November 02, 2026 • • 3 min read

Deduplicating and Merging Musical Work Registrations Without an ISWC

SegmentMusic Rights Organization PlatformNoctil Platform EnginesMatchIQ ProcessEngine
In short: Noctil's MatchIQ engine finds every registration of the same musical work, even without an ISWC, and clusters them against each other and the organization's existing record. ProcessEngine then merges each cluster into one complete work under the organization's own rules, sending only uncertain cases to human review. In a 100,000-pair benchmark, work-to-work matching produced fewer than 0.01% false positives or false negatives. 98% of duplicate registrations were clustered and merged automatically, and manual review fell by up to 35%.

Turn many registrations of the same song into one complete, trusted work. For PROs, MROs and other music rights organizations that receive the same musical work from several publishers, sub-publishers and administrators.

The Situation

Music rights organizations represent songwriters, composers and music publishers, and receive musical works through the registration process. A single work often has several writers, and each writer can be represented by a different original publisher, sub-publisher, co-publisher or administrator.

Each of those parties registers the work separately, at different times and in different formats. A song with several writers can therefore reach the organization as many separate registrations, each with its own partial view of the writers, publishers and shares.

The Problem

Turning those registrations into one accurate work is a technical and operational challenge:

  • No reliable identifier at first. New works often arrive without an ISWC, or even a provisional ISWC or society tune code, so registrations can’t simply be joined on a shared key.
  • Many partial versions. Each registration covers only the writers and publishers that submitter represents. The full picture only emerges once all of them are combined.
  • Conflicting information. Titles, writer names, IPI numbers and shares can differ between registrations. The organization must decide which source to trust for each attribute and leave out unreliable data.
  • Duplicates against the existing repertoire. New registrations have to be checked against each other and against works already on file, or the repertoire fills with duplicates.
  • Costly errors. A wrongly merged or incomplete work misstates ownership and shares. That delays royalties, or pays them to the wrong people.

The Solution

The Noctil Platform combines MatchIQ for matching with ProcessEngine for workflow, to build one source of truth for every musical work.

  1. Ingest from every source. Registrations from publishers, sub-publishers, co-publishers and administrators are loaded in their native formats, including CWR.
  2. Match and cluster with MatchIQ. MatchIQ deduplicates incoming registrations against each other and against the organization’s current work records. It uses work title and alternate titles, writers and composers, IPI numbers, publishers, performing artists and linked recordings, and links every registration of the same work into one cluster, even without an ISWC.
  3. Rule-based merging. Each cluster is merged into a single work using the organization’s own rules for ownership, source authority, data quality and completeness. The most reliable value for each attribute is kept, and unreliable data is left out.
  4. Workflow orchestration with ProcessEngine. ProcessEngine runs the end-to-end merge workflow. It sends clusters to human review only when confidence is low or ownership conflicts, and records every decision for audit.
  5. Continuous enrichment. As new registrations and identifiers arrive, such as an ISWC assignment, they are matched to the existing work and enrich it rather than creating new duplicates.

The Result

  • Fewer than 0.01% false positives or false negatives in work-to-work deduplication, based on a 100,000-pair matching benchmark
  • 98% automatic clustering and merging, in a benchmark of 100 unique works each registered in 6 to 9 versions
  • Up to 35% less manual review, through fewer candidate matches and an automated enrichment workflow
  • Accurate resolution of versions and clusters at any repertoire size
  • Complete ownership and shares on every work, so royalties flow without delay

Frequently Asked Questions

How do rights organizations deduplicate musical work registrations?

Noctil's MatchIQ compares each incoming registration with the others and with existing works, using title, writers, IPI numbers, publishers, performers and recordings. It clusters every version of the same work, and ProcessEngine merges each cluster into one record.

Can works be matched without an ISWC?

Yes. MatchIQ matches on the combination of work attributes, so registrations without an ISWC or tune code are still clustered correctly. When an ISWC is assigned later, it is linked to the existing work.

How are conflicting registrations merged?

Merging follows the organization's own rules for ownership, source authority, data quality and completeness. Clusters with low confidence or conflicting ownership go to human review through ProcessEngine.

How accurate is work-to-work deduplication?

In a 100,000-pair benchmark, false positives and false negatives were below 0.01%. 98% of duplicate registrations were clustered and merged automatically.

How much manual review does it save?

Up to 35%, through fewer candidate matches and an automated enrichment workflow.

Who is this for?

PROs, MROs, music CMOs and publishing administrators that maintain a musical work repertoire built from many registration sources.

← Back to all use cases

Talk to us about your metadata and rights operation.

Whether it's metadata, matching, rights or royalties, we'd be happy to explore what's possible.

Book a demo