The Anatomy of a 60-Second Video: Why the First 3 Seconds Decide Everything
Viewers decide to scroll or stay in 1.7 to 3 seconds. A 60-second video isn't one piece — it's four: a hook (0–3s), tension (3–15s), payoff (15–50s) and CTA (50–60s), held together by a pattern interrupt every 3–5 seconds. This is the structure we cut every client video to.
Most raw footage we receive has the good part at second 20. The creator warms up, introduces themselves, sets context — and by the time they say the interesting thing, 80% of viewers are gone. Editing for retention is mostly about moving second 20 to second 0. Here's the full anatomy.
0–3 seconds: the hook
One job: earn the next 3 seconds. Three hook types that work reliably:
- The payoff first. "Here's why your Reels get 200 views." Say the result, then explain.
- The contradiction. "Most people think X. They're wrong." The viewer wants to know if they're one of the wrong people.
- The specific number. "94% of patients check you online before booking." Specific beats vague every time.
Editing rules: speech starts on frame one — cut every breath before it. Hook text on screen, upper-middle, large. No logo, no intro card, no name. If the creator said "hey guys" first, it's gone.
3–15 seconds: the tension
You've earned attention — now make the viewer need the answer. Describe the problem, raise the stakes, or show the "before". This is where most videos lose people, because the creator explains too slowly.
Editing rules: cut every pause longer than 0.3 seconds. Punch in 10–15% on the second sentence. Drop in one b-roll clip at second 8–10 — this is the first pattern interrupt.
Retention graphs on short-form almost always show a steep drop between 8 and 12 seconds. That's where the novelty of the hook wears off. A visual change at second 8–10 — b-roll, a zoom, a caption colour flip — carries viewers over the cliff.
15–50 seconds: the payoff
Deliver the answer. One idea, fully explained, no tangents. If the creator covers three points, we cut to the strongest one and save the others for separate videos.
Editing rules: pattern interrupt every 3–5 seconds — alternate between the talking head and b-roll, add a caption highlight, use a subtle sound effect on key words. Word-by-word captions with the active word highlighted. Around 85% of social video plays muted; the captions are the audio.
50–60 seconds: the CTA
One ask. "Follow for part 2." "Comment your biggest question." "Link in bio." Not all three. The CTA text goes inside the safe zone — bottom 300px is blocked by the platform UI on every app.
Editing rules: hard cut to end. No fade, no outro card. The last frame should still be the creator's face — that's what shows in the loop.
The pattern interrupt toolkit
| Interrupt | When to use | Cost to attention |
|---|---|---|
| Punch-in (10–15% zoom) | Emphasis on a sentence | Very low |
| B-roll cut | Illustrating a point | Low |
| Caption colour change | Key word or number | Very low |
| Sound effect (whoosh, pop) | Transitions, reveals | Low |
| Text overlay / graphic | Lists, comparisons | Medium |
| Camera angle switch | Long talking segments | Low (if two angles shot) |
Use two or three types per video, not all six. Too many interrupts feel frantic and viewers scroll for a different reason.
What this looks like in the timeline
A typical 45-second edit from our studio: 14–18 cuts, 4–6 b-roll inserts, 2 punch-ins, 1 caption colour shift on the key number, 1 sound effect at the payoff, hard cut at the end. That's about 25 editorial decisions in 45 seconds — which is why "just trim it" is never what the job actually is.
Frequently asked questions
What should happen in the first 3 seconds of a short-form video?
State the payoff or the problem immediately — no intro, no logo, no 'hey guys'. Put the hook as on-screen text in the upper-middle of the frame and start speaking on frame one. Viewers decide to scroll within 1.7–3 seconds.
What is a pattern interrupt in video editing?
A visual or audio change every 3–5 seconds — a zoom, a cut to b-roll, a caption colour change, a sound effect — that resets the viewer's attention. Without them, retention drops sharply after 10 seconds.
How long should a short-form video be?
Between 20 and 45 seconds for most content. Long enough to deliver one complete idea, short enough that completion rate stays high. Completion rate is what the algorithms reward.
Should captions be word-by-word or full sentences?
Word-by-word with the current word highlighted, centred in the frame. Roughly 85% of social video is watched on mute, and word-by-word captions keep the eye moving and reading.
Send us your footage. We'll handle the rest.
10 short-form videos a month from $799 — edited, colour graded, captioned and published. No contract.
View Pricing →