

Alongside the KPI project, other projects could be implemented. Music creators, educators and industry staff often lack awareness of how metadata works, why it matters, and how it affects payments and licensing. DIY creators in particular rely heavily on aggregator platforms, where guidance on metadata is limited or easy to miss.

Improve education and training

The following points on improving education and training are recommended:

• educate the educators - many teaching staff are not confident in metadata issues - we recommend targeted training and accreditation for higher education course leaders
• teach music students - universities and colleges could introduce a simple certification on metadata, similar to training required before using studio equipment
• support DIY creators - aggregators such as TuneCore and DistroKid should provide clearer and more prominent metadata guidance, as most self-releasing artists use these platforms
• train industry staff - employers could adopt a mandatory certification for metadata, following the example of one major label that trained all staff, including executives

Run regular surveys

The following points on using surveys are recommended:

• creator surveys - short, well designed surveys would track progress and identify different needs across demographics and socioeconomic groups - collaboration with organisations such as World Intellectual Property Organization (WIPO) and UK music industry bodies could help reach more creators
• platform guidance surveys - routine reviews of the metadata guidance provided by aggregators and DSPs would help measure improvements and capture changes in mandatory and optional data fields

Support further research

The report notes the positive developments of further research through the Music Futures programme, a five-year Arts Human Rights funded project launched in 2025. The programme is developing models and tools to improve metadata interoperability, assess licensing options and inform policy through a regulatory sandbox approach.

Glossary of terms

CMOs are licensing bodies which grant rights on behalf of multiple rights holders in a single blanket licence for a single payment. They collect any royalties owed to the rights holders are licensing bodies which grant rights on behalf of multiple rights holders in a single blanket licence for a single payment and collect any royalties owed to the rights holders. 

DSP are providers of online music streaming services to consumers.

ISRC is a code which is used to identify sound recordings.

ISWC is a code which is used to identify a song or composition.

IPN is a code to identify a performer and is used in data exchanges between CMOs.

IPI is a code which is assigned to songwriters, composers and music publishers by Performing Rights Organisations. These are attached to ISWC numbers to include information about authorship to works with an ISWC.

ISNI is a code used to identify persons including persona stage names and organisations involved in creative activities such as record labels.