# Matching TV Broadcast Usage to Audiovisual Works at Scale > How Noctil's MatchIQ engine matches TV broadcast and audiovisual usage to the exact film, series, season, episode and version — even when ISAN/EIDR identifiers are missing — with under 1.8% error and 15–20% less manual review. - Segment: Audio Visual Rights Organization - Platform: Noctil Platform - Engine: MatchIQ - Author: Jacob Varghese - Published: 2026-11-01 - Source: https://www.noctil.com/use-cases/matching-tv-broadcast-usage-to-audiovisual-works **In short:** Noctil's MatchIQ engine matches TV broadcast and audiovisual usage reports to the correct audiovisual work, down to the exact series, season, episode and version. It works even when identifiers such as ISAN or EIDR are missing or wrong. In benchmark runs of 100,000 usage lines, MatchIQ keeps false positives and false negatives below 1.8% on average. Compared with legacy, rule-based matching engines, it also cuts manual review of partial matches by 15–20%, and runs in parallel to finish any volume inside the distribution window. --- _Match every broadcast to the right film, series, season and episode. For audiovisual (AV) rights societies, broadcasters' licensing teams and AV licensing companies matching TV broadcast logs to their film and series repertoire._ ## The Situation AV rights societies and licensing companies receive large volumes of usage data from TV broadcasters, cable and satellite channels, streaming services and partner societies abroad. It often runs to hundreds of thousands or millions of lines per distribution period. Each line has to be matched to the right audiovisual work in the organization's repertoire, so that producers, directors, screenwriters and other rightsholders are paid for the exact film or episode that was broadcast. ## The Problem Audiovisual works are harder to match than music. One title can refer to many different works, and broadcast logs rarely give enough detail to tell them apart: - **Series, season and episode hierarchy.** A log line may name only the series, such as a long-running drama, with an episode number but no season, or an episode title but no number. Payment depends on the exact episode, not the series. - **Same title, different films.** Remakes, reboots and unrelated films often share a title. Release year, director and production company are needed to tell them apart, and logs rarely include them. - **Localized and alternate titles.** Works are broadcast under translated, dubbed or regional titles that don't match the original title in the repertoire. - **Versions and edits.** Theatrical cuts, TV edits, director's cuts, dubbed and subtitled versions can each carry different rights or durations. - **Missing or inconsistent identifiers.** Industry identifiers such as ISAN (International Standard Audiovisual Number) and EIDR (Entertainment Identifier Registry), along with broadcasters' own internal IDs, are often missing, incomplete or point to the series rather than the episode. - **Noisy timing data.** Broadcast durations include ad breaks or partial airings, so runtime alone can't confirm a match. Unmatched lines leave royalties unclaimed. Partial matches pile up for manual review, which puts the distribution deadline at risk and erodes rightsholders' trust. ## The Solution MatchIQ, the Noctil Platform's matching engine, understands how audiovisual works are structured, not just their titles. 1. **Hierarchy-aware matching.** MatchIQ models series, seasons, episodes and versions as linked records. It resolves a log line to the exact episode using whatever clues are present: an episode number, episode title, air date, channel or running order. 2. **Contextual disambiguation for films.** Release year, director, cast, production company, country of origin and duration are weighed together, so remakes and same-title films are told apart. 3. **Title intelligence.** Original, translated, localized and alternate titles are matched across languages and spellings. 4. **Identifier reconciliation.** MatchIQ uses ISAN and EIDR wherever they exist, including their series, episode and version levels, as well as broadcaster house IDs. It infers the right work when identifiers are missing, incomplete or wrong. 5. **Configurable thresholds and levers.** Organizations tune confidence thresholds and matching rules to balance automation against review. 6. **Parallel processing at scale.** Matching runs in parallel and can be configured to finish any volume within the distribution period, at a low cost per match. ## The Result - **Under 1.8% 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 - **Episode-level accuracy**, so royalties follow the exact film, episode and version broadcast - **Matching completed within the distribution period**, at any volume, through parallel processing - **More royalties claimed** and stronger rightsholder trust, with fewer lines left unmatched ## Frequently Asked Questions **How do AV rights societies match broadcast logs to the correct episode?** Noctil's MatchIQ models series, seasons, episodes and versions as linked records. It resolves each log line using clues such as episode number, episode title, air date and channel, even when the log names only the series. **How does MatchIQ tell apart films with the same title?** It weighs release year, director, cast, production company, country of origin and duration together, so remakes and unrelated films with the same title are matched correctly. **Does MatchIQ use ISAN and EIDR?** Yes. It uses ISAN and EIDR identifiers, including their series, episode and version levels, and broadcaster house IDs where available. When identifiers are missing or wrong, it infers the correct work from context. **Can it match translated or localized titles?** Yes. Original, translated, dubbed and alternate titles are matched across languages and spellings. **How accurate is MatchIQ for audiovisual works?** In benchmark runs of 100,000 usage lines, false positives and false negatives average below 1.8%. Manual review of partial matches is 15–20% lower than with legacy, rule-based engines. **Who is this for?** Audiovisual rights societies and CMOs, producer and director collecting societies, broadcasters' rights and licensing teams, and AV licensing companies that distribute royalties for films and TV series.