Part 3 · Operate
Short video
Scripts bounded by what can be said aloud in the time on screen.
01What it is
A script contract for short vertical video: a hook, timed segments and one ask. Every bound is arithmetic rather than taste.
02Why it exists
Short video is the one format where a bad decision stays invisible until the render is paid for. So the rules are calculated: a segment whose words cannot be spoken in the time it holds the screen is refused at generation.
The speaking rate is the load-bearing number and everything else follows from it. The failure this prevents is a word still on screen when the next one starts, which is invisible in a script and obvious in a render.
03How to use it
- 01Keep the hook to what can be said in two seconds.
- 02Use the data composition only with real dated figures.
- 03Approve the script before anything renders. Rendering is the expensive step.
04What good looks like
- You give it
- A five-word hook and six segments, each with words that fit its seconds.
- You get
- A script inside the platform duration bounds with no density or dead-air violations.
- Why
- Word count divided by the speaking rate fits inside each segment’s time.
Derived from the rules in the code
05What weak looks like, and what it costs
- You give it
- A twelve-word hook.
- You get
- Refused, with the arithmetic in the message: the words need more seconds than the hook has.
- Why
- The hook is bounded by what can be said in two seconds, not by what fits on screen, which is far more.
Derived from the rules in the code