youtube shorts · shorts algorithm · shorts views · creator growth
YouTube Shorts Algorithm 2026: What Actually Matters
How the YouTube Shorts algorithm works in 2026, why Shorts get no views, and which signals matter — every claim cited to YouTube's own docs.
· Everpop
YouTube Shorts distribution is driven by viewer satisfaction, not follower count. YouTube shows your Short in people's feeds and measures whether they watch or swipe away, expanding reach when the response is strong. Since 31 March 2025, a view counts every time a Short starts to play or replay, so raw view numbers rose while "engaged views" now measures who stayed.
How does the YouTube Shorts algorithm actually work?
YouTube's recommendation system doesn't run on a fixed formula. It learns "from over 80 billion pieces of information we call signals" and uses them to predict what each viewer will find satisfying, per YouTube's own explainer. The signals it names are "clicks, watchtime, survey responses, sharing, likes, and dislikes." There is no follower threshold and no posting quota in that list — the feed is a per-person prediction of what is worth watching next.
Watchtime carries real weight. YouTube's own example: "if our tennis fan watched 20 minutes of Wimbledon highlight clips, and only a few seconds of match analysis video, we can safely assume they found watching those highlights more valuable." For Shorts, that same logic plays out in seconds instead of minutes.
Picture two Shorts shown to the same number of feeds. The one people keep watching earns wider distribution; the one people swipe past stalls — no matter who posted it or how many subscribers they have.
Why do my Shorts get no views?
Because the Shorts feed is a swipe test, and most Shorts lose it in the opening moment. YouTube hands you the exact metric in Studio: "Viewed (vs. swiped away)," which it defines as "the percentage of times that viewers viewed your Shorts versus swiped away." Alongside it is "Shown in feed" — "the number of times that your Shorts showed in the Shorts Feed."
Read those two together and the diagnosis gets honest:
- If "Shown in feed" is climbing but "Viewed (vs. swiped away)" is low, YouTube is testing your Short and viewers are swiping. That is a hook problem, not a reach problem.
- If "Shown in feed" is near zero, distribution never really started. That points upstream — a still-processing upload, a "Made for Kids" setting that removes feed placement, or a Short that is not eligible — rather than at your content.
That distinction saves weeks. A creator convinced their content is "shadowbanned" often just has a weak first second and a healthy "Shown in feed." A creator stuck at literally zero usually has an eligibility or settings issue. If yours flopped even with a strong open, why did my YouTube Short flop walks the full diagnostic.
Did YouTube really change what counts as a view?
Yes, and it matters more than the headline suggests. Since 31 March 2025, "views will count the number of times a Short starts to play or replay, with no minimum watch time requirement." In plain terms: a loop, a replay, or a split-second scroll-by now counts. That is why many channels watched raw view numbers jump without changing anything.
YouTube kept the old, stricter metric and renamed it: "We will keep the existing Shorts view metric, now called 'Engaged views,' in YouTube Analytics so you can see how many viewers chose to continue watching your video." Engaged views are the honest number for comparing one of your Shorts to another.
The part creators miss: this does not change your money. YouTube states the change "will not directly impact the YouTube Partner Program (YPP) eligibility and Shorts ad revenue sharing. They will both be based on 'Engaged views.' So creators' eligibility and earnings will not change." A Short with a huge view count and few engaged views is not the win it looks like. Because an inflated number can flatter a flop, a durable record of engaged views beats a screenshot — which is exactly why screenshots of view counts lie.
What signals actually matter, and which are myths?
The signals YouTube names are "clicks, watchtime, survey responses, sharing, likes, and dislikes." A few honest reads:
- Watchtime and satisfaction lead. YouTube runs viewer surveys and trained a model to "predict potential survey responses for everyone." Holding attention beats any single tap.
- Likes and shares are signals, not switches. They appear on the list, but no YouTube documentation says a like quota triggers distribution.
- Views are not engaged views. After March 2025, chasing the raw view count optimises for the wrong number.
- What YouTube never lists: hashtags, follower count, posting time, or "beating the algorithm." No official source ties Shorts feed distribution to any of them. Treat growth-hack claims about them as unproven until a platform source says otherwise.
What should I check before posting my next Short?
| Check | Where in YouTube Studio | What good looks like |
|---|---|---|
| Hook holds the opening moment | Viewed (vs. swiped away) | Watch rate rising, swipes falling |
| Distribution actually started | Shown in feed | Number climbing, not stuck near zero |
| Real retention, not inflated views | Analytics → Engaged views | Engaged views tracking with total views |
| Not mislabelled "Made for Kids" | Video settings | Audience set correctly |
| Clean file, no recycled watermark | Your upload | Native export, no other platform's logo |
One practical note on that last row: if you cut Shorts from a longer video, keep a clean master and export without another platform's logo baked in. Everpop is built around that habit — drag-and-drop or a shared Google Drive Finals folder in, review-first approval before anything posts, and signed 48h and 7d Analytics receipts that record engaged views and watch time even when a clip flops. The takeaway stands even if you never open it: keep the source file, and judge Shorts on engaged views rather than the headline number.
Claims table
| Claim | Source |
|---|---|
| A view counts each time a Short "starts to play or replay, with no minimum watch time requirement" (since 31 March 2025). | https://support.google.com/youtube/answer/10059070?hl=en |
| YouTube kept the old metric as "Engaged views," showing "how many viewers chose to continue watching your video." | https://support.google.com/youtube/answer/10059070?hl=en |
| The change "will not directly impact the YouTube Partner Program (YPP) eligibility and Shorts ad revenue sharing... based on 'Engaged views.'" | https://support.google.com/youtube/answer/10059070?hl=en |
| "YouTube has changed how it counts organic views for YouTube Shorts"; Shorts average view duration is "calculated from engaged views." | https://support.google.com/youtube/answer/9313698?hl=en |
| The recommendation system learns "from over 80 billion pieces of information we call signals": "clicks, watchtime, survey responses, sharing, likes, and dislikes." | https://blog.youtube/inside-youtube/on-youtubes-recommendation-system/ |
| Watchtime example: a fan watching "20 minutes of Wimbledon highlight clips, and only a few seconds of match analysis video" signals what they value. | https://blog.youtube/inside-youtube/on-youtubes-recommendation-system/ |
| "Viewed (vs. swiped away)" = "the percentage of times that viewers viewed your Shorts versus swiped away"; "Shown in feed" = "the number of times that your Shorts showed in the Shorts Feed." | https://support.google.com/youtube/answer/12942217?hl=en-GB&co=YOUTUBE._YTVideoType%3Dshorts |
Frequently asked questions
- Do loops and replays count as views on Shorts?
- Yes. Since 31 March 2025, YouTube counts a view every time a Short "starts to play or replay, with no minimum watch time requirement." A loop or a quick scroll-by now registers, which is why many channels saw raw view numbers rise without changing anything.
- What is the difference between views and engaged views?
- Views count every play or replay. "Engaged views" is YouTube's renamed older metric, showing "how many viewers chose to continue watching your video." It is the better number for comparing one of your Shorts against another.
- Did the March 2025 change increase Shorts earnings?
- No. YouTube says the change "will not directly impact the YouTube Partner Program (YPP) eligibility and Shorts ad revenue sharing," because both are based on engaged views. A bigger raw view count alone does not pay more.
- Why is my Short stuck at zero views?
- Zero usually means distribution never started, not that the content is bad. Check "Shown in feed" in YouTube Studio; if it is near zero, look at a still-processing upload, a "Made for Kids" label that removes feed placement, or eligibility, rather than assuming a shadowban.
- How many subscribers do I need for Shorts to get views?
- None, in YouTube's stated system. Recommendations are a per-viewer prediction built from signals like watchtime and survey satisfaction; follower count is not on YouTube's published signal list for the feed.
- Which single metric should I watch?
- "Viewed (vs. swiped away)" — your watch-versus-swipe rate — paired with engaged views. Together they tell you whether your hook holds and whether people actually stayed.