Text on the Screen: How On-Screen Text Works in Video Editing and How to Use Titles, Captions, Labels, and Graphics Effectively
6 min read
Use on-screen text only when it helps the viewer understand, remember, or act faster. Every title, caption, label, and graphic should earn its place. Good text supports the video without stealing attention from the story. Bad text clutters the frame, repeats the obvious, and makes even polished footage feel amateur.
TLDR: On-screen text works best when it is short, readable, timed well, and visually consistent. For example, a two-minute product tutorial might use a title card, three step labels, captions for speech, and one final call to action instead of filling every shot with text. In social video tests, captions often raise completion rates because many viewers watch with sound off; some publishers report that more than 60% of mobile views start muted. Treat text as guidance, not decoration.
What on-screen text actually does
Text on the screen is not one thing. It has several jobs. Each job needs a different style and level of priority.
- Titles introduce the video, a section, a speaker, or a topic.
- Captions show spoken words, useful for accessibility and muted viewing.
- Labels identify people, places, objects, steps, or data points.
- Graphics explain relationships, numbers, timelines, and ideas that footage alone cannot show.
The best editors separate these uses clearly. A title should not look like a subtitle. A label should not fight with a caption. A call to action should not look like a random note. Viewers read hierarchy before they read words. Size, weight, placement, and timing tell them what matters first.
Start with hierarchy, not font hunting
It drives me crazy when editing software offers hundreds of flashy text presets but makes basic alignment take four clicks too many. Presets can help, but they are not a strategy. Before choosing a font, decide what the viewer must notice first.
A simple hierarchy might look like this:
- Main title: largest text, brief, often centered or placed in a clean area.
- Section title: medium size, used to mark a shift in the video.
- Caption: smaller, consistent, placed near the bottom.
- Label: compact, close to the item or person it identifies.
- Fine detail: used sparingly, only when the video can pause or slow down.
Use no more than two font families in most videos. One is even better for tight brand work. Vary size, weight, color, and spacing instead of adding more fonts. Typography should feel controlled. If every phrase shouts, nothing is heard.
Readability beats style every time
Video text has a brutal constraint: viewers do not get much time. They may be watching on a phone, outdoors, while scrolling, or with half their attention missing. Your text must survive poor conditions.
Use high contrast. White text over bright footage fails often. Add a dark shadow, a semi-transparent box, a blur panel, or place the text in a darker part of the frame. Avoid thin fonts for captions. Avoid all caps for long lines. Keep line lengths short.
A useful rule is the three-second test. Show the frame to someone for three seconds. If they cannot read and understand the text, simplify it. For social clips, even two seconds may be a fairer test.
Readable text is not boring. It is respectful. The viewer should not have to pause the video to decode a label or subtitle.
Timing is where many edits fail
Text timing should match the viewer’s reading speed and the visual rhythm of the edit. If text appears too late, it explains what the viewer has already missed. If it vanishes too soon, it creates irritation. Honestly, it feels like a small betrayal when a data label disappears half a second before the audience can read the number.
For most title cards, leave text on screen for at least two seconds. Longer titles may need four or five. Captions should follow speech closely, but not flash word by word unless there is a stylistic reason. Labels should appear before the viewer needs them and stay long enough to connect with the object or person.
Use motion with care. A clean fade, slide, or scale change can guide attention. Overdone animation makes the text feel cheap. If the viewer watches the animation instead of reading the message, the effect has failed.
Captions are not optional extras
Captions serve accessibility first. They also help comprehension, search, localization, and silent viewing. For interviews, explainers, training videos, and social clips, captions should be treated as part of the edit, not an afterthought.
- Keep caption lines short: one or two lines is usually enough.
- Break lines by meaning: avoid splitting phrases in awkward places.
- Use speaker labels when needed: especially in panel videos or dialogue scenes.
- Check spelling manually: auto-caption tools still damage names, brands, and technical terms.
- Respect safe areas: captions should not be hidden by platform buttons.
Auto-generated captions save time, but they are not finished work. Expect to waste time on corrections if the audio has accents, music, noise, or industry terms. Still, corrected captions are worth it. They make the video usable for more people and reduce confusion.
Titles and lower thirds should identify, not distract
A title announces context. A lower third identifies a person or role. Both should be clean and brief. A good lower third might show a name on the first line and a role on the second line. That is enough. Do not add a biography unless the viewer truly needs it.
Place lower thirds away from faces, hands, product details, and platform overlays. Keep them consistent across the whole video. If the first speaker gets a blue nameplate, the next speaker should not get a spinning gold ribbon. Consistency builds trust.
For title cards, choose words with care. “How to Clean a Camera Sensor Safely” is better than “Camera Sensor Cleaning Tips and Things to Know.” It is sharper. It promises value. It fits on screen.
Labels and graphics should reduce mental effort
Labels work best when they point to something specific. Use them for ingredients, parts, locations, steps, warnings, or data. Place labels close to the item they describe. If you need arrows, keep them simple. If you need five arrows, consider a pause, a diagram, or a cutaway shot.
Graphics are useful when the camera cannot show the idea clearly. Use charts for numbers. Use icons for categories. Use simple maps for location. Use progress indicators for tutorials. But do not turn every statistic into a busy infographic. Motion graphics should clarify, not show off.
For data, round numbers when precision is not needed. “About 75%” is often easier to absorb than “74.83%.” If precision matters, give the viewer enough time to read it.
Use a repeatable text system
A text system saves editing time and keeps the video coherent. It can be simple:
- Font: one clear sans serif for most uses.
- Colors: one primary color, one accent color, white, and dark gray.
- Caption style: fixed size, position, background, and line spacing.
- Title style: fixed placement and animation.
- Label style: small, direct, and close to the subject.
Save these settings as templates when your software allows it. This prevents small errors from spreading across a project. It also makes revision faster when a client asks for a new color, a larger caption size, or a different logo lockup.
Final checks before export
Before exporting, watch the video on the smallest screen your audience is likely to use. Then watch it with sound off. Check every text element against these questions:
- Can it be read quickly?
- Does it add information?
- Is it consistent with the rest of the video?
- Does it block faces, products, or key action?
- Does it remain visible after platform controls appear?
- Is the spelling correct?
On-screen text is most effective when it feels almost invisible in its craft. The viewer understands more, remembers more, and moves through the video with less effort. That is the real goal. Not more text. Better text.
