Burned-In Captions or Sidecar Files: A Guide to Vertical Video Exports

Two caption formats serve different delivery needs

Burned-in captions, also called open captions, are rendered directly into the video image. Once exported, they behave like any other visual element. A viewer cannot switch them off, and the text looks the same in a compatible video player regardless of its subtitle controls. This makes them useful for short-form social video, where automatic playback and muted viewing are common parts of the viewing environment.

Sidecar captions are stored separately from the video, usually in a text-based subtitle format such as SRT or WebVTT. The video remains unchanged while a compatible player reads the separate file and displays its contents. This arrangement allows viewers to select subtitles and can make localisation and corrections easier. However, the presence of a sidecar file does not establish that a particular social platform will import or retain it during upload.

The practical distinction is between predictable appearance and flexible delivery. A promotional clip designed for a single social feed may benefit from burned-in captions. A longer video distributed through a player with subtitle support may benefit from a separate caption track.

• Burned-in: captions are always part of the picture.
• Sidecar: captions remain separate from the video.
• Both: text needs accurate timing, spelling and punctuation.

The 9:16 safe zone is an interface problem

A vertical export normally uses a 9:16 canvas, commonly 1080 by 1920 pixels. That defines the video frame, not the usable area for captions. Social applications place interface elements over the picture, including account information, descriptions, engagement controls and navigation. Their positions can vary between applications, devices and interface updates.

A caption positioned at the extreme bottom of a 1920-pixel frame can therefore be obscured even when the exported file is technically correct. The same problem occurs near the right edge, where controls may cover words or punctuation. A safe zone is a working margin inside the frame that protects essential text and visual information from these overlays.

For example, an editor could reserve the lower 300 pixels and the outer 100 pixels on each side as an initial design precaution. These are illustrative margins, not universal platform specifications. The final placement should be checked against current platform guidance and representative device previews.

• Keep captions away from the bottom edge.
• Avoid placing important words beneath right-side controls.
• Leave enough horizontal space for long words and two-line captions.
• Preview the composition at actual mobile viewing size.

Caption design must work at phone size

A subtitle that looks balanced on a desktop editing monitor may become difficult to read on a phone. Small type, narrow outlines and low-contrast colours are particularly vulnerable when the image is reduced. A practical starting point is a clear sans-serif typeface, strong contrast and consistent spacing between lines. The text should remain readable without competing with the subject or covering essential visual information.

Two lines are usually easier to manage than a long sentence stretched across the screen. Break captions at natural grammatical boundaries, keeping closely connected words together. For example, a narration line reading 'The camera records each movement before the editor selects the final frame' could become two consecutive captions with a pause aligned to the speaker's delivery.

Avoid changing the position and styling on every shot without a reason. A consistent caption system gives viewers a stable reading location. Speaker labels, emphasis colours and sound descriptions can be introduced where they communicate useful information, but decorative treatments should not make the main dialogue harder to follow.

Prepare separate files for different publishing destinations

The export workflow should begin with a clean timeline master. Keep the original footage, graphics, narration and editable caption text available in the project. Then create the delivery version required by each destination. This prevents one platform's caption layout or compression requirements from becoming permanent across every subsequent upload.

A typical vertical video master might use a 1080 by 1920 frame, a progressive scan, a frame rate matching the source footage and H.264 video inside an MP4 container. AAC audio is a common delivery choice. These settings provide a useful starting point, but platform-specific limits and preferred codecs can change. Check the destination's current upload documentation before finalising a production preset.

For a platform that accepts separate subtitle uploads, prepare the appropriate text file and confirm that its encoding, timing and language metadata meet the destination's requirements. For a platform that expects captions in the picture, render a separate burned-in version. Retain an uncaptioned master so the same edit can be adapted later.

• Editable project master with separate caption text.
• Burned-in MP4 for destinations that require visible captions.
• MP4 and compatible sidecar file for supported subtitle workflows.
• Localised caption files for additional language versions.

Check the export after rendering

Caption problems can appear during rendering even when the editing timeline looks correct. A line may disappear at a cut, a font may be substituted, or a final caption may remain on screen after the narration has finished. Automated caption generation can also introduce transcription errors that are difficult to notice while concentrating on the picture.

Review the exported file from beginning to end, paying particular attention to the opening seconds, transitions, speaker changes and final line. Check that each caption appears when the relevant words are spoken and disappears before the next unrelated shot. Listen to the audio as well. A technically readable caption can still misrepresent the dialogue if punctuation or transcription changes its meaning.

For sidecar delivery, open the video and subtitle file together in a compatible player. Confirm that the text appears at the correct time and that special characters display correctly. For burned-in delivery, inspect the actual rendered pixels, including the bottom and side margins.

Build a reusable caption export preset

A repeatable preset reduces decisions at the end of each production. Record the canvas dimensions, frame rate, codec, audio format, caption font, approximate size, line limit and safe-zone margins. Keep platform-specific variations separate, especially when one destination accepts sidecar subtitles and another requires a fully rendered picture.

For a one-minute explainer, the final delivery might include a clean 1080 by 1920 master, a burned-in social version and an SRT file for a compatible player. The project should preserve the original text so corrections can be made without rebuilding the caption sequence.

The next practical step is to create a short 10-second vertical sample with dialogue, two-line captions and a final title card. Export it using the proposed settings, preview it on a phone and adjust the safe-zone margins before applying the preset to the complete film.

Questions readers ask