How do I measure the editing pace of my YouTube video?

The steps at a glance

Editing pace is how often the picture changes, and how that rate moves through the video. You can measure it in a few minutes with a free tool, as long as you know what to compare it with:

1. Export the version you want to measure.
2. Count the cuts with a cut counter.
3. Read the median shot, not only the average.
4. Find the slow minutes on the chart.
5. Measure two or three reference videos the same way.
6. Compare the curve, the transitions and the text.
7. Change one thing, then measure again.

Step 1: Export the version you want to measure

Measure a flattened file, not the project. The export is what a viewer sees: every transition, title card, zoom and layer of B-roll is baked into it, and a timeline view can hide how often the picture really changes. A quick low-resolution render of a draft works too, so you can check the pace before the final pass.

The file has to be one your browser can play: MP4, MOV or WebM. Some MKV and ProRes files won’t open; export an MP4 copy for the check.

Step 2: Count the cuts with a cut counter

The free Cuts Per Minute Counter plays your file inside the browser tab and looks at it several times a second. When the picture jumps between two checks, it narrows the change down to the pair of neighbouring frames where it happened, so a jump cut in a talking-head video counts even when the framing barely moves. The file is never uploaded.

What you get back: cuts per minute, the number of cuts found, the median and the longest shot, a bar chart with the cuts in each minute, a list of the longest shots, a CSV of every cut and an EDL with one event per shot. Three settings matter:

• Checks per second. 4 is the normal setting; 5 is the most precise; 2 is faster for long videos. A talking-head or documentary file takes roughly half its length to scan; fast gameplay takes about its full length. Keep the tab in front while it runs.
• Sensitivity. Low for fast gameplay, where camera turns and explosions change the whole frame at once; High for soft footage. You can change it after the scan without scanning again.
• EDL frame rate. Set it to your timeline’s frame rate if you plan to use the EDL.

Treat the result as close, not exact. The tool’s own hand check on October 19, 2026, on 2 minute clips, found 17 of the 21 cuts in a heavily edited Roblox video, 10 of the 13 jump cuts in a talking-head lesson and 6 of the 8 in a first person gameplay clip, where it also counted about 9 very fast camera turns. Slow dissolves and fades of a second or more are often missed, so a dissolve-heavy edit reads lower than it feels.

Step 3: Read the median shot, not only the average

Cuts per minute is an average. One busy montage can pull it up while the rest of the video sits still, so read it next to the median shot length: half the shots are shorter than that, half are longer. Then open the longest shots list. A few long shots in a fast video are normal and often deliberate. Many long shots in a row are a slow stretch.

Variety is part of a style. The Interpreter, a faceless disaster channel, has a median shot under 5 seconds, yet every 10 minutes I measured still held 2 to 7 shots longer than 10 seconds.

Step 4: Find the slow minutes on the chart

The minute by minute chart is the most useful part of the result. Red bars mark minutes with 2 cuts or fewer. Click one to jump there and watch it with fresh eyes. Something else may be carrying that minute: a strong line, a reveal, a slow move across a photo. If so, leave it. If nothing new reaches the viewer for that whole minute, it is a candidate for a trim, a punch-in or one piece of B-roll.

Check the same spot in YouTube Studio once the video is live. YouTube Help describes a dip in the retention chart as viewers abandoning or skipping at that part of the video, and a flat line as viewers watching that part from start to finish. A slow minute that lines up with a dip is worth fixing on the next video. How to read the retention graph covers the other shapes.

Step 5: Measure two or three reference videos the same way

A cut rate means nothing on its own. Pick two or three videos in your format that your audience already watches and that held viewers well, and measure them with the same settings. Compare a full video with a full video: openings usually run faster than the rest, so the first ten minutes of one video against the whole of another is not a fair test.

If you can’t get a file of the reference, count by hand. Play three separate 60 second windows (one in the opening, one in the middle, one near the end), tally every time the picture changes, and average the three. That is how I checked the automatic count on a Philip Thompson documentary: in three windows counted by eye, the detector had caught 83% of the changes, and 95% of what it flagged was real.

Step 6: Compare the curve, the transitions and the text

Here is what the same measurement looks like on 4 videos from 3 narrated documentary channels, measured frame by frame in August and September 2026:

All three channels tell long, narrated true stories, and two of them cover spies. If one pace were right for the genre, they would sit close together. They don’t. Full details and the method are on how often to cut in a YouTube video. When you compare your video with your references, look past the headline number at three things:

• The curve. In the Philip Thompson video, the cold open ran at about 19 changes a minute, minutes 0 to 10 at 14.5, minutes 10 to 25 at 12.5, minutes 25 to 30 at 10.8 and the closing at 8.2. If your chart is flat from start to finish while your references slow down after the opening, that is a choice you made, maybe without noticing.
• How shots join. Hard cuts at a given rate feel faster than dissolves at the same rate. Librarian’s 2 changes a minute work partly because each two-second dissolve keeps the picture moving.
• Text and sound. They do some of the work cuts do elsewhere. The Interpreter shows text on screen 34 to 42% of the time over a music bed that never drops out; Librarian adds no text and leans on a continuous voice at about 157 words a minute.

Step 7: Change one thing, then measure again

Fix one thing per pass: tighten the red minutes, add B-roll over a long explanation, or speed up the opening if your references do. Then export and run the counter again, so you compare the new cut with the old one instead of with your memory of it. If a minute is slow because the script has nothing new to say there, more cuts won’t save it; that fix belongs in the script.

Dead air often sits in the same places as slow minutes. The free silence detector lists every pause in the audio, so you can line the two up.

What about gaming videos?

Measure them, but read the result with care. In gameplay, camera movement and busy interfaces change much of the picture on every frame. A cut detector is built to tell that apart from a real cut, and in fast footage it sometimes can’t. Use the Low sensitivity and trust the shape of the chart more than the total. I haven’t published cut-rate measurements for gaming channels, so there is no gaming reference number on this page, and I would be wary of anyone who gives one number for every game and format.

How I use pace checks on client videos

When I edit long-form Roblox and Minecraft videos, I don’t aim for a cut rate. I let the moment set the speed: quick through the action, slower where a story beat has to land. Before delivery I scan the chart for one thing, a near-empty minute sitting in an otherwise busy edit. That kind of gap is where people tend to drop, and the fix is usually small: a trim, a punch-in, one piece of B-roll.

Narrated documentaries are different, because the reference comes first. I measured the channels in the table above as part of matching reference styles for documentary edits. In The Spy Show, the narrated spy-history episodes I cut from Vincent Meijer’s script and voiceover, the picture changes every few seconds because the narration keeps moving through places, people and documents. The Interpreter, a client channel, cuts hard every 4 to 5 seconds on its own uploads. When a creator sends me a video they want their edit to sit closest to, I start from its pacing, not from a house style.

Common mistakes when measuring pace

• Borrowing a number from another format. A documentary rate copied onto a talking-head lesson, or the other way round, makes an edit feel wrong without anyone knowing why.
• Reading only the average. The median shot and the chart tell you more than cuts per minute alone.
• Comparing an opening with a whole video. Openings run faster; compare like with like.
• Calling a dissolve-heavy edit slow. Long dissolves are often missed by detectors and still read as movement to a viewer.
• Adding cuts to fix a script problem. If nothing new is said for a minute, faster cutting just decorates the gap.

Editing pace FAQ

How many cuts per minute should a YouTube video have?

No single number works across formats. Even inside one genre the range is wide: in my 2026 frame by frame counts of 4 narrated documentaries, Librarian changed picture about twice a minute, while Philip Thompson and The Interpreter changed it 13 to 15 times. Measure two or three videos in your own format that hold viewers well and start from their pace.

Can I count the cuts in a video without downloading it?

Yes, by hand. Play three separate 60 second windows (opening, middle and end), tally every picture change and average the three. It takes a few minutes and is close enough to compare a reference with your own export, which you can measure with the free cut counter.

Why does the cut counter find extra cuts in my gameplay video?

In gameplay the whole frame can change at once without a cut: a quick camera turn, an explosion, a menu opening. Set the sensitivity to Low for that footage and read the minute by minute chart, because a slow stretch still shows up as slow.

Is a faster cut rate better for retention?

Not by itself. The two spy documentary channels I measured sit about six times apart on cut rate, and both are established. What matters is whether something new reaches the viewer in every stretch. A minute with almost no change in an otherwise busy video is the place to look first.

Does the cut counter upload my video?

No. The file plays in your browser tab and the analysis runs on your computer or phone. It never goes to a server, and the tool keeps no copy.

Sources

Reference pace: cuts per minute measured on 4 videos, my own frame by frame measurements of three narrated documentary channels (Philip Thompson, Librarian and The Interpreter) in August and September 2026, including the hand count of three 60 second windows. Tool settings and the hand check of October 19, 2026: the Cuts Per Minute Counter page. Retention chart terms: YouTube Help, Measure key moments for audience retention, read October 19, 2026. No cut rate is claimed here for gaming, vlogs or talking-head videos, because I haven’t published measurements for them.