The Anatomy of a Scroll-Stopping Hook
The first two seconds of a short-form video decide whether it lives or dies. Before a single word of your message lands, the viewer has already made a subconscious call: keep watching, or keep scrolling. Everything else you shot is downstream of that decision. You can have perfect lighting, a sharp edit and a genuinely useful message, and none of it matters if the opening frame gives someone permission to leave.
I have made hundreds of videos across content categories that people do not naturally seek out, from legal services to founder brands. The single biggest lever on performance was never the camera or the grade. It was the hook.
Say the stakes, not the setup
Most hooks fail because they warm up. They introduce, they contextualise, they clear their throat. "Hi guys, so today I wanted to talk about something that a lot of people ask me about" is four seconds of nothing, and four seconds is an eternity on a feed.
A strong hook skips all of that and opens on the tension: the promise, the problem, or the pattern-break. Compare "today I want to talk about wills" with "if you die without a will, the government decides who raises your children." Same subject. One is an introduction, the other is a stake. Lead with the thing the viewer would be worse off for not knowing.
Show motion in frame one
A static opening frame reads as a photo, and photos get scrolled. Movement signals that something is happening and rewards the eye for staying. That movement does not need to be expensive: a gesture, a step toward camera, a hard cut on the first syllable, or a slow push in will all do the job.
Pair that motion with on-screen text that raises a question the viewer needs answered. The caption is not a summary of the video, it is a headline for it. If your text could be swapped onto any other video in your niche, it is too generic to earn a stop.
Make the first line impossible to half-hear
Most people watch with the sound low, in a noisy room, with half their attention elsewhere. Your first line has to survive that. Keep it short, front-load the concrete noun, and cut every qualifier. "There is a fairly common mistake people tend to make" becomes "this mistake costs families thousands." Specific beats clever, every time.
Earn the next line
Retention is a chain: every line exists to buy the next one. Once the hook lands, the job becomes keeping the momentum without deflating it. That means cutting dead air ruthlessly, holding the pace, and never explaining what you can show. If a sentence does not add tension, information or payoff, it is costing you viewers.
The practical test I use before publishing: mute the video, watch the first two seconds, and ask whether I would keep watching. If the answer is no, the problem is not the algorithm. It is the opening.
Test hooks, not videos
The most useful habit I have built is treating hooks as their own unit of work. When a video underperforms, I do not always rebuild it. I rewrite and reshoot the first two seconds and post it again. Often the body of the video was fine and the door was simply closed before anyone got inside.
Do that consistently and you stop guessing. Over a few dozen videos you build a private library of openings that reliably earn attention in your specific niche, which is worth more than any general advice, including this.