Skip to the tool
CaptionSuiteCaptionSuite home

Guides

Caption reading speed: how long a line should stay on screen

Updated

For English, aim for at most 42 characters per line, at most two lines at once, and a reading speed of about 17 characters per second. Keep every cue on screen for at least five sixths of a second and rarely more than seven. These are the streaming and broadcast conventions, and they exist because a viewer is watching the picture as well as reading.

The numbers, and where they come from

Every captioning style guide converges on roughly the same figures, most visibly Netflix's timed-text specification for English. They are conventions rather than laws, but they encode a real constraint: reading competes with watching, and a caption that wins that competition has cost the viewer the shot.

  • 42 characters per line — beyond this the eye has to travel too far back for the next line
  • 2 lines maximum — a third line starts covering the thing you are captioning
  • 17 characters per second — comfortable for an adult reader; past about 21 they are choosing between reading and watching
  • 5/6 of a second minimum — below this a caption flashes rather than reads
  • 7 seconds maximum — a caption held long after the line was spoken makes the whole track feel out of sync

What to do when a line is too fast

Reading speed is characters divided by duration, so there are only two levers: fewer characters, or more time.

Giving it more time usually means borrowing from the gap before the next cue, which works when there is a pause and not when there is not. Splitting the cue in two is the better fix for a long sentence — each half gets its own screen time, and the split should land on a natural clause boundary rather than in the middle of a phrase.

Cutting characters is legitimate too. Captions are not a transcript: dropping a filler word or a repeated false start loses nothing and buys reading time.

Where to break a line

Break lines to balance their lengths, not to fill the first one. Greedy wrapping produces a full line and a one-word stub underneath it, which looks broken burned into a video.

Where the meaning allows, break at a clause boundary so each line reads as a unit. Avoid splitting an article from its noun, or a preposition from what it governs — the eye stumbles on the line that begins with a dangling 'the'.

This tool wraps automatically and balances the lines, and it re-wraps when you change caption style, because a big social-video preset fits far fewer characters across a vertical frame than a standard subtitle does across a wide one.

Vertical video changes the numbers

The 42-character convention assumes a wide frame and a subtitle-sized font. Large captions on a 9:16 clip fit closer to 20 characters a line, which means more cues, each shorter — and the reading-speed limit becomes the binding constraint much sooner.

It also means the bottom of the frame is unusable: a platform's own interface — the handle, the caption, the buttons — covers it. Captions for a vertical feed belong above the lower third, which is where the social preset here puts them.