How long should each shot be?
There is no single magic number, but there are workable ranges by genre, and one number that matters more than any single cut. Here is how long a shot should hold, and how to find the ones quietly killing your pace.
By Thomas, founder of CutScore · Updated June 2026
Everyone wants one number, and I understand why. It would be lovely if I could say "hold every shot for 4.2 seconds" and we could all go home. But shot length is not a setting, it is a decision you make hundreds of times in a single edit, and the right answer changes with what you are saying. A reaction needs a quarter of a second. A reveal might need ten.
I have shipped videos that died from both directions. Early on I held shots far too long, because I had recorded them and felt I owed them screen time. Later I overcorrected and cut so fast the video felt like a strobe light: lots of motion, nothing to hold onto. Both versions lost viewers in the same place, and for the same underlying reason. The shot length was not following the meaning.
So the honest answer has two parts. There are sensible ranges by genre, which we will get to. And there is one rule that beats all of them: a shot lives exactly as long as it keeps giving the viewer something new. The second it stops, you cut. Let me show you the ranges, then the rule.
How long should each shot be, by type of video?
These are starting points, not laws. Use them as a sanity check, then let the content stretch or shorten any individual shot.
| Video type | Typical shot length | Why it sits there |
|---|---|---|
| Fast social short | ≈ 1 to 3s | A vertical feed punishes hesitation, so movement and new frames keep the thumb still. |
| Talking-head / explainer | ≈ 3 to 6s | Long enough to follow a thought, short enough that you are never staring at a frozen face. |
| Vlog / lifestyle | ≈ 2 to 5s | Energy and place change often, so shots turn over to keep the scene alive. |
| Tutorial / how-to | ≈ 4 to 8s | Hands and screens need time to read, and rushing a step just means a re-watch. |
| Cinematic / B-roll | 6s and up | A composed shot earns the time when light, motion or depth keep revealing detail. |
| Reaction / comedy beat | < 1s | The joke is the cut. Hold a reaction too long and you flatten the timing. |
Measuring average shot length and spotting the one shot that overstays is tedious to do manually. CutScore measures both in one pass and points at the exact timestamp, so you can fix the pacing instead of auditing it.
When should a shot end? The moment it stops giving you something.
Every shot has to earn its seconds
Here is the test I use on a held shot. While it plays, is anything changing? New information, new movement in the frame, a shift in your face or your voice, a beat in the music. If yes, the shot is still working and can stay. If nothing is changing and you are just filling time, that is dead air, and dead air is where attention leaks out. A long shot is not a problem. A long static shot of you searching for the next word is. Length and boredom are different things, and we confuse them constantly.
Cut on meaning, not on a timer
The fastest way to make an edit feel robotic is to cut at a fixed interval, every three seconds like a metronome. Real pacing breathes. You stack a few quick cuts to build energy, then hold a longer shot so a point can land, then move again. The rhythm should follow what you are saying. When a thought is dense, give it room. When it is a list or a punchline, speed up. If you want the mechanics of trimming dead air without reshooting, the humble jump cut does most of that work, and I go deeper on tempo in how fast should I cut my video.
The first shot is the one that has to be tightest
Your opening shot does not get the genre average. It gets less. Most of a video's drop-off happens in the first few seconds, so a slow logo or a three-second throat-clear at the top is the most expensive long shot in the whole edit. Start on the most interesting frame you have, and keep the cuts honest until the viewer has committed. There is more on this in what the first 3 seconds should be, and on holding attention after the hook in hook retention.
How to find the shots that drag
Watch your rough cut at normal speed and put a finger on the spacebar. Every time you feel the tiniest urge to skip ahead, mark it. Those marks are almost always a shot held a few seconds too long, and they cluster in the same spots: the middle of a sentence you fumbled, a B-roll clip you liked too much, the bit after the point already landed. You do not need to cut them shorter everywhere. You need to find the specific offenders. That is the difference between a vague "tighten the edit" note and a list of timestamps.
Here is a real CutScore report for an everyday video: average shot length, the cut rhythm, and the exact timestamps where a shot sits too long.
Three shot-length fixes that do the most.
Most of the jump from "this drags" to "this moves" comes from these three. Do them in order.
By feel, by counting, or in one pass.
By feel, after a break
Free, and useful if you step away first. Watch the rough cut a day later with a finger on the spacebar and mark every urge to skip. The catch is the one I keep hitting: after forty viewings, my own edit always feels faster to me than it really is.
By counting cuts
Accurate, but tedious. Count your shots, divide the runtime, and you have an average shot length. Then scrub for the longest gaps by hand. It works, and it teaches you a lot, but almost nobody keeps doing it on every video. More on the maths in cuts per minute.
With a coach in one pass
Hand the file or a link to CutScore. It measures your average shot length, reads the cut rhythm against your genre, and flags the exact shots that sit too long, with timestamps and a fix for each. See a sample report.
Frequently asked.
Stop guessing how long your shots should be.
CutScore measures your shot lengths and cut rhythm and tells you exactly which shots to trim, with the timestamps to back it up. Join the waitlist for early access.
Join the waitlist