How Long Should YouTube Videos Be? Lessons From 200+ Films

How long should YouTube videos be? After publishing 200+ films and earning 60M+ views across four channels, we can give you a number: for documentary-style content, 20 to 37 minutes. Every episode we release lands in that band, and it isn't a creative preference. It's the output of watch-time math, retention graphs, and the hard economics of producing original 3D animation every week.

But our number is not your number. The honest answer to video length is a formula, not a constant — and once you understand the formula, you can stop copying other channels' runtimes and start reading your own data. Here's the math, the reasons we landed where we did, and the specific cases where shorter flatly wins.

How Long Should YouTube Videos Be? Do the Watch-Time Math

YouTube does not rank videos by length. It ranks them by satisfied watch time — how many minutes a viewer happily gives you per impression, and whether they keep watching the platform afterward. Length is just the container; watch time is the currency.

Run the numbers. A 10-minute video holding 60% average retention delivers 6 minutes per view. A 30-minute film holding 45% — a "worse" percentage — delivers 13.5 minutes per view, more than double the watch time. That's why a longer film with an uglier-looking retention graph can still beat a tight short video in the algorithm's eyes.

Monetization stacks on top. As of 2026, videos over 8 minutes can carry mid-roll ads, which is where most long-form ad revenue actually comes from. Public RPM figures vary wildly by niche, but the pattern holds: more watch minutes per view means more ad slots per click of the thumbnail. (Typical public figures — not financial advice.)

Why Our Documentaries Run 20–37 Minutes

Across Blackfiles (126 films, 53M views), Breakfiles, Outplayed, and Outlived, every episode we publish runs between 20 and 37 minutes. We didn't set that band in a strategy meeting and then defend it. It emerged from 200+ films' worth of retention data.

Below 20 minutes, an investigative story can't breathe. You need setup, escalation, the reveal, and the aftermath — compress that arc into 12 minutes and it plays like a Wikipedia summary. Our biggest films, like "The FBI Agent Who Warned Everyone About 9/11" (482K views) and "The ONLY Person Who Survived 133 Days Stranded at Sea" (475K views), all live inside the band.

Above 37 minutes, two things degrade at once. Back-half retention sags faster than the extra minutes add watch time, and production cost climbs in a straight line — we animate every frame in original 3D with zero stock footage, so each additional minute is real money and real days. With weekly uploads on every channel and a roughly 25-person team, runtime discipline is calendar discipline.

The Retention Curve Decides, Not the Stopwatch

We never decide a film is 28 minutes in advance. We research long — 16 to 20 hours per film before a single frame exists — script long, then cut until every minute earns its place. The runtime is whatever survives the edit.

• Nail the first 30 seconds. Restate the promise of the title and thumbnail, then complicate it. The steepest drop-off happens here no matter how long the video is.
• Cut by segment, not by vibes. When a story beat sags in the retention graph, the next script fixes that beat. We treat every published film as data for the next one.
• Never pad to hit a threshold. Stretching 7 minutes of story to cross the 8-minute ad line costs more in retention than it gains in mid-rolls.
• End before the energy does. A viewer who finishes wanting more clicks the next film. A viewer who finishes exhausted doesn't come back.

Our Scriptwriter pipeline gets us from research to draft fast, but the value is in what gets deleted. A 30-minute film usually starts life as a 45-minute script.

When Shorter Wins

Long-form documentary is our entire business, and we'll still tell you it's not always the answer. There are formats where adding minutes only subtracts value.

• Search-intent answers. If the viewer asked a specific question, the best video is the shortest one that fully answers it. A 5-minute tutorial beats a 15-minute one carrying the same information.
• Fast-moving topics. News and reaction content rewards speed to publish, not depth. A multi-week production cycle loses the window entirely.
• Unproven channels. If your last ten videos couldn't hold viewers for 8 minutes, a 30-minute upload won't fix that — it will document it.
• Shorts as a discovery layer. Shorts rarely pay like long-form, but they're a cheap top-of-funnel. As of 2026, the YouTube Partner Program threshold is 1,000 subscribers plus 4,000 public watch hours or 10 million Shorts views — Shorts can get you to the door faster, but long-form pays the rent once you're inside.

Typical Video Lengths by Format (As of 2026)

If you need a starting point before you have your own data, these are the typical public ranges across formats. Treat them as defaults to test against, not laws.

• Tutorials and how-tos: 5–12 minutes — exactly as long as the answer requires.
• Vlogs and daily content: 8–15 minutes.
• Video essays and commentary: 15–35 minutes.
• Documentaries and true crime: 20–45 minutes — our own band sits at 20–37.
• Podcasts and interviews: 45 minutes and up, usually clipped into shorter cuts.

One caveat overrides the whole table: your retention data beats any benchmark. If your audience holds for 22 minutes, you make 22-minute videos — whatever the format is "supposed" to run.

How Long Should YouTube Videos Be When You're Starting Out?

Shorter than your ambition. We launched Blackfiles in February 2025 straight into full-length films, but we did it with an in-house team and a tested pipeline behind it. If you're solo, don't copy our runtime — copy our discipline.

The ladder is simple. Publish at a length you can sustain every single week, watch your average view duration across 8–10 uploads, and extend runtime by 20–30% only when the curve holds. It's the same runtime ladder we teach inside Sentris Academy, because the math doesn't care whether your team is one person or twenty-five.

FAQ: How Long Should YouTube Videos Be?

Is 10 minutes still the magic number for ads? No — that threshold dropped years ago. As of 2026, videos over 8 minutes qualify for mid-roll ads, but crossing the line with padding hurts retention more than the extra slot helps.

Do longer videos rank better? Not because of length. They rank better when they generate more total watch time per impression, which longer videos can do even at lower retention percentages. A boring 40 minutes loses to a gripping 12 every time.

Should beginners make Shorts or long-form? Both, with different jobs. Shorts are discovery; long-form is where watch time, ad revenue, and audience loyalty actually live. We built 500K+ subscribers almost entirely on 20–37 minute films.