Linear TV Metadata broadcast schedule metadata for every channel you carry

Linear television has two distinct layers, and Linear TV Metadata models both — which is what makes it accurate where a flat feed falls apart.

• The event — a point in time and space: a specific channel, at a specific moment, for a specific duration. "TV7, 20:20, 95 minutes." The event is the slot.
• The content — what actually fills that slot: the film, episode, match or show, with its title, people, description, images and classification.

The two are linked but separate. The same piece of content — say, a given film — can air as many different events: on different channels, in different countries, on different dates. Because content is described once and reused across every event it appears in, the guide stays consistent: the same film carries the same synopsis, cast and artwork wherever and whenever it is scheduled, while each event keeps its own exact timing, channel and rights.

Underneath every record is an identifier that does not change. This is the connective tissue of the whole product, and it is what separates Linear TV Metadata from a feed you have to constantly re-match.

• Content ID — every film, series, season and episode has one permanent ID. When the same title appears on another channel, another platform or in another market, it is the same ID — not a new, unconnected record to reconcile.
• Person ID — every actor, director and presenter has one ID, so "the same person" is unambiguous across every title they appear in, regardless of how their name is spelled in a given market.
• Franchise / brand IDs — related titles are grouped under one identity, so a universe or franchise holds together across its films, series and spin-offs.
• Sports entity IDs — disciplines, tournaments, teams and players each carry their own ID where sport is in the schedule.

Because identity is permanent and shared, your systems match once and stay matched. Updates, enrichment and corrections propagate automatically to every place the same content appears.

Linear TV Metadata is built around a separation that most feeds don't make: language-independent facts versus language-dependent text. Core facts are stored once; user-facing text exists in a proper version for every market.

This is why the guide is right in every market at once. Home Alone is one Content ID: "Home Alone" as its original title, "Kevin sam w domu" in Poland and "Maman, j'ai raté l'avion !" in France, all from the same record, sharing the same year, director and cast. A single Person ID renders as "Nicole Kidman" in English markets and "Nicole Kidmanová" in Czech, or in native script (章子怡) where that is what the audience expects. Nothing has to be translated on your side, and no two markets can drift out of sync, because the facts behind them are shared.

Linear TV Metadata connects to more than the schedule itself. Titles and people link to Festival & Awards Metadata — festivals and awards with categories, nominees and winners — so a film in tonight's line-up can surface as, say, a Cannes winner, and an actor's profile can show what they have won. The same permanent identifiers make these connections automatic rather than something you assemble yourself.

Thematic and mood tags don't only describe a title; they combine into ready-made browsing collections, such as "Retro 80s action comedies" or "Feel-good weekend watches", generated automatically from tag combinations rather than curated by hand. You get a browsing rail your editorial team didn't have to assemble, on top of the same tagging data already used for search and recommendations.