Visual hook rules: first frame, text on screen, sound off, and how long you have
The checklist for what the viewer sees before any word is heard: first frame, text, the sound-off case, and the attention windows the creator-growth skill states (about 0.1 s, 0.5 s, 1-2 s by age).
Check the first 3 seconds of this video against the rules below and return a pass/fail per rule with the fix.
How long you have, from the creator-growth skill: younger audiences decide in about 0.1 s, millennials in about 0.5 s, older viewers in 1 to 2 s. So the frame that is on screen before anyone has heard a word is the hook; the sentence is the second hook.
FRAME 1
- The subject (result, object, face mid-action) is already in frame at full size. No fade in, no logo, no title card, no walking into frame.
- Something is already moving or about to: a hand, steam, a cursor. A still frame that lasts 1 s reads as a photo.
- Framing survives the platform's overlays: subject in the centre 60 percent, nothing important in the bottom 20 percent or the right edge.
TEXT ON SCREEN
- The hook line is on screen from frame 1, in the same words as spoken, max 9 words, big enough to read at phone size (48-60 px at 1080 wide).
- One line, not a paragraph. If the hook needs two lines it is two hooks.
- Text sits in the upper third or the centre, never over the subject's face or hands.
SOUND OFF
- With the audio muted, frame 1 plus the text still tells the viewer what the video is and why to stay. Test: screenshot frame 1, show it to someone for one second, ask what the video is about.
- If the opening depends on a sound (a sizzle, a voice), the text carries the same promise.
SOUND ON
- The first word is spoken within 0.5 s. No breath, no 'so', no 'hey guys'.
- If there is music, it starts on frame 1, not after.
Video to check: {describe frame 1, the on-screen text, the first spoken words, and the audience age} Paste into your assistant as the first message of a new chat; fill the braces first.
Steps
- Export frame 1 as an image and look at it at thumbnail size on a phone. That is what decides in the first 0.1 to 0.5 s.
- Mute the video and watch the first 3 s. If you cannot say what it is about, rewrite the text, not the voice.
- Trim everything before the first frame that passes: fade-ins, logo stings and the half-second of someone sitting down are the most common cuts.
Example
Checked: a 30-second tutorial for a photo app, audience 25-40.
Frame 1: a face looking at the phone, then the product photo appears at 1.2 s. FAIL: the result must be frame 1; move the finished photo to 0 s, face at 2 s. Motion in frame 1: none until 1.2 s. FAIL: start on the thumb already mid-tap. Overlays: text in the bottom 15 percent. FAIL: move to the upper third. Text: ‘How I make my product photos look professional in seconds with this app’ (13 words). FAIL: ‘Listing photo. Two taps.’ (4 words). Sound off: with the audio muted the first 3 s show a face and a long caption. FAIL: with the result as frame 1 and the 4-word line it passes. First spoken word: ‘So, hey, today…’ at 0.9 s. FAIL: cut to ‘Two taps.’ at 0.2 s.
After fixes: frame 1 is the finished mug on white with ‘2 TAPS’ in the upper third, the thumb enters at 0.4 s, the voice says ‘Two taps’ at 0.2 s.
Tips
- The attention numbers come from the creator-growth skill and are the reason frame 1 matters more than the sentence: a 25-year-old has decided before the first word is finished.
- Camera basics decide weak against strong in that skill's list: depth in the background, light, energy, audio. Frame 1 is where all four are visible at once.
- In Claude Code, `/creator-growth` applies these rules when it plans the formats; use this prompt to audit a cut you already have.
| works with | By hand: No tool needed: a checklist or a technique. · Claude: Paste into claude.ai or the Claude app. · ChatGPT: Paste into ChatGPT; works with any GPT model. · Any LLM: Model-agnostic: Gemini, Mistral, a local model. |
| skill | npx skills add fortunto2/solo-factory/creator-growth · source |
| cost | Free. No API, no key. |
| status | Unverified: written from the technique, not yet run end to end by us. |
| license | CC BY 4.0 for the text here; skills are MIT. |
| updated | 2026-11-01 |
| for agents | visual-hook-first-frame-rules.md |
Related
Result-first hook: show the finished thing in frame 1, say how long it took
Template for the hook that opens on the outcome, the one that works with sound off and for products whose output is the content; the first frame is the result, not you holding it.
Hook for a generated video: what must be in frame 1 when a model draws it
The first-frame rules for Veo, Kling and Runway: start mid-action, name the motion already happening, keep text out of the generation, and generate the hook shot as its own 4-second prompt first.
Silent text-on-screen story: 25 seconds, one line per shot, paced for reading
The no-voice format that works with sound off because the text is the narration; the prompt paces each line to its shot at 2 words a second and puts the turn on the second-last card.
Hooks for a product with nothing to show: text and voice carry the first 3 s
For an API, a CLI, a service, a habit, anything without a screen worth filming: five hook shapes built on a typed line, a number and a voice, and what goes on screen instead of the product.
Generate 20 hooks for a topic and rank them by a stated rubric
One prompt that produces 20 openers across the 8 archetypes, scores each on four named criteria, and returns the top 5 with the score shown, so the ranking can be argued with instead of trusted.
A/B test hooks with campaign links: judge after a week on installs, not views
The creator-growth method: same video, different hook, one store campaign link per variant, installs per campaign read after a week. Includes the iOS ct= link shape and the noise rule.