Forced subtitles, explained: what a forced narrative (FN) track is, when a delivery needs one, and why rejections trace back to it
A subtitle delivery for a feature is rarely one file per language. For most platforms and most professional buyers it is two: the full dialogue track, and a much shorter second track called the forced narrative, usually abbreviated FN. Viewers meet the same thing under a different name: the "English (Forced)" entry in a player's subtitle menu, the forced subtitles that appear even when subtitles are switched off. The dialogue track everyone understands. The FN track is the one that produces confused emails, because it is short, oddly formatted, mostly in capitals, and apparently redundant. It is also, in our experience, where a disproportionate share of QC rejections actually live.
This is the explainer for the person at the distributor fielding the note that says "FN missing or non-conforming." What the track is, why it exists as a separate file, what goes in it, what must stay out of it, and the three failure shapes we see when one gets rejected.
What a forced narrative track is
Netflix's partner documentation defines a forced narrative subtitle as "a text overlay that clarifies communications or alternate languages meant to be understood by the viewer." In practice the track carries two kinds of content.
The first is on-screen text: title cards, location and date supers, inserts, plot-relevant signs, a text message filling the frame, a phone screen the camera lingers on. A Dutch viewer watching an English-language film with subtitles off still needs to know that the card that just appeared says ONE YEAR LATER. The FN track is where EEN JAAR LATER lives.
The second is foreign-language dialogue: any line spoken in a language other than the film's primary language, where the audience is meant to understand it. If two characters in an English-language thriller switch to Russian for a scene the original audience was given subtitles for, every localized version needs that same help, in its own language, even when the viewer has full subtitles switched off.
That last clause is the whole reason the track exists. "Forced" means the text renders whether or not the viewer has subtitles enabled. It is the subtitle you cannot turn off, because the filmmaker never intended you to watch that moment without it.
Forced subtitles vs SDH vs burned-in text
Three terms get conflated in briefs, and they are three different deliverables.
Forced subtitles (FN) carry only what every viewer must read: plot-pertinent on-screen text and foreign-language dialogue. A few dozen cues on a typical feature, delivered as their own timed text file per language.
SDH (subtitles for the deaf and hard of hearing) is the opposite end of the scale: the full dialogue plus speaker identification and sound description, everything a viewer needs when the audio is unavailable. An SDH track contains the FN events; an FN track contains almost none of the SDH track.
Burned-in subtitles are not a track at all: the text is rendered into the picture itself and can never be switched off or localized. Platforms have moved away from them for exactly that reason; Netflix requires the primary video to be textless, with both full and forced subtitles delivered as separate overlay files. A master with the original's English cards burned in cannot carry a clean Dutch FN track over them, which is why the textless requirement and the FN requirement arrive together.
If a brief says "forced subs," it means the FN track. If it says the master has burned-in text, that is a problem to raise before localization starts, not after.
Why it has to be a separate file
The instinct at many vendors is to fold the cards into the dialogue track and call it done. The platforms' display logic is exactly why that fails.
On Netflix, forced narratives display when the viewer has full subtitles and closed captions switched off. The moment a viewer activates a full subtitle or SDH track, the FN track is suppressed, and the full track is expected to carry every FN event itself. The partner spec is explicit on both halves: FN subtitles are delivered as separate timed text files against a textless primary video, and "all Forced Narrative events are also included in each full Subtitle and SDH/CC file."
So the card content appears in two places by design: once in the standalone FN track (for viewers watching with subtitles off, or listening to a dub), and once inside each full subtitle track (so nothing disappears when the viewer switches subtitles on). What the spec does not allow is the shortcut of shipping only the merged version. A dialogue track with cards mixed in, and no standalone FN file next to it, fails the package.
There is a second reason the separate file matters, and it is the one distributors feel first: dubs. A dubbed viewer has localized audio and, typically, no subtitles on screen. The dub covers every spoken line. It covers nothing the audience reads. Every location super, every time jump, every plot-relevant text message is invisible to the dubbed viewer unless an FN track in the dub's language is loaded over the picture. The moment a title sells with dubbed audio, the FN track stops being a formality and becomes the only thing standing between the viewer and a plot hole.
What belongs in it, and what must stay out
The editorial rule is plot pertinence, and it cuts both ways. Netflix's Timed Text Style Guide requires forced narratives for on-screen text to be included "only if plot-pertinent," rendered in ALL CAPS (mixed case is reserved for long written passages), never combined with dialogue in the same subtitle event, and deleted when redundant, for example when the on-screen text is already in the target language or the line is spoken aloud.
In practice, on a feature, that resolves to a shortlist. In: the time-jump card, the location super, the phone screen the plot turns on, the news chyron a character reacts to, the foreign-language exchange. Out: set-dressing signage (the bar's name above the door, unless someone is looking for that bar), the billing block, and any text the camera does not ask you to read. The main title card is a judgment call with a convention behind it: keep the original title on screen unless the client supplies an official localized title, because the title is a brand asset, not a translation problem.
Timing is the other half of conforming. An FN cue for on-screen text is expected to mimic the text it translates: on when the card cuts in, off when it cuts out, adjusted only as far as minimum-duration rules require. A card that lingers for four seconds with an FN cue that flashes for one reads as a defect, because it is one. The only way to verify this is against the picture, cue by cue, which is why we time and check FN tracks on a timecoded screener rather than trusting a spotting list's stated in and out points.
The three ways FN deliveries fail
We spent part of this summer rescuing a rejected delivery on an 89-minute feature: two subtitle languages, produced by another vendor, bounced by the client's QC. The dialogue translations were largely fine. The FN side showed all three of the classic failure shapes at once, which is what makes it a useful case study.
First: the missing track. One of the two languages simply had no FN file. Every card in the film was dark for that territory the moment a viewer watched with subtitles off. Nothing in the dialogue track's own QC would ever surface this, because the defect is a file that does not exist.
Second: the mislabeled track. The other language had a file with FN in its name, which turned out to be a duplicate of the full dialogue track. To a checksum it is a healthy file; to a spot check it even plays. But a platform ingesting it as forced narrative would render the entire dialogue track over the picture for every viewer with subtitles off. The label said FN. The content said otherwise.
Third: the editorial misses. Once we built the tracks properly, the film needed exactly four forced cues per language: a time jump, a location super, and two phone-screen inserts, each verified frame-accurately against the screener's burnt-in timecode, each rendered in caps, each excluded from nothing (the same four cues were folded into the full subtitle tracks, per spec). The set-dressing signs stayed out. The billing block stayed out. Four cues is a normal number: an FN track is usually tiny, which is exactly why it gets skipped, forgotten, or padded with the wrong content by vendors who never priced it into the job.
The delivery went back out with real FN tracks and passed. The lesson for a distributor is not the fix, it is the intake question: when you commission subtitles, ask explicitly whether the quote includes forced narrative tracks per language, as separate files, timed against picture. If the vendor asks what you mean, that is your answer.
What to put in your delivery brief
Three lines cover it. State the primary language of the film and every scene of foreign-language dialogue you know about. State that you require, per target language, a full dialogue track and a standalone FN track, with FN events duplicated into the full track. And state the delivery formats you need them in, because professional buyers often want frame-based broadcast formats alongside the SRT, and those conversions have their own failure modes that surface as sync drift rather than missing files.
Common questions
What are forced subtitles?
Forced subtitles (forced narrative, or FN, subtitles) are the subtitle events that display even when a viewer has subtitles switched off: translations of plot-relevant on-screen text and of dialogue spoken in a language the audience is not expected to understand. They are delivered as a separate timed text file per language, alongside the full dialogue track.
What does "English (Forced)" mean in a player's subtitle menu?
It is the English forced narrative track. Selecting it, or leaving subtitles off on a platform that auto-applies it, shows only the forced events: foreign-language lines and on-screen text cards. It is not a broken or incomplete subtitle track; carrying only a handful of cues is its job.
Are forced subtitles the same as SDH?
No. SDH carries the full dialogue plus speaker IDs and sound description for viewers without access to the audio. Forced subtitles carry only the on-screen text and foreign-dialogue translations every viewer needs. They sit at opposite ends of the same spectrum, and platforms require the forced events to also appear inside each full subtitle and SDH file.
Does a forced narrative track have to be a separate file?
For most platforms, yes. Netflix requires FN subtitles as separate timed text files against a textless primary video, because the display logic (FN shows when subtitles are off, and is suppressed when a full track is on) only works when the platform can address the track independently.
Fonti Studio delivers FN tracks as standard on every title that carries on-screen text or foreign-language dialogue: extracted from your CCSL's supers when you have one, spotted against the picture when you don't, timed to the card, verified against burnt-in timecode, and delivered per language alongside the dialogue masters in the formats your buyer specifies. If you have a delivery that bounced on FN, or a brief where nobody has mentioned the second track yet, email us the brief. We will tell you what the package needs before you commit to anything.
Fonti Studio is a subtitle and dubbing studio run by post-production engineers, for film distributors and sales agents. €400 per language for feature subtitles, forced narrative tracks included where the title needs them. Flat. EUR. Free preview before you pay.
Heading into a delivery window?
Send a film master, trailer, and CCSL or original-language SRT. We send back a free preview in your target languages and tell you up front whether the master will pass.
Email us a brief