ChumokuA Short loses viewers in the first two seconds because the feed gives it no click to trade on. On long-form, the viewer chose the video from a thumbnail and a title before it started playing, so the opening gets a few seconds of patience it did not have to earn. In a Short feed there is no choosing. The video is already playing, the only two options are keep watching or swipe, and the opening frame is doing the entire job the thumbnail used to do. Most Shorts that lose people early are not badly made. They open with a wind-up, a logo, a greeting, or a restatement of what the caption already said, and every one of those spends the two seconds that decide the video.
The feed is not a click
On a normal YouTube video the viewer has already committed. They read a title, looked at a thumbnail, decided this was worth their time, and clicked. That decision buys the opening several seconds of goodwill.
A Short arrives with none of that. It starts playing on its own, in a stack of other videos, and the viewer is holding a thumb over the screen. Nothing has been chosen. The question is not "is this worth continuing" but "is this better than whatever is one swipe away", and that question is answered before most creators have finished their first sentence.
This is why advice written for long-form openings transfers badly. "Set up the premise" is good advice when someone has already decided to watch. It is a way to lose a Shorts viewer.
The four things that spend the first two seconds
A wind-up. Anything before the subject appears: a logo sting, "hey guys", a slow zoom onto an empty frame, a beat of silence while the music starts. None of it is information, and all of it costs the same two seconds as information would.
A repeat. The caption or the on-screen text has already told the viewer what this is, and then the first line of narration tells them again. The viewer has now been given the same fact twice and no reason to stay for a third.
A frame that looks like the last one. Shorts are watched in a run. An opening frame that is visually generic - a person centred in a plain room, a gameplay clip with no distinguishing detail - reads as the video they just swiped away from, and gets treated the same way.
A delayed subject. The thing the video is actually about arrives at four seconds. Whatever is on screen before it is a trailer for the video the viewer is already leaving.
What to change in the edit
Find the most specific frame in the video and put it first. Specific means it could not be the opening of anyone else's Short: a strange object, a number, a face mid-reaction, the moment something breaks. The video can then explain itself backwards.
Cut until the first sentence carries a fact or a question. If the first line could open any video in your niche, it is not an opening, it is a throat-clear.
Say a different thing than the on-screen text says. The two should work together, not repeat: text states the situation, narration starts the argument.
Check the first frame as a still. Pause the export on frame one and look at it alone. If it does not make you ask a question, the video is starting one edit too early.
How to tell whether this is actually your problem
Do not guess at it. YouTube Studio shows a Short's audience retention curve, and a first-seconds problem has an unmistakable shape: a near-vertical drop at the very start, before the curve settles into its normal decline.
That number is not available through any public API, and no third-party tool has access to it. Anything quoting your early swipe-away rate without you connecting your own channel is estimating it from public view counts. Read it yourself in Studio, under the video, in the engagement tab.
If the curve is flat at the start and drops later, the opening is fine and the problem is somewhere in the middle. Fixing the hook will not help that video.
Common questions
How long is the hook of a YouTube Short?
In practice the decision is made in the first one to two seconds, which is roughly the first sentence or the first two cuts. There is no official window; it is just how long a viewer takes to decide whether to move their thumb.
Should a Short open with the conclusion?
Often yes. Opening on the result and explaining how you got there gives the viewer a reason to stay that a chronological opening does not. The exception is when the result only means something after a setup, in which case the setup has to be one line, not five seconds.
Does the caption or title matter for Shorts?
Far less than for long-form. In a feed the video is already playing before anyone reads the caption, so the opening frame and the first line do the work a title and thumbnail do elsewhere. Keyword work that pays off on long-form has very little to do with whether a Short holds people.
Where do I see how many people swiped away?
YouTube Studio, on the individual video, in the engagement or audience-retention view. It is not exposed by the public YouTube API, so a tool can only show it to you if you have connected your own channel to it.
Hooks · Retention · Editing
Published 12 September 2026
Chumoku does this reading for you: send it a Short and it watches the whole thing, marks where attention drops, and tells you what to change in the edit. With your channel connected it reads your real retention curve rather than guessing at it.
Try itKeep reading
20 September 2026 · 6 min read
What view jail on Shorts actually tells you
Shorts that keep stopping at the same view count. What the ceiling narrows the problem down to, and which of the usual explanations for it nobody can actually check.
Distribution · Retention
12 September 2026 · 4 min read
How to read a Shorts retention curve
What the shapes in a Shorts audience-retention graph mean, why it can go above 100 percent, and which conclusions the curve does not support.
Retention · Analytics