Home › Guides

CCSL: the Combined Continuity & Spotting List, what it is, what it isn't, and why your localization vendor asks for it first

By the Fonti Studio post-production team · 7 October 2025

When a localization vendor asks you for the CCSL, two things are usually true. The film has one, because most professionally finished masters do. And the person fielding the request at the distributor isn't entirely sure what is in it, or why a subtitle house would care.

This is the explainer for that person. What a CCSL is, what it isn't, what it does for the subtitle pass and the dub pass, and what changes when you can only send us part of it. It is the document we ask for first, and the reason is narrow: it is the source-side ground truth that lets us lock every name, place, and term across a whole deliverable before a single line is translated.

Where it comes from

The CCSL is a US studio post-production tradition, and its roots are in copyright law, not localization.

Before 1912, American copyright had no separate category for motion pictures. Films could only be registered as a series of still photographs. The Act of August 24, 1912 created a category for motion pictures, and from that point registrants had to deposit both a copy of the film and a written description of it with the Copyright Office. Because nitrate stock was flammable, the Library of Congress did not retain the prints, but the descriptions were transferred to the Library and kept. For many films of the nitrate era, now considered lost, that paper record is the best or only surviving trace of the work. The convention of writing down what a film says and shows was a legal habit before it was an editorial one.

That requirement still stands. Under 37 CFR 202.21 and the Copyright Office's Circular 45, the description filed to register a motion picture may be a continuity, a press book, a synopsis, or a final shooting script, and it has to indicate the title, the general content, and the nature of the work. A continuity satisfies that. The CCSL is the modern, frame-accurate form of the same document, and it has become a standard line item in the delivery package that distributors, sales agents, and platforms expect.

What's in it

A CCSL combines two documents that used to be separate.

The continuity list is the picture side: a shot-by-shot description of the film, every camera cut logged with a timecode, every piece of on-screen action and text described in order. The spotting list is the sound side: the verbatim dialogue, plus narration, music cues, sound effects, and deliberate silences, each pinned to a frame-accurate in and out point. In a conventional CCSL the two run as parallel columns, action on the left, spotting on the right, so each line of dialogue sits next to the shot it is spoken over.

Out of those two columns fall the things a localization vendor actually mines:

• A scene log with timecodes, which is the spine of the timing.
• A spotting list of frame-accurate dialogue, which is the source text.
• A character map, naming who speaks each line, with the canonical spelling of every character's name.
• The proper nouns in their authoritative spelling: places, organisations, invented terms, anything the script capitalises and the audience will see repeated.

For a feature, this typically runs well over a hundred pages. It is dense, it is precise, and it is exactly the kind of thing a model will not reconstruct from audio.

Why it matters for subtitling

The single most useful thing a CCSL gives the subtitle pass is the glossary lock.

A feature has a few dozen names, places, and terms that recur. If those are not pinned to one canonical spelling before translation, they drift. The same model, sampling line by line, will spell a character one way in an early cue and another way forty minutes later. Across a multi-episode series, with different linguists on different episodes, the drift is worse. This is not a hypothetical failure mode; it is why platforms built tooling specifically to prevent it. Netflix calls its version a Key Names & Phrases glossary, a consistency sheet that holds the approved rendering of every recurring term so everyone touching the title uses the same string.

The CCSL is where that glossary comes from cleanly. The character map already tells us who is named what; the spotting list already shows us the terms in their authored spelling. We extract those into a locked glossary, review it once with you, and constrain the entire translation pass to those exact strings. The name in the subtitle, the name in the dub casting roster, and the name in the CCSL are then identical across every language and every cue. On our reference run, The Last Kumite delivered in French and Dutch, that lock held 47 cast, place, and term entries with zero drift.

The second thing the CCSL gives subtitling is source-side ground truth. Our pipeline can transcribe a film from audio when it has to, and it does it well, but automatic transcription is weakest exactly where it matters most: proper nouns. A speech model that nails ordinary words will still mis-hear an invented term, because the term carries no acoustic redundancy and no context the model has seen. A CCSL removes that problem. The dialogue is already written, already spelled, already attributed. There is nothing to recover and nothing to mis-hear.

Why it matters for dub

On the dub side the character map earns its keep.

A dub has to know who is speaking each line before it can route that line to the right cloned voice. We have an audio-visual stack that infers this from the picture and the diarized audio, and it is good, but it is inference. The CCSL states it. Every line is already attributed to a named character, and that character is already mapped to the cast. We use that attribution as the closed-set ground truth for speaker identity, and reserve our diarization and face-identity tools for confirming it and for the lines the CCSL leaves ambiguous, rather than for guessing from scratch.

The character naming matters for the same reason it matters in subtitling: the voice that carries a character in French has to be the same character's voice the audience met in the trailer and the synopsis. The CCSL is the document that makes the casting list, the subtitle glossary, and the on-screen name agree.

What happens when you don't have one

Plenty of indie masters arrive without a CCSL. That is workable. The fallback is the source-language SRT, a clean professional subtitle in the spoken language. If a film has been through any prior festival or platform pass, it usually exists.

A good source SRT gives us most of what the CCSL would: professionally authored timing that tracks the speech, clean dialogue text, and a stable set of proper nouns to extract the glossary from. We treat it as the authoritative source for both text and timing, use automatic transcription only as a cross-check against it, and lock the glossary from the subtitle's own spelling rather than from a character map. The result is close to the CCSL path. What we lose is the explicit speaker attribution and the explicit term list, both of which we then reconstruct rather than read off the page. It costs a little more verification work; it does not cost quality.

What happens when you have neither

When there is no CCSL and no usable source-language subtitle, only the audio, we will say so up front, and depending on the film we may decline the brief.

This is not a weakness of the pipeline. It will transcribe the film, bias the decoder toward your cast list, and correct mis-heard proper nouns against an online context pass, and on our reference title that chain reached high entity recall on the proper nouns that matter. The point is honesty about the floor. A dialogue-dense, accent-heavy, or heavily overlapped film with no written source has a real error budget that no amount of post-correction fully closes, and we would rather tell you that before you pay than hand you a master that needs a human re-read you didn't budget for. Some titles are fine on audio alone. Some are not. We will tell you which yours is.

Where to get one

We do not build CCSLs, and that is deliberate. A CCSL is a manual editorial trade, a person watching a film and logging every cut and every line by hand. It is a craft, it is priced accordingly, and it is not what we sell. Conflating "we localize your film" with "we author your continuity script" would muddy both.

If your master doesn't have one, it is a standard post-house deliverable. The specialist CCSL houses that the delivery requirements at the major platforms point toward will produce one to the spec your distributor or platform requires, and your post supervisor will usually know the vendor already, because the platform delivery package asks for it regardless of localization.

What Fonti Studio does sits one step downstream of the CCSL. We take the continuity, the spotting list, and the character map, lock the glossary from them, and turn them into broadcast-grade subtitles and original-cast dubs in your target languages, with every recurring name spelled the same way in every cue and every voice routed to the right character. Send the CCSL if you have it, the source SRT if you don't, and we'll tell you on day one which path your title is on. Email us a brief.

Fonti Studio is an AI-native subtitle and dubbing service for film distributors and sales agents. €400 per language for subtitles. €3,500 per language for dubs. Flat. EUR. Free preview before you pay.