The One Metric That Reports on People Who Left
Nearly every analytics number you see describes people who stayed. Views, watch time, retention, engagement: all of them are measured on an audience that already decided to give you a moment. Viewed vs swiped away is the exception, and that is what makes it worth understanding properly.
YouTube's own description of the metric is short. It "Highlights the percentage of times that viewers viewed your Shorts versus swiped away." No click is involved. The Short is already on screen and already playing, and the viewer is deciding, in something close to a reflex, whether this is the thing they came for.
Because the decision happens that early, this metric isolates your opening more cleanly than anything else available to you. Retention tells you where people left across the whole video. This tells you how many never gave it a chance at all, which is a different problem with different fixes.
What YouTube Confirms About Its Role
The metric is not decorative. In its guidance on how Shorts are discovered, YouTube states that "Our systems use the signals for % of viewers who chose to view, avg. view duration and avg. % viewed to inform ranking." Likes and post-watch survey responses are named alongside them.
Read that list carefully, because the order describes a funnel. First the share of people who chose to watch at all. Then how long they stayed. Then how much of the video that represents. Each one gates the next, and a video that fails the first never gets assessed on the others.
It also tells you something reassuring about what is not on the list. Posting frequency is not a ranking signal in that sentence, and YouTube addresses it directly elsewhere by stating that there is no minimum posting cadence required for videos to perform well. Volume is a strategy, not a requirement.
What Nobody Can Tell You, Including Us
Search for a good viewed vs swiped away rate and you will find confident numbers. Seventy percent. Eighty. A threshold at which YouTube supposedly starts pushing a Short. None of them come from YouTube.
YouTube defines the metric and publishes no figure for what a good one is. There is no documented level at which distribution changes, and stating one would be inventing a fact about somebody else's system. That absence is worth saying out loud, because the alternative is repeating a number whose only source is that other people repeated it first.
What can honestly be offered is a target: a level worth aiming at, chosen for how well it separates openings that work from openings that do not. That is a recommendation rather than a measurement, and the difference is not pedantry. A measurement can be proven wrong by data. A target is advice, and it stands or falls on whether following it improves your videos.
Retensis Targets for Chose to View
These are the bands we hold a Short to. They are deliberately demanding at the top, because the purpose of a target is to be worth reaching rather than to be comfortable.
Where you sit matters less than which direction you are moving. A Short at 55% that used to be at 45% has a working opening and a creator who is learning something. A Short at 72% on a channel that normally sits at 80% is worth a second look even though the number reads as strong in isolation.
| Range | Rating | What it means |
|---|---|---|
| Above 80% | Exceptional | The opening frame is doing the work of a thumbnail and winning |
| 70–80% | Strong | Most people who were shown it decided to stay for it |
| 60–70% | Healthy | A workable opening for most kinds of Short |
| 50–60% | Needs work | The first frame is losing people before the idea arrives |
| Below 50% | Problem | More viewers are swiping past than choosing to watch |
What Actually Moves It
The opening frame carries more weight than any other single element, because it is doing the job a thumbnail does on long-form video with none of the advantages. It has no title beside it, no channel name in a position anyone reads, and roughly a second before the decision is made. A frame that shows a person about to speak communicates far less than a frame that shows the thing the video is about.
The opening line is the second lever, and the most common mistake is spending it on context. Introductions, greetings, and explanations of what you are about to explain all consume the exact window in which the viewer is deciding whether there is anything here. Starting at the interesting part and backfilling the context afterward almost always reads better.
The third is differentiation, and it is the one creators overlook. Your Short is not being judged against your last Short. It is being judged against whatever the viewer would have swiped to instead. An opening that looks like a category of video the viewer has already seen a dozen times today will lose to one that does not, regardless of how good the rest is.
This is where the metric stops being able to help you. It tells you the opening lost people; it cannot tell you which of those three things went wrong. Retensis reads the opening frame, the first spoken line, and the pacing of the first few seconds, and reports which one is costing you the decision. That is the difference between knowing your hook is weak and knowing why.
Frequently asked questions
It measures the decision a viewer makes in the first moment your Short appears, before they have seen anything except the opening frame and heard anything except the opening sound. YouTube describes the metric as the percentage of times viewers watched rather than swiped past. It is the only metric in Shorts analytics that reports on people who did not watch, which makes it the closest thing you have to a measurement of your opening in isolation.
No. YouTube documents the metric and explains what it represents, but it publishes no figure for what a good one looks like, and it does not describe a level at which distribution changes. Every specific threshold circulating for this metric was invented by somebody outside YouTube. The bands on this page are Retensis targets, which is a different kind of claim: they are what we hold a Short to when we analyze it, not a description of what YouTube requires.
Easily, and this is the most common shape of the problem. A strong opening frame can win the decision to stay while the rest of the video gives people nothing to stay for, which produces a high chose-to-view rate sitting next to a collapsing retention curve. YouTube names the share who chose to view, average view duration and average percentage viewed as separate ranking signals, so winning one of the three while losing the other two is a real and unrewarding outcome.
Treat the first frame as a thumbnail that is already playing. It has to communicate a subject and a reason within about a second, without a title to help it, and with sound that may or may not be on. The most common fixable causes are a slow establishing shot, an opening line that sets up context nobody asked for, and a first frame that is visually indistinguishable from the hundred Shorts before it.
Get deeper insight into your YouTube Shorts
Analyze your Shorts before you publish: hook score, predicted retention, and specific fixes. Find out what your YouTube Studio data can't tell you.
Analyze your Shorts free →