SuperAPI AI Directory

visual-match

Design & creation

View repository

Updated 10-29 17:32

About this skill

Visual Match

Most people decide with their eyes before they hear a word, and many watch with the sound off. If the picture says "man on a sofa" and the words say "this garden took one weekend", the picture wins and they swipe.

Three channels, one message

Channel Job Common miss
Visual (0 to 1 s) Show the subject, ideally the most surprising or wanted frame Talking head at a desk, logo, black screen, slow walk in
On-screen text Make the topic clear with sound off, 3 to 7 words, adds to the speech Repeats the speech word for word, or a vague label ("Tips!!")
Spoken The hook line Starts with a greeting or a name

They must point at the same subject. If they don't, change the cheapest one.

Step 1: Find the key frame

List what the footage already shows. Pick the frame that best proves or teases the payoff: the result, the before and after, the moment something changes, the strange object. Usually it sits in the middle or end of the video. Move it, or a 1-second flash of it, to second zero. Prefer a frame where something moves (a hand, an object, a change) over a still one.

Step 2: Pick a visual hook type

Type Example
Text overlay that sets the question "Why this cake sank" over the sunken cake
A change of shot in the first second Cut from the face to the finished room
Subject movement A hand drops a cracked tile, a door swings open
Effect or transition A split screen of before and after
An arresting shot A frame so unusual it needs explaining

Step 3: Write the text

  • 3 to 7 words. Readable in a second. Sits clear of the platform's buttons and captions (keep it in the upper middle on vertical video).
  • Adds to the spoken line: the words name the problem, the text names the twist, or the reverse.

Step 4: Check the match

Play it in your head with the sound off. Does a stranger know the topic? Play it with the eyes closed. Do they still know? Both should be yes.

Output

Key frame: [what, and where it is in the footage now]
First shot (0 to 1 s): [shot]
On-screen text: "[words]"
Spoken: "[words]"
Edit: [move / cut / reshoot, exactly]

If the idea is the problem (the visuals can't show what the words claim), say so. Drop or reshape the hook rather than promise what the picture can't prove.