You want your video captioned, and somebody has asked whether you mean open or closed captions. Or a platform has a captions switch and you’re not sure whether it makes the text you burned in pointless. The two kinds solve different problems, and on short-form video the answer is often both.
The short answer. Open captions are rendered into the video frames, so every viewer sees them and nobody can turn them off. Closed captions are a separate text track the player draws on top, so viewers can switch them on, off, and often restyle them. Open captions win on a muted social feed. Closed captions win on accessibility controls and searchability. Where a platform offers its own captions, using both covers the gaps in each.
What open and closed actually mean
The words describe where the text lives, not what it says.
Open captions are part of the picture. The text is drawn into each frame while the video is encoded, the same way a logo or a filter would be. On social video this is usually called burning captions in. Whatever player shows the file shows the captions, because they are the file.
Closed captions are data that travels next to the video. That might be a sidecar file such as SRT or WebVTT, a track inside the video container, or text a platform generates itself after you upload. The player reads the track and draws the words over the video at the right moments. That’s what the CC button controls.
Neither term says anything about quality. Both can be accurate or sloppy, styled well or badly. The difference is who is in control of the text once the video leaves you.
What open captions do well
They’re always there. A lot of short video is watched in the feed with the sound off, and most viewers never open a captions menu. Open captions reach them without anyone tapping anything.
They look the way you designed them. Font, size, colour, outline and position are fixed in the frame. That matters on vertical video, where each app covers a different part of the screen with buttons and text, and you need the words to sit in a known safe area.
They survive the trip. A clip gets downloaded, reposted, shared into group chats and uploaded again somewhere else. A caption track rarely makes that journey. Burned-in text does, because it’s the same pixels everywhere. The guide to burn subtitles into a video on iPhone sets out where each approach is accepted, side by side.
Where open captions fall short on accessibility
Open captions help a lot of viewers, including many who are deaf or hard of hearing. But they hand every choice to the creator, and some viewers need to make those choices themselves.
Viewers can’t change them
Closed captions can be adjusted by the viewer in many players. On YouTube, for example, the caption options at the time of writing include font, size, colour and background. On an iPhone, Settings has a Subtitles & Captioning section under Accessibility that restyles captions in apps that support it. A viewer with low vision can make closed captions bigger. With open captions, the size you picked is the size they get.
The same goes for language. A closed track can be one of several, and the viewer picks. Open captions come in one language, and a second one has to be burned in alongside it.
Machines can’t read them
A caption track is text. A platform can index it, translate it on the fly and feed it to other tools. Burned-in captions are just shapes in the picture. Some platforms may run text recognition on frames, but you can’t count on it, so treat burned-in words as invisible to search.
Compression can blur them
Open captions go through every re-encode the video does, and platforms re-encode everything you upload. Thin or brightly coloured text can soften along the way. A closed track is redrawn sharp by the player every time. If you’re burning captions in, the post on why captions look blurry after upload covers the settings that keep them legible.
Where closed captions fall short on social video
Closed captions are the accessibility standard for broadcast, streaming and the web. On short-form social video they run into three practical problems.
First, the viewer has to turn them on, or the platform has to turn them on for them. Defaults vary by app and change from time to time, so you can’t assume a muted viewer will see them.
Second, you often can’t supply your own track. On the major short-form apps, captions are generally generated by the platform from your audio after upload, and at the time of writing there’s no field in the phone apps to attach a caption file you’ve corrected. YouTube does accept caption file uploads, but through YouTube Studio rather than the Shorts upload screen.
Third, the platform decides how they look and where they go. That’s fine for many videos. It isn’t fine if they land on top of a product you’re showing or the text in your hook.
When to use both
For most short videos, burn captions in for the muted feed, and leave the platform’s own captions available for viewers who rely on them. They don’t compete. One reaches people scrolling past. The other gives control back to the people who need it.
There is one catch. If a viewer turns on the platform’s captions over a video that already has burned-in text, they see two sets of words at once, sometimes overlapping. You can’t stop that, but you can make it less awkward:
- Place open captions where platform captions don’t usually go. Look at where the app draws its own captions on your phone, and keep yours clear of that area as well as the buttons.
- Make sure the words match. If your burned-in captions are corrected and the auto track isn’t, a viewer sees two different versions of the same line. On YouTube, review the automatic captions in Studio after upload.
- Put the important line in the post text. It’s readable by screen readers and search, and it’s there even if neither set of captions is on.
A quick way to decide by destination:
| Where the video goes | Open | Closed |
|---|---|---|
| TikTok, Reels, Shorts | Yes, for the muted feed | Leave the platform’s on, check it |
| YouTube long-form | Optional | Yes, upload a corrected file |
| Your own website | Optional | Yes, a WebVTT track the player can show |
| Sent in a chat or reposted | Yes, it’s the only kind that travels | Usually lost |
| Broadcast or a client with an accessibility spec | Ask first | Usually required |
If you’re making the open half on a phone, Capster transcribes the audio, lets you fix the words and style them, and renders the captions into the frames on the device. For the platform side, the page on subtitles for YouTube Shorts explains how YouTube’s automatic subtitles sit alongside captions you’ve burned in yourself.