How Long Should a YouTube Video Be? The Honest Answer

You've got a video planned and one question you can't settle: how long should it be? Search for the answer and every page gives you a confident number. Eight minutes. Ten to fifteen. Under seven for retention, over twenty for watch time. They can't all be right, and almost none of them cite YouTube, because YouTube has never published an ideal length.

What YouTube has published is more useful. It has documented what it ranks on, the handful of length rules it actually enforces, and a report that shows you exactly where your own viewers stop watching. This post covers all three, then gives you a method for sizing a video to its topic instead of to a number someone else picked.

Is there an ideal length for a YouTube video?

No. YouTube's own page on how search works names three things it weighs: relevance, engagement and quality. Watch time appears, but as an engagement signal, measured as “the watch time of a particular video for a particular query”. Video length isn't on the list at all.

Recommendations work on the same idea, with a twist. In a 2021 post on YouTube's recommendation system, YouTube's VP of Engineering Cristos Goodrow put it plainly: “not all watchtime is equal.” YouTube measures what it calls valued watchtime, “the time spent watching a video that you consider valuable”, through surveys that ask viewers to rate what they just watched from one to five stars.

Put those together and the question changes. A 20-minute video that people abandon at minute four isn't earning 20 minutes of anything. A six-minute video that fully answers the search is doing its job. The goal isn't a length. It's a video where every minute is worth watching.

The length rules YouTube actually sets

There are a few hard limits, and they're worth knowing before you plan anything:

Only one of these touches strategy, and it's the mid-roll line. That is a monetization rule, not a ranking one, and it only matters once you're in the YouTube Partner Program.

Why longer videos seem to win

Search a broad topic and the top results are often long. It's easy to read that as “long videos rank better” and set your target at 15 minutes. That gets the cause backwards.

A search like “how to set up a home server” asks for several things at once: the hardware, the operating system, storage, remote access, backups. A video that covers all of them properly is long because the topic is. The length is a side effect of the coverage. Copy the length without the coverage and you get the downside on its own: more minutes for viewers to leave in.

The same mistake shows up at the 8-minute line. Stretching a six-minute video to 8:01 gets you a mid-roll slot, plus two minutes where the viewer has no reason to stay. If they leave there, it's the first thing the retention report will show you. Cross the line when the topic genuinely needs the time, and not otherwise.

Short isn't automatically safer, either. A video that skips a sub-topic the viewer expected sends them back to search for one that doesn't. The right length is the one that covers what the search asks for, with nothing added and nothing missing.

How to work out how long your video should be

Instead of picking a number and filling it, build the length up from the topic. Five steps:

1. List the sub-topics the search demands.Search the term you're targeting and open the videos already ranking. Their chapters are the fastest way in, which is one more reason your own chapters should work. A sub-topic that appears in almost every top video is table stakes. Questions viewers keep asking in the comments are the ones nobody has answered yet.
2. Give each sub-topic a time budget. Weight it toward what the viewer came for. A setup step people follow along with needs more room than a definition. A sub-topic everyone covers in a sentence probably deserves a sentence from you too.
3. Turn the budget into a word count. At a conversational 140 to 160 words a minute, ten minutes of talking is roughly 1,400 to 1,600 words. The script length calculator does the conversion at slow, conversational or fast pace, in either direction.
4. Add the time words don't measure. Demos, screen recordings, b-roll and deliberate pauses all add runtime no word count predicts. The script gives you a floor, not a finished length.
5. Cut anything that doesn't earn its place.Every section should map to something on your step-one list. If you can't say which sub-topic a section covers, it's padding.

Add it up and that's your length. It probably won't be a round number, and that's fine. Nobody watching a 9:40 video wishes it were ten minutes.

Steps one and two are the slow part: five competing videos, a notepad, and a lot of scrubbing to time each section. That research step is what VidHalo automates. Point it at a search term and it reads the transcripts, chapters and comments of the videos ranking for it, then shows which sub-topics each one covers, how much time each spends on them, and what viewers are still asking for. Your time budget comes from evidence instead of a guess.

Check your length against audience retention

Once a video is live, YouTube tells you whether the length was right. The audience retention report in a video's Studio analytics highlights key moments for any video at least 60 seconds long with at least 100 views:

• Intro: the percentage of viewers still watching after the first 30 seconds.
• Top moments: stretches where almost no one dropped off.
• Spikes: moments viewers rewatched or shared.
• Dips: moments viewers skipped or stopped watching.

Each one tells you something about length. A weak intro usually means the video isn't what the title promised. YouTube's own advice there is to make the title and thumbnail better reflect the content, and to rework the first 30 seconds. A dip in the middle marks a section that ran long or didn't deliver; YouTube suggests reviewing dips to see why interest dropped. Spikes are the sections worth more room next time.

That's the honest endpoint of the length question. The retention curve on your last video is a better guide to your next one than any article, this one included.

How long should a YouTube Short be?

The hard rules are simple. Per YouTube's three-minute Shorts page, a Short is a square or vertical video up to three minutes long. One catch if you use music: a Short over one minute with an active copyright claim is blocked worldwide until the claim is resolved.

Within that range, which length performs best is a different question with its own data. The Shorts length checker benchmarks your runtime against a large study of published Shorts, with the source and its caveats on the page. The same principle applies, though: a Short should be as long as the idea holds attention, and no longer.

A quick check before you shoot

1. You can name every sub-topic the video covers, and each one matches something the search asks for.
2. Each section has a time budget, weighted toward what viewers came for.
3. The script's word count fits that budget at your own pace.
4. If you're close to 8 minutes, the extra time is content, not stretch.
5. After publishing, you'll check the retention report once the video passes 100 views, and adjust the next one.

The length follows the coverage

The real answer to “how long should a YouTube video be” is: as long as it takes to cover what the search is asking for, and no longer. The hard part is knowing what that coverage is for a given search. Guessing leaves you with a padded video or a thin one.

VidHalotakes the guessing out. It shows you what the top-ranking videos for your term cover, how long they spend on each sub-topic, and where they leave gaps, so you can plan a video that's complete without being long for the sake of it. The full manual version of that process is in how to outrank a YouTube video.