How to Edit a Script Down by 30% Without Losing the Point (With a Worked Example)

Almost every script is too long on the first pass, and almost always by the same amount: somewhere around a third. The draft that reads at 85 seconds needed to be 60; the two-minute explainer should have been 80 seconds. That is not a talent problem, it is how writing works. You put the thinking on the page first, then you have to take the thinking back out and leave the point.

This post is the method. It is deliberately mechanical, because 'make it tighter' is useless advice at 11 pm with a recording to finish. There is a worked example, a 120-word paragraph cut to 85, followed by the seven rules that did the cutting, a way to check length against how fast you actually talk, the cases where you should not cut, and how to use the app to record the long and the short version back to back so the camera, not your ego, decides.

The worked example: 120 words to 85

Here is a first draft for a short productivity video. It is fine. It is also 120 words, which at a normal speaking pace is about 50 seconds for what should be a 35-second point.

So today I want to talk about something that I think a lot of people struggle with, which is actually getting started on big tasks. What I personally do, and this has honestly changed the way I work, is I set a timer for just ten minutes. The idea is basically that you only have to work for ten minutes, and after that you are totally allowed to stop if you want to. But the interesting thing is that most of the time you don't actually stop, because once you have started, continuing is a lot easier than starting was. So in order to beat procrastination, try the ten-minute timer. That is really all it takes to get going.

The same point, 85 words:

Big tasks are hard to start, so stop trying to start them. Set a timer for ten minutes. For ten minutes you work; when it rings, you are allowed to stop. Most of the time you won't, because continuing is easier than starting, and the hard part is already behind you. I have used this for two years and it is the only productivity trick that survived. Ten minutes. Set the timer, then decide.

Nothing was lost. The hook is now in the first sentence instead of the fifth. The number is said once, hard, at the end. And the version with the sentence 'that is really all it takes to get going' is gone, because the sentence before it had already said that.

The seven rules that did the cutting

• Cut the throat-clearing. 'So today I want to talk about', 'something I think a lot of people struggle with', 'what I personally do': none of it is the video. Your first sentence should be the first thing you would say if the viewer were about to swipe away, because they are.
• One idea per sentence. The draft's second sentence carried three: what I do, that it changed my work, and the timer. Split them and two of the three turn out to be optional.
• Delete adjectives and adverbs that do not change the picture. 'Actually', 'honestly', 'basically', 'totally', 'really', 'a lot'. Read the sentence without each one. If the meaning is identical, the word was padding you added to sound conversational, and it is cheaper to be conversational by talking to the lens.
• Replace the long form of short words. 'In order to' is 'to'. 'The idea is basically that' is nothing. 'At this point in time' is 'now'. 'Due to the fact that' is 'because'. These phrases are where a third of most scripts hides.
• Kill the recap. The last sentence of a draft is usually a restatement of the one before it, written because endings feel abrupt. On video, abrupt is good. End on the instruction or the image, not on a summary of it.
• Say the number, not the sentence about the number. 'Set a timer for just ten minutes' became 'Set a timer for ten minutes', and the close became 'Ten minutes.' Numbers are the stickiest words in a script; do not wrap them in cushioning.
• Move the point to the first line. If the draft's real claim is in the middle, cut everything before it and see whether the beginning is even needed. In the example, the point was 'stop trying to start, set a timer'; everything before that was warm-up.

How to check the length against the way you talk

Word counts only mean something next to your speaking rate. Most people land somewhere between about 130 and 160 words a minute when they are talking naturally to a camera, slower for emotional or instructional content, faster for energetic hooks. So a 60-second video is roughly 130 to 150 words, and a 30-second one is 65 to 75. These are typical ranges, not rules; the only number that matters is yours.

To find yours, read a paragraph you know the word count of into the phone at your normal pace and time it. Do it twice. Then every future script has a target: words divided by your rate equals seconds. Our earlier post on the ideal script length for a 60-second video goes deeper on platform-specific targets; the short version is that 'a third too long' is almost always what the stopwatch says about a first draft, which is why 30% is the number in this title.

One trap: cutting words and then speaking faster to 'fit' the rest. If the cut version still runs long, you have not cut enough. Pace should stay the same; the script gets shorter.

When not to cut

The 30% rule is for the explainer, the opinion, the tip, the hook-driven short. There are three kinds of script where it does damage.

• Compliance-heavy wording. If you work in medicine, law or finance, some sentences are there because they have to be, and 'tightening' them changes what they promise. Cut around them, not through them. We covered this in the post on scripts for compliance-heavy topics.
• Step-by-step tutorials. When the viewer is following along with their own hands, every step needs its words, including the boring ones like 'now wait for it to finish'. What you cut in a tutorial is the commentary between steps, never the steps.
• Emotional or story-driven pieces. Rhythm carries these, and rhythm sometimes needs a 'slow' sentence. Read the story aloud before cutting it; if a line earns a pause, it stays even if it is technically redundant.

Record both versions, then decide

The honest test is not the word count. It is watching the two takes. Teleprompter: Camera Overlay makes that cheap: paste the long draft as one script and the cut as another, record the first, swap to the second, record again, and both land in the takes library where you can play them back to back. Because the script floats over the camera preview right next to the front lens, you are looking at the viewer in both takes, so the comparison is about the words and not about where your eyes went.

Two settings matter for an edit session. Voice-follow scrolling, which uses the iPhone's on-device speech recognition to move the script at the pace you actually speak and finds your place again if you stumble, is the one to use here: it tells you immediately when a sentence is awkward in the mouth, because you will trip on it and feel the script wait for you. Classic auto-scroll with the speed slider is better once the script is final and you want a consistent pace across takes. In either mode the text size is adjustable, which helps when a cut version suddenly has more white space.

The app is free, iOS-only, and works entirely offline, so nothing you write or record leaves the phone; recordings are portrait 4K and the script never appears in the saved video. Free users see an occasional ad and a small one-time purchase removes them, with no subscription and no watermark either way. If you are still at the drafting stage, the post on writing a script that doesn't sound written is the step before this one.

A ten-minute cutting routine

• Count the words and write the target next to it (words divided by your rate).
• Delete the first sentence. Read it. Delete the second if the piece still makes sense.
• Go through once removing only padding words and long forms of short words.
• Go through once splitting double sentences and deleting the weaker half.
• Delete the last sentence if it restates the one before it.
• Record it. If the stopwatch still says long, go again; do not speed up.

Next in this series: using a teleprompter for product demos, where the script has to follow what is on the screen rather than the other way round, and the cutting rules change accordingly.