Grow Creator Field Notes

Shorts: high stay rate but no views? A 5-gate check

High stay rate but no views on YouTube Shorts means distribution stalled, not a bad edit. Five gates to check against your own YouTube Studio numbers.

A high stay rate with low views points at distribution, not at your edit. Stay rate is measured only among the people the feed already served your Short to, so it is a percentage of a group whose size it never reveals. Score well with a hundred viewers and you still have a hundred viewers.

That is the answer. The rest of this page is the check: five gates that stop distribution, each with a test you can run against numbers already sitting in your YouTube Studio. Budget twenty minutes. The Instagram Reels version is near the bottom.

Find your quadrant first

Two numbers, four outcomes. Work out which box you are in before you change anything.

Only one of those four boxes is a distribution problem. One is not a problem at all. The remaining two are content problems, and a content problem is never fixed in your upload settings.

Stay rate is not retention

Creators use the two words interchangeably, and they measure different things. YouTube's own content performance definitions keep them apart:

• Stayed to watch is the percentage of times viewers stayed past the opening seconds of a Short.
• Engaged views counts how many times viewers stayed past those opening seconds, loops excluded.
• Average percentage viewed is the average share of the video watched among the people who stayed. For Shorts, YouTube calculates it from engaged views and their watch time.

Read the denominators rather than the numbers. Stay rate is a share of the people the feed already served. Average percentage viewed is a share of the people who stayed, which is a smaller group again. Neither one can tell you how many people were offered the Short, because neither one counts them. Views is the only figure in that set that moves when distribution moves.

The same stay rate can sit on top of a few hundred views or a few hundred thousand, and the percentage gives no hint which. A strong stay rate next to a flat view count is a small sample scoring well.

YouTube publishes what these metrics mean. It does not publish how many people a new Short goes out to at first, or what makes the system widen that group. Any page quoting an exact seed number is guessing, however confidently it is written. The working model most creators run on is that a Short starts with a limited audience and travels further when that first group responds well. That model fits what people see in Studio, but it is a model, not documented mechanics.

Some Studio versions label this card differently. YouTube introduced the Shorts split as viewed vs swiped away, which is the same measurement read from the other end.

Where these numbers sit in YouTube Studio

Do this on desktop. The mobile Studio app hides several of these fields.

1. Open YouTube Studio, go to Content, then the Shorts tab.
2. Click the Short you are worried about, then Analytics.
3. Overview holds views plus the Shorts card that splits viewers who stayed from viewers who swiped away.
4. Engagement holds average view duration, average percentage viewed, and the retention curve.
5. Audience holds the new versus returning split and the share of views from subscribers.

Studio renames and moves cards more often than it announces, so hunt by metric name rather than by position on the page. Write down five things for this Short and for a typical Short of yours from the last three months: views, stayed to watch, engaged views, average percentage viewed, and non-subscriber share of views. For the wider tour of what each field is good for, see our walkthrough on how to read your Shorts analytics.

Your numbers against your own baseline

We are not going to hand you an industry benchmark. Published Shorts benchmarks are mostly repackaged guesses, and your own channel's normal is the only comparison that carries real information. Compare against yourself.

One caveat on that last row. A Short that flattens quickly is not proof of suppression, and YouTube surfaces no indicator that says a video was down-ranked. All the shape tells you is whether the push is over. If it is over and the stay rate was strong, gates 2 to 5 are where to look. For a longer take on what retention figures do and do not tell you, see our Shorts retention rate benchmarks.

The five gates

Gate 1: your channel is still on a small starting audience

Test: open Content, Shorts tab, and look at your last 10 to 20 uploads. If nearly all of them land in the same narrow band no matter how the stay rate varies, that band is where your channel currently lands. It is a fact about the channel, not a verdict on one video. This is the ordinary case on new and small channels, and the least interesting, because nothing is broken. What to expect from your first 100 Shorts covers the timeline.

Gate 2: Made for Kids is switched on

YouTube's guidance on determining if your content is made for kids states that videos you set as made for kids are more likely to be recommended alongside other kids' videos. The setting also greys out several features on the video, which you can see for yourself in Studio.

Test: open the video in Studio, then Details, then Audience. Then check Settings, Channel, Advanced settings for a channel-wide default, because a default set once applies to everything you upload afterwards and people forget they set it. This is a legal declaration about your content, not a growth lever. If your videos really are made for kids, leave the setting alone.

Gate 3: the subject sits in restricted territory

Two different things get confused here.

Ads: a yellow monetisation icon in Studio points at the advertiser-friendly content guidelines. That icon is about which ads can run, not about how far the video travels.

Age restriction: this one cuts reach directly, because an age-restricted video is not viewable to signed-out viewers or viewers under 18. Part of the possible audience is removed before the feed starts.

Test: in Studio, Content, read the Restrictions column. Anything other than None deserves a click. For subject sensitivity itself there is no flag to read, so the only honest test is a controlled one: post the same format on a plainly uncontroversial subject, then compare distribution.

Gate 4: re-upload, watermark, or reused footage

YouTube does not publish a Shorts re-upload detection policy, so treat this as what creators consistently report rather than as confirmed mechanics: a Short carrying another app's watermark, or the same file posted a second time, tends to travel less than the original did. Instagram documents the equivalent rule openly in its tips for improving your reach, telling creators that content with a visible watermark gets less promotion and that Reels already on Instagram are unlikely to be recommended. The direction is at least documented on one of the two platforms.

Test: check the Restrictions column for a copyright claim, since a claim can block the video in some territories and that shows up there. Then re-export from your original project with no third-party watermark and post it as a new upload rather than editing the existing one.

Gate 5: the Short is being served to the wrong people

Test: video analytics, Audience tab, two fields. The subscriber share of views tells you whether the Short ever left your own audience. Other videos your audience watched gives you a rough read on the neighbourhood your Short landed in. A high stay rate with a heavy subscriber share describes a Short your existing viewers enjoyed and the feed never carried further.

When a channel's uploads jump between unrelated subjects, the system has less to go on when deciding who to try next. That is a hypothesis you can test rather than a published rule: post three Shorts on one narrow subject in a row, then compare their distribution against your scattered uploads.

If tabbing through five Studio screens is more than you want to do, our free YouTube channel audit runs without a signup and reads your recent uploads as a set rather than one at a time, which is the level this problem lives at.

The same check on Instagram Reels

Instagram Insights has no stayed to watch field, so the diagnosis uses different numbers.

• Views and accounts reached tell you how far the Reel travelled. Views counts plays, accounts reached counts unique accounts.
• Average watch time measured against the length of the Reel is the nearest available stand-in for a hold rate.
• The follower and non-follower split beneath views matters most here, because non-follower views are the distribution number.

The Reels version of this exact problem looks like strong average watch time for the length with non-follower views close to nothing. That is distribution, not editing. Instagram's reach guidance, linked above, lists the flags most often behind it: recycled posts, visible watermarks, and accounts that mainly reshare other people's content. Our fuller diagnostic on Reels not getting views works through the rest of them.

What to change, in what order

1. Change nothing about the edit yet. If your stay rate already sits at or above your own median, more cutting and more zooms buy nothing on that axis.
2. Clear the binary flags: Made for Kids, age restriction, copyright claim, third-party watermark. Each is either on or off, and each takes minutes.
3. Move the subject, keep the format. Same hook style, same pacing, different topic, three uploads in a row inside one narrow lane.
4. Judge the result over ten uploads. One Short cannot separate a channel ceiling from a single miss, because one Short is one sample.

Keep the log while you do it: date, subject, views, stayed to watch, average percentage viewed, non-subscriber share. Ten rows in, one of two pictures is visible. Either a single row is an outlier, which means one video to redo, or every row sits under the same ceiling, which means no amount of re-editing that one Short will move anything.

Canonical: https://growcreator.pro/blog/high-stay-rate-but-no-views-youtube-shorts

Leumos Private Limited · hello@growcreator.pro