Published 27 October 2026 · Figures measured 6 October 2026 · Every figure carries the number of Shorts behind it
The question, as creators actually ask it
"Over 50% of my views are coming from the 'Browse Feature' instead of the Shorts Feed. Is this a good sign or a bad sign? ... 175k Views with 86k coming from the Browse Feature."
"my latest short has gotten 70% views from browse feature, and 20% from ..."
Nobody publishes a baseline to answer that against. So here is ours.
The breakdown, for 409 Shorts
Two columns, and the gap between them is the actual finding, so it is worth thirty seconds.
Median means the middle Short. Half were above it, half below. It describes the typical Short. Mean is the plain average, which a handful of unusual Shorts can drag a long way.
| Where the views came from | Typical Short | Average |
|---|---|---|
| The Shorts feed | 70.00% | 52.76% |
| YouTube search | 0.70% | 8.77% |
| Subscribers | 0.00% | 1.97% |
Three things fall out of that, and the third is the one worth your attention.
1. The feed is the whole game
Seventy percent of the typical Short's views came from the Shorts feed. Nothing else is close.
And it skews further at the top end. In our data, every Short that passed 4,000 views got between 93.6% and 98.6% of its views from the feed.
Read that as composition, not as a target. It is tempting to conclude that browse traffic is a bad sign and feed traffic is a good one. Our data cannot support that, and the reason is that feed share and view count climb together: "high performers are feed-heavy" may simply be another way of saying "the feed is where the volume is". We can tell you what our high performers looked like. We cannot tell you that looking like them causes anything.
2. Your subscribers are not watching, and that is normal
The typical Short got 0.00% of its views from subscribers. Not a rounding artefact. Zero is the middle value.
This surprises people who moved over from long-form, where the subscriber feed does real work. On Shorts it does essentially nothing for the typical video, and that holds across every size of channel we looked at.
So a Short underperforming is not your audience abandoning you. They were very likely never shown it.
3. Search looks bigger than it is, and this is the interesting one
Look at the search row again. The typical Short got 0.70% of its views from search. The average Short got 8.77%.
That is more than a ten-fold gap between the middle and the average, and it only happens one way: a small number of Shorts get a lot of search traffic, and almost everything else gets virtually none.
It matters because the average is the number that tends to get quoted. "Shorts get nearly 9% of their views from search" would be a true statement about the average and a badly misleading one about your Short. For the typical Short, search is under one percent.
We found the same thing a year earlier on a smaller base: a median search share of 1.0% across 257 Shorts. Two measurements, two sizes of catalogue, same answer.
We are deliberately not telling you what to do about this. A page that reports a low search share is one sentence away from advice about what you write in your title or description, in either direction. We tested that whole family of questions on our own data and found nothing, so we are not going to make a recommendation we cannot support, and we would treat anyone else's with caution too. This page tells you where the views came from and stops.
What this means if you are staring at your own breakdown
Mostly it means you now have something to compare against, which you did not before.
- Feed-dominated is the ordinary case. 70% for the typical Short in our data.
- A browse-heavy mix is not a known problem. It is different from what our high performers looked like, and that is as far as our data goes. One creator in the quotes above was asking whether a browse-heavy mix was bad while reporting 175,000 views, which is a useful reminder that the mix is not the outcome.
- Zero from subscribers is normal. Not a signal about you.
- Low search is normal. If search is a real share of your views, you are the exception, not behind.
The limits, before you use any of this
These are small and mid-sized channels. 409 Shorts from creators who connected a channel to us. Nothing here describes a channel that routinely reaches six figures, and the mix may genuinely differ there.
This is composition, not causation, everywhere on the page. We did not run an experiment. We looked at where views arrived from and reported it. No sentence here says that changing your mix would change your outcome, because nothing we measured could establish that.
Traffic sources are YouTube's own categories and we report them as given. Where a Short's reported sources do not sum tidily, that is in their data and we have not adjusted it.
The one thing we genuinely cannot see
Knowing a Short got 70% of its views from the feed does not tell you how many people the feed showed it to and did not count, because they swiped past. That share, viewed versus swiped away, is visible to you in YouTube Studio and is not in the data a creator can authorise us to read.
So we can tell you where the views that happened came from. We explain what happened to the people who started. Studio tells you how many never did.
Questions creators ask about this
Taken from public creator forums in October 2026, close to the wording they were asked in.
Most of my Shorts views are from Browse Features instead of the Shorts feed. Is that bad?
It is different from what our best-performing Shorts looked like, and that is as far as we can honestly take it. Every Short above 4,000 views in our base drew 93.6% to 98.6% of its views from the feed (n = 409 Shorts, measured 6 October 2026).
But feed share and view count rise together, so that may just be describing where volume lives rather than identifying a cause. We would not treat a browse-heavy mix as a diagnosis, and the creator who asked this version of the question was reporting 175,000 views at the time.
Why do my Shorts get no views from my subscribers?
Because that is how Shorts behaves for almost everyone. The median subscriber share across our 409 Shorts was 0.00%.
It is not a sign your audience lost interest. The typical Short is distributed to people who do not follow you, and your subscriber count is close to irrelevant to how it travels.
Do Shorts get views from YouTube search?
Hardly any, for the typical Short. Median search share was 0.70% across 409 Shorts, and 1.0% on an earlier measurement of 257.
The average is much higher at 8.77%, but that is a handful of Shorts pulling it up. If you see a figure near 9% quoted as typical, it is an average being passed off as a middle.
Should I change anything based on this?
Nothing on this page is a recommendation. It is a baseline so you can tell whether your own breakdown is unusual, and in most cases the answer will be that it is not.
If your Shorts are being watched and still going nowhere, the more useful question is whether retention predicts which of your Shorts wins, which we have measured and which has a clearer answer.
What is the difference between Browse Features and the Shorts feed?
They are two separate categories in YouTube's own traffic-source reporting and should not be added together.
Shorts feed is the vertical swiping experience: someone was moving through Shorts and yours came up. Browse Features covers the surfaces where a viewer was choosing what to watch, such as Home and Subscriptions. The practical difference is the posture of the viewer. One was being served the next thing; the other picked something.
In our base the feed dominates for the typical Short at 70.00% (n = 409, measured 6 October 2026), so a mix weighted the other way is unusual relative to ours. Unusual is not the same as bad, and we have nothing that would let us call it bad.
My traffic sources look completely different from one Short to the next. Is something wrong?
That variation is normal in our data, and it is the reason this page reports a median rather than a single headline figure. The gap between the median and the mean on every row of the table is the variation showing up: search is 0.70% for the middle Short and 8.77% on average, which can only happen if Shorts differ enormously from each other.
So one Short arriving mostly from one source and the next from another is the catalogue behaving the way ours does, not a signal about either Short.
Does a Short's traffic mix change as it gets older?
We have not measured it, and we would rather say so than guess. The figures here are a snapshot of where each Short's views had come from at the time we read it, not a trace of how that mix moved.
What we have measured about time is how long a small Short keeps accruing views at all, which is a different question and has a clearer answer.
Do more subscribers mean more Shorts views?
Not through the subscriber feed, on our data. The median subscriber share across 409 Shorts was 0.00%, and that holds across the sizes of channel we see.
We have not measured whether a larger subscriber count helps indirectly, so we will not claim it does or does not. What the number does rule out is the simple version: the typical Short is not being carried by the people who already follow you.
Two neighbouring questions this page does not answer. Why a Short with better retention still lost is about outcomes, not sources. How long a Short keeps accruing views is about time.
How this was measured
Why we looked at this at all
We were testing whether anything about how a Short is described predicts how it performs. It did not, on every measure we tried. The traffic-source breakdown came out of that work, and it turned out to be the more useful result: not a lever, but a baseline that nobody publishes with a sample size attached.