Caption Readability for Vertical Video: Size, Contrast, Fonts and Timing

Readable captions on vertical video come down to five things: big enough text, strong contrast, short lines, enough time on screen, and a position the platform interface does not cover. Get those right and almost any style works; get them wrong and even a beautiful template becomes noise.

This guide gives practical defaults for each, based on accessibility guidance and how vertical video is actually watched — on a phone, at arm's length, often in a hurry and often muted.

1. Size

On a 1080×1920 frame, caption text that is roughly 60–90 px tall for capital letters is a practical starting range for bold sans-serif fonts. Go larger for single-word styles, smaller only if lines get too wide. The test that matters: hold your phone at a normal distance and read it without effort.

Avoid thin or light font weights. They look elegant in a design tool and disappear after compression.

2. Contrast

Text must stand out against moving, changing footage. Options, from lightest touch to strongest:

• Drop shadow — subtle, works over darker footage.
• Outline / stroke — a thin dark stroke around light text reads over most backgrounds.
• Background box or pill — the most reliable over busy or bright footage.

The W3C's Web Content Accessibility Guidelines recommend a contrast ratio of at least 4.5:1 for normal text. 1 That guidance was written for web content, but it is a sensible benchmark for caption text against its immediate background.

3. Line length

Shorter is easier. For vertical video:

• One to five words per caption is typical for short-form.
• Keep a line within about 80–88% of the frame width.
• Use at most two lines.
• Break lines at natural phrase boundaries, not in the middle of a name or number.

4. Time on screen

Viewers need time to read. Very fast speech is better handled with shorter chunks than with tiny, rapidly changing text. As a rough editorial guide, a caption that disappears in less than a third of a second for a multi-word phrase is probably too fast to read comfortably; single highlighted words can move faster because the line stays visible.

Avoid flashing captions on and off with gaps between them. A small overlap or holding the last line until the next appears feels smoother.

5. Position

Platform interfaces cover the bottom and right side of vertical video. The lower-middle third — roughly 60–72% of the way down — is a dependable default. Move captions higher if the speaker's face is low in the frame. Details and margins are in our Reels safe zone guide.

Fonts that work

Bold, simple sans-serifs with clear letter shapes and open counters read best on small screens. Condensed fonts fit more words per line but can get hard to read at small sizes. Script and decorative fonts are best saved for one accent word, not whole lines.

Uppercase can look punchy for short phrases; mixed case is easier for longer phrases because word shapes are more distinct.

Colour and highlights

• White or off-white text with a dark stroke is the safest base.
• Use one highlight colour for the active or emphasised word.
• Check colours on footage with similar tones — yellow highlights vanish over sunlit sand.
• Don't rely on colour alone to convey meaning.

Motion

Animation should help the eye track speech, not compete with it. Subtle fades, small slides and gentle scale changes read better than large bounces on every word. See word-by-word animated captions for how the styles differ.

Accuracy is part of readability

A misspelled name or wrong number stops a viewer more than any style choice. Always proofread auto-generated captions. Our guide to fixing auto caption errors has a checklist.

Applying this in CaptionLabs

In the CaptionLabs studio you can change font size, words per line, word spacing, position, background and animation speed, and the exported MP4 uses the same layout as the preview. Templates start from readable defaults that you can adjust. Open the studio.

Quick checklist

• Readable at arm's length on a phone
• Contrast holds over the brightest part of the clip
• No more than two short lines
• Each caption stays on screen long enough to read
• Nothing covered by the platform interface
• Words and names spelled correctly
• Watched once with sound off

Frequently asked questions

What font size should captions be on Reels?

There is no official size. On a 1080×1920 frame, bold text with capitals about 60–90 px tall is a practical starting point. Test on a phone.

What is the best font for captions?

A bold, simple sans-serif with clear letter shapes. Avoid thin weights and decorative scripts for full lines.

Should captions be uppercase?

Uppercase suits short, punchy phrases. Mixed case is easier to read for longer lines.

How many lines should a caption have?

One or two on vertical video.