01
What actually changes between them?
Whether the words survive the journey. Text drawn into the picture appears everywhere the video appears, including when it is downloaded, reposted or shown by an app that ignores separate tracks. A separate track can be styled by the viewer and switched off, and it is frequently absent when a clip travels.
02
What survives the journey
For short clips the practical point is that words in the picture are always present, and always present beats switchable but missing.
| Event | Words in the picture | Separate track |
|---|---|---|
| Downloaded and reposted | Kept | Usually lost |
| Shared to another app | Kept | Usually lost |
| Viewed where tracks are ignored | Kept | Not shown |
| Viewer wants them off | Cannot | Can |
| Read for search indexing | Not directly | Sometimes yes |
03
When a separate track is the better choice
Long-form videos on a channel page, courses and training material often benefit from a separate track. Viewers can switch it on or off, choose a language, and resize the text on some players. For people using screen readers or other assistive tools, a proper track can also be more useful than words in the image.
Some platforms also use a separate track to understand what a video is about. For content that lives on a searchable page for years, uploading an accurate file alongside the video is worth the extra step, even if the vertical clips cut from it use words in the picture.
04
Doing both for the same content
Nothing stops you from using both approaches in one project. The full episode gets an uploaded track for accessibility and search; the short clips get words in the picture so they survive every repost. The same corrected text can serve both, so the checking work is only done once.
Keep the corrected text file with the original recording. If you later translate the episode, cut new clips or publish a written summary, that file is the starting point for all of them.
05
Avoiding double captions
If a clip already has words in the picture and a platform also generates its own automatic captions, viewers may see two sets on top of each other. Most apps let you switch off automatic captions for a post. Check each platform's settings when you upload a clip that already has words burned in.
The same issue appears when a source recording already had words on screen. Adding a second layer on top makes the frame crowded and hard to read. Either crop the original words out or choose moments where they are not present.
06
Words that travel with the clip
- Draw them in for social clips. It is the only approach that survives being shared.
- Keep a separate track for long form. Viewer control and search both benefit there.
- Do not rely on a platform to add them. Automatic captions vary and are not always shown.
Make one clip and repost it to check. Nothing to pay for the first.
Questions people ask
- Which is better for accessibility?
- A separate track offers viewer control, which matters on long form. On short clips the practical answer is that words in the picture are always present, and always present beats switchable but missing.
- Do platforms read a separate track?
- Some use it for search and indexing, which is a genuine advantage on long form.
- Can I switch words off here?
- There is no option to remove them. Every clip carries its words drawn into the picture.
- Does drawing them in reduce quality?
- The words become part of the picture, so they are subject to the same compression as everything else. At normal sizes it is not visible.
- Which do most social clips use?
- Words in the picture, essentially universally, because a clip that loses its text when reposted has lost most of its content.