Matching Music Usage to Sound Recordings at Scale

In short: Noctil's MatchIQ engine matches music usage reports, such as playlists, broadcast logs and partner reports, to the correct sound recordings in an organization's repertoire. It works even when identifiers like ISRCs are missing or wrong. In benchmark runs of 100,000 usage lines, MatchIQ keeps false positives and false negatives below 1% on average. Compared with legacy, rule-based matching engines, it also cuts manual review of partial matches by 15–20%, and runs in parallel so millions of lines can be matched inside any distribution window.

Claim more of your repertoire’s royalties, within every distribution cycle. For CMOs, neighbouring rights societies and music licensing companies matching high-volume usage reports to their sound recording repertoire.

The Situation

Collective management organizations (CMOs) and music licensing companies receive huge volumes of usage data: hundreds of thousands to millions of lines per day, week or month. It arrives from their local territory and from international partner societies as playlists, broadcast logs, digital service reports and other usage reports.

Each line has to be matched against the organization’s own sound recording repertoire, so that the right rightsholders can claim and receive their royalties.

The Problem

Usage data is often self-reported or produced by third parties, so its quality is low:

• Minimal metadata. Many lines carry little more than a title and an artist name.
• Missing or wrong identifiers. ISRCs and other identifiers are often absent or incorrect, so simple identifier matching fails.
• Unmatched usage. Lines that can’t be matched mean royalties go unclaimed, leaving money on the table for the organization’s rightsholders.
• Manual review bottlenecks. Partial matches need human review. At high volume, that stretches teams beyond what they can finish within the distribution period.
• Distribution and trust at risk. Late or inaccurate matching delays payments and weakens members’ trust in the organization.

The Solution

MatchIQ is the Noctil Platform’s metadata matching engine, built for high-volume, low-quality usage data.

1. Contextual matching. MatchIQ infers data and relationships from context, such as title variants, artist and contributor names, duration, release details and surrounding usage. It links each usage line to the correct sound recording even when the identifier is missing or wrong.
2. Fewer partial matches. Higher-confidence matching resolves more lines automatically, so fewer go to manual review.
3. Configurable thresholds and levers. Organizations set confidence thresholds and matching rules to balance automation against review, according to their own risk tolerance.
4. Parallel processing at scale. MatchIQ runs in parallel and can be configured to process any volume within the required time frame, keeping matching inside the distribution window.
5. Lower cost per match. The engine is designed to run fast and efficiently, so matching millions of lines stays affordable.

The Result

• Under 1% false positives and false negatives on average, in benchmark runs of 100,000 usage lines
• 15–20% less manual review of partial matches, compared with legacy, rule-based matching engines, through tunable thresholds and levers
• Matching completed within the distribution period, at any volume, through parallel processing
• More royalties claimed for the organization’s rightsholders, with fewer lines left unmatched
• Stronger member trust, from on-time, accurate distributions