If you are asking how long should youtube shorts be, the strongest answer from this May 2026 sample is still: start shorter than you think. In the final duration comparison set, 0-15 second Shorts were the only bucket where more than 60% of sampled videos had already crossed 10,000 views inside the 14-day collection window. The weakest bucket was 41-50 seconds.
That does not mean every winning Short has to be under 15 seconds. The 51-60 second bucket still produced real winners, especially when the idea needed explanation or a fuller payoff. But the May pull was clear: the default edge leaned short, and longer Shorts had to earn every extra second.
What the Data Says
May 2026 Findings
- Best default bucket: 0-15 seconds
- Weakest bucket: 41-50 seconds
- Best longer bucket: 51-60 seconds
- Best small-channel default: 0-15 seconds
- Niche pattern: 0-15 seconds led all seven faceless-friendly lanes
- Most important caveat: YouTube supports Shorts up to 3 minutes, but this analysis isolates the sub-60-second format on purpose
Dataset Snapshot
Live pull date
May 9, 2026 UTC
recent public Shorts published in the prior 14 days
Search calls completed
56 calls
across seven faceless-friendly niches and four query variants each
Estimated quota used
5,686 units
including enrichment calls
Broad faceless-style sample
1,915 Shorts
after duration, title, and format filtering
Duration comparison sample
1,915 Shorts
final comparison set used for the ranking table
Data freshness: This article was updated with a live public-data pull completed on May 9, 2026 UTC using the YouTube Data API. YouTube classifies eligible vertical videos up to three minutes as Shorts, but this post focuses on the 0-60 second range because that is still where this faceless-style sample had the cleanest like-for-like comparison. See Understand three-minute YouTube Shorts.
How Long Should YouTube Shorts Be: The Ranking
Here is the main duration table first.
| Rank | Shorts length | Share hitting 10K | Median views | Small-channel win rate | Read |
|---|---|---|---|---|---|
| 1 | 0-15 seconds | 66.0% | 28,351 | 53.6% | Best default by a wide margin |
| 2 | 16-20 seconds | 48.3% | 8,959 | 34.5% | Strong short range, but behind the leader |
| 3 | 51-60 seconds | 40.7% | 2,825 | 25.8% | Best longer bucket when the idea needs room |
| 4 | 21-30 seconds | 34.8% | 1,437 | 21.6% | Workable, but no longer a standout |
| 5 | 31-40 seconds | 34.1% | 1,202 | 24.5% | More room did not translate into stronger median results |
| 6 | 41-50 seconds | 25.6% | 868 | 19.0% | Weakest range in this sample |
This is the main answer. If you need one rule instead of six, make your first test in the 0-15 second range, then stretch longer only when the viewer needs more setup to understand the payoff.
Share Hitting 10K by Shorts Length
0-15 seconds
best broad default
66.0
16-20 seconds
useful short range
48.3
21-30 seconds
workable but weaker
34.8
31-40 seconds
no clear edge in this pull
34.1
41-50 seconds
weakest bucket
25.6
51-60 seconds
best longer bucket
40.7
0-15 Seconds Was The Clear Default
Verdict: If you do not know where to start, start very short.
The 0-15 second bucket separated from the rest on every practical metric that mattered in the main comparison set:
- 66.0% crossed 10,000 views
- median views: 28,351
- 53.6% of the whole bucket were small-channel wins
- 81.3% of winners came from channels under 100,000 subscribers
That last number is important. A bucket can look strong if it is dominated by large channels. Here, the shortest bucket also looked accessible. Most winners in the 0-15 second range came from channels under 100,000 subscribers.
The production lesson is simple: a lot of Shorts do not need more explanation. They need a faster opening decision. If the value, surprise, or emotional turn is obvious right away, the shorter cut has a better chance of getting to the payoff before the viewer leaves.
41-50 Seconds Was The Weakest Bucket Again
Verdict: This range still looked like the awkward middle.
The 41-50 second bucket was the weakest of the six:
- 25.6% crossed 10,000 views
- median views: 868
- small-channel win rate: 19.0%
That does not make 45-second Shorts impossible. It does suggest that this range had the hardest time justifying its runtime in the May sample.
Very short Shorts had a cleaner hook-and-payoff pattern. Full-minute Shorts sometimes had enough room for a utility tip, story, or explanation. The 41-50 second group landed between those two strengths.
If you are batching Shorts, do not treat 45 seconds as a safe compromise. Either make the idea much tighter, or give it the full minute and use the extra time for a clear reason.
51-60 Seconds Was The Best Longer Bucket
Verdict: Longer Shorts still had a place, but they were not the best default.
The strongest longer range in this pull was 51-60 seconds:
- 40.7% crossed 10,000 views
- median views: 2,825
- 63.5% of winners came from channels under 100,000 subscribers
That is meaningfully weaker than 0-15 seconds, but stronger than the 21-30, 31-40, and 41-50 second buckets.
Use the full minute when the idea needs it:
- a quick tool demo
- a before-and-after explanation
- a story with one turn
- a list where each item needs context
- a claim that needs proof before the payoff
This is where creators often ask the runtime question backward. The better question is not "Can this be 60 seconds?" It is: does the extra time make the Short clearer, more useful, or more satisfying?
The Best Length Was Consistent By Niche
Verdict: In this May pull, every niche leaned shortest.
The April version of this article had a clearer niche exception. AI Tips leaned longer then. In the May refresh, the shortest bucket led all seven lanes.
| Niche | Best bucket in this sample | Share hitting 10K | Videos in best bucket | Read |
|---|---|---|---|---|
| Texting Stories | 0-15 seconds | 91.7% | 60 | Fast conflict setup won clearly |
| Facts and Trivia | 0-15 seconds | 88.9% | 54 | Immediate surprise worked best |
| History Facts | 0-15 seconds | 75.9% | 87 | Quick shock-and-reveal packaging led |
| Motivation | 0-15 seconds | 61.1% | 167 | Fast emotional payoff still fit the lane |
| Psychology Facts | 0-15 seconds | 60.6% | 66 | Short hooks beat slower framing |
| AI Tips | 0-15 seconds | 52.0% | 50 | Even utility clips leaned shorter in this pull |
| Finance Facts | 0-15 seconds | 39.4% | 66 | The shortest bucket led, but the lane was tougher overall |
These niche splits are directional, not permanent laws. Public search results can move quickly, and some query families are noisier than others.
Still, the May pattern is useful: when the idea can be understood fast, the shortest cut usually deserves the first test.
What I Would Actually Test First
If I were publishing faceless Shorts from scratch right now, I would use a simple runtime testing rule:
- Start with 0-15 seconds as the default.
- Use 16-20 seconds when the hook needs one extra beat.
- Save 51-60 seconds for ideas that need explanation, proof, or a fuller payoff.
- Be skeptical of 41-50 seconds unless the script has a very clear reason to land there.
That is also the cleaner production setup if you are testing multiple Shorts angles from one repeatable system. If you want to see what that workflow looks like in practice, Overview walks through how EchoEtch turns a script into a finished video pipeline.
What This Does Not Mean
This post is about a sampled public dataset, not a platform-wide law.
It also does not mean YouTube rewards one magic duration. YouTube's own Shorts guidance points creators back toward performance analysis in Studio rather than a single recommended runtime. See Content tab analytics tips for Shorts.
The safest read is:
- length matters
- runtime still interacts with topic
- hook clarity still matters more than one exact number
- longer Shorts need to earn their extra seconds
That last point is the one most creators should care about. A longer Short can still win. It just has to justify the extra runtime immediately.
Methodology
This article was updated from an April 2026 dataset to a fresh public-data pull completed on May 9, 2026 UTC using the YouTube Data API.
We searched recent public Shorts across seven faceless-friendly lanes:
- Facts and Trivia
- Motivation
- History Facts
- Psychology Facts
- Finance Facts
- AI Tips
- Texting Stories
The collection used:
- 56 `search.list` calls
- about 5,686 estimated quota units
- 2,487 unique video discoveries before enrichment filters
- a broad faceless-style sample of 1,915 Shorts
- a duration comparison sample of 1,915 Shorts for the final ranking table
- 1,763 enriched public channel records
For the main analysis, a video had to be:
- 60 seconds or less
- published within the 14 days before collection, after April 25, 2026 UTC
- public at the time of the API pull
- not obviously podcast, vlog, reaction, livestream, full-episode, music-video, or similar face-driven format noise
For the ranking table, a Short counted as a 10K winner if it had already reached 10,000 views or more at the time of collection during that first 14-day window.
The ranking used three decision variables:
- share of the bucket that crossed 10K
- median views in the bucket
- small-channel win rate, defined here as the share of the bucket made up of Shorts from channels under 100,000 subscribers that had already crossed 10K
This is still a sampled read of public Shorts surfaced through YouTube search, not a platform-wide census. The sample includes real public search noise, multilingual results, uneven niche supply, and channel-size differences. The results should guide testing, not replace channel-level analytics.
If you want to test multiple Shorts lengths without rebuilding the whole production stack each time, create your first faceless Short with EchoEtch.