Best Clip Length for TikTok, Reels, and YouTube Shorts

AutoClip Team6 min read

Updated

Illustration for Best Clip Length for TikTok, Reels, and YouTube Shorts

Length Is a Proxy for the Thing That Actually Matters

Ask ten clippers for the ideal short length and you'll get ten confident, contradictory answers. That's because length isn't the variable the platforms care about. Completion is.

Every major short-form feed rewards the same basic thing: people watching to the end, and ideally looping. Duration matters only because it's the easiest lever you have on completion. A 22-second clip that holds 90% of viewers to the end will outperform a 45-second clip that holds 45%, every time, on every platform.

So the useful question isn't "what length wins" — it's "what's the longest this clip can be while still finishing strong." That reframing changes what you do in the edit. You stop padding to hit a number and you start cutting until nothing is left that a viewer would skip.

One caveat worth stating up front: longer clips, when they hold, are worth more. They accumulate more watch time per view, which is why platforms have all pushed their maximum durations up. A 60-second clip that completes is a better outcome than a 20-second clip that completes. It's just much harder to pull off.

TikTok

The sweet spot for clipped content on TikTok sits around 21–34 seconds. Long enough to land a story or a full exchange, short enough that a loop feels natural rather than like a chore.

Under 15 seconds, you can get very high completion but the clip often reads as thin — fine for a punchline or a single reaction, weak for anything with a setup. Over about 45 seconds, drop-off gets steep unless the content is unusually strong.

TikTok also rewards rewatches more visibly than the other two. If your clip's last line loops cleanly into its first, you get watch time for free. It's worth designing for occasionally, not every time.

The one length rule that isn't negotiable: whatever the duration, the hook has to land in the first second or two. Not the first three seconds — sooner than the folklore suggests. Read why 3-second hooks are overrated for the more nuanced version of that argument.

YouTube Shorts

Shorts tolerates length better than the others. 30–58 seconds is the productive range for clipped content, and Shorts audiences are noticeably more patient with setup — a story that needs fifteen seconds of context can work here and would die on TikTok.

Two practical notes. First, stay under the 60-second line if you want the clip to behave predictably in the Shorts feed rather than sliding toward long-form treatment. Second, Shorts viewers arrive from a platform where they're used to watching things, not scrolling past them, so a clip that ends on a genuine conclusion outperforms one that ends on a cliffhanger.

If most of your distribution is Shorts, it's worth understanding how the Shorts algorithm treats clipper content specifically — reused-content handling is a real factor there in a way it isn't on TikTok.

Instagram Reels

Reels is the least forgiving of the three. 15–30 seconds is where clipped content performs, and drop-off past 30 seconds is sharper than on either competitor.

Reels audiences scroll faster and are more visually driven. That has an editing consequence: clips with strong visual movement — reactions, gameplay, anything with a face doing something — outperform static talking heads more dramatically here than elsewhere. If you only have a talking head, keep it short and lean hard on captions and punch-ins to create movement.

Reels also puts real weight on shares and saves, which favors clips with a single quotable takeaway over clips that are merely funny.

Choosing the Length for a Specific Clip

Forget the ranges for a moment and run the clip through four questions.

Where does the interesting part begin? Cut everything before it. Almost every clip you'll ever make is improved by removing its first two seconds.

Does it need setup? If yes, either compress the setup into a caption overlay or accept the extra ten seconds and put it on Shorts rather than Reels.

Where does the energy peak? End within a few seconds after that. Everything past the peak is decay.

Is there a line that could be the last line? If a clip has a natural closer — a punchline, a verdict, a mic-drop — end there even if that means a strange duration like 27 seconds. Odd durations aren't penalized; padding to a round number is.

Practically, this is easier when the tool proposes the boundaries and you adjust rather than scrubbing a timeline from scratch. In AutoClip, each candidate clip comes with a suggested in and out point plus a virality score you can override in the timeline editor. Around nine clips come back from a typical video in about 10–15 minutes, and the length trimming is where your judgment earns its keep.

Then post the same clip at different lengths across the three platforms and watch completion rate, not views. Two weeks of that data on your own niche beats every general-purpose length guide, including this one.

Frequently Asked Questions

For your best clips, yes — a tighter cut for Reels, a longer one for Shorts. For routine posts it's not worth the time; pick a length in the 25–35 second overlap and post the same file everywhere. Vary the caption per platform instead, which matters more and costs less.

Not directly. Long clips hurt reach when they lose viewers, which correlates with length but isn't caused by it. If your 55-second clips hold completion, keep making 55-second clips — they earn more watch time per view than short ones.

Roughly seven seconds. Below that most feeds struggle to serve it properly and viewers barely register it. Sub-10-second clips can spike on loops but rarely build a channel.

Post the same source clip at two lengths on the same platform a few days apart, and compare completion rate rather than views. Do that a dozen times and you'll have a length rule for your niche that's worth more than any published benchmark. [Clip length vs completion rate data](/blog/clip-length-vs-completion-rate-data) covers how to set that test up properly.

Stop guessing where a clip should start and end

AutoClip proposes in and out points for around nine clips per video, scores each one, and hands you a timeline editor to adjust before you post.

Get started for free