How long should a YouTube video be? Length by format, niche and script size
There is no ideal length, but there is a right length for each format. Typical ranges, the 8-minute mid-roll rule, and how to turn a target runtime into a word count.

The short answer to how long a YouTube video should be is: as long as the story needs and no longer. That is true and not much help when you are staring at an empty document. So this guide gives working ranges by format and niche, explains the one hard number YouTube attaches to length, and shows how to turn a target runtime into a script word count before you write.
What YouTube rewards
YouTube does not publish a preferred length, and it has said for years that it has none. What its recommendation system responds to is whether viewers choose a video and then stay with it. Two measures in Studio describe the staying part.
- Average view duration: the minutes a typical viewer watches
- Average percentage viewed: that figure as a share of the video's length
A longer video can collect more minutes per viewer, which helps it. A longer video also gives every viewer more chances to leave, which hurts it. A 20-minute video that people abandon at minute four is worth less than an 8-minute video they finish. The right length is the longest one at which the retention line still holds.
Typical lengths by format and niche
These ranges describe what works for narrated, faceless formats. They are starting points drawn from how each format is structured, not rules.
| Format | Typical length | Why |
|---|---|---|
| News explainer | 8 to 12 minutes | One question, answered while it is still current |
| Archive documentary (history, space) | 10 to 15 minutes | Room for setup, three or four turns and an ending |
| Engineering and "inside the machine" | 10 to 18 minutes | Each system needs explaining before the next makes sense |
| Profile or biography | 12 to 20 minutes | A life has more chapters than an event |
| True crime case file | 15 to 25 minutes, often longer | Viewers expect detail and the audience watches in long sessions |
| Sleep and slow history | 30 minutes to several hours | Played in the background, where length is the product |
| Shorts | 20 to 60 seconds, up to 3 minutes allowed | One beat, and replays count |
Before settling on a number, look at your own niche. Take ten recent videos from small channels that outperformed their subscriber count and note the lengths. If they cluster between 12 and 16 minutes, a 35-minute debut is a gamble and so is a 5-minute one.
Mid-roll ads and the 8-minute line
The one fixed number is eight minutes. Monetized videos of eight minutes or longer can carry mid-roll ads, placed automatically or by you in Studio. Shorter videos are limited to ads before and after. Mid-rolls generally raise what a video earns per view, which is why so many videos run to 8:05.
Two cautions. First, the rule is a threshold and nothing more: a 6-minute story stretched to eight will lose more in retention than it gains in ad slots. Second, placement matters. A mid-roll dropped in the middle of a sentence pushes viewers out. Put breaks at natural pauses between sections, and do not put one in the first couple of minutes while the viewer is still deciding whether to stay.
If you want to see what mid-rolls might mean for income at your view counts, the YouTube money calculator lets you try different assumptions.
Why padding kills retention
Padding is anything the viewer would not miss. It is easy to spot in someone else's video and hard to see in your own.
- Intros that explain what the video is about after the title already did
- Background the viewer did not ask for, placed before the story starts
- The same point made three times in different words
- Recaps of what was said two minutes ago
- Long outros that begin with a signal that the content is over
Each of these shows up as a dip on the retention graph. The damage is larger than the lost seconds, because a viewer who leaves at a padded stretch does not see anything after it either. Lower retention then means fewer recommendations for the whole video.
The test for every paragraph in a script: if it were deleted, would the next paragraph still make sense, and would anyone feel something was missing. If the answer is yes and no, cut it. A tight 11 minutes beats a slack 15. The opening is where padding costs the most, and it gets its own piece in what the first 60 seconds must do.
How to plan script word count
Length is decided in the script, not in the edit. Documentary narration runs at about 140 to 160 words per minute once natural pauses are included. Slower suits sleep content and heavy subjects. Faster suits news and countdowns.
| Target length | Words at 140 per minute | Words at 150 per minute | Words at 160 per minute |
|---|---|---|---|
| 8 minutes | 1,120 | 1,200 | 1,280 |
| 10 minutes | 1,400 | 1,500 | 1,600 |
| 12 minutes | 1,680 | 1,800 | 1,920 |
| 15 minutes | 2,100 | 2,250 | 2,400 |
| 20 minutes | 2,800 | 3,000 | 3,200 |
| 25 minutes | 3,500 | 3,750 | 4,000 |
For any other length, or to check a finished draft, paste it into the words to minutes calculator. Two practical notes on using the numbers.
- Find your real pace first. Record yourself, or your chosen synthetic voice, reading 300 words and time it. Use that figure from then on
- Leave headroom. Title cards, pauses on a strong image and short music breaks add 5 to 10 percent to the runtime, so a script counted for 8:00 exactly will usually land a little over, and one counted for 7:50 may or may not clear the mid-roll line
Budget the words by section
A word count is more useful when it is divided up. For a 12-minute, 1,800-word archive documentary, a workable split looks like this.
- Hook and central question: 150 words, the first minute
- Setup and context: 300 words
- Three or four main turns of the story: 1,050 words between them
- Resolution and what it changed: 250 words
- Close and pointer to the next video: 50 words
If the setup has grown to 700 words in your draft, you know where the padding is before a single shot is cut.
Short or long for a new channel
New channels are often told to start short because short is easier. For narrated channels aiming at the Partner Program, there is a case for starting at 8 to 12 minutes instead. Watch hours accumulate faster, the videos are eligible for mid-rolls on the day the channel is monetized, and you learn the long format from the first upload. The 4,000 public watch hours the program asks for come much sooner at six minutes per view than at 90 seconds.
The case against is skill. Holding attention for 12 minutes is harder than holding it for four, and a weak long video is worse than a good short one. A sensible compromise is to open at 8 to 10 minutes, watch where the retention line falls away, and extend only once most viewers are reaching the end. Software has limits to plan around as well: JetMotion, for example, renders up to 25 minutes, and the options for longer formats are compared on the long-form AI video generator page.
A quick way to decide
- Write the outline, with one line per beat of the story
- Estimate words per beat honestly, and add them up
- Convert to minutes at your real narration pace
- If the total is just under eight minutes, check whether one more real beat exists, and if it does not, publish the shorter video
- If the total is far above your niche's normal range, split the story into two videos
- After publishing, check where the retention line drops and adjust the next script
Length is an outcome of the outline. Get the beats right, count the words, and the number of minutes mostly takes care of itself. If you are still at the stage of choosing a format, how to start a faceless YouTube channel covers that decision alongside the rest of the setup.


