The first script I ever wrote for a magic performance was, objectively, terrible. Not because the ideas were bad. The ideas were fine. It was terrible because I wrote it the way I write everything else: like a consultant.
I had spent fifteen years writing strategy documents, pitch decks, and executive summaries. I knew how to construct an argument on paper. I knew how to organize information. I knew how to make a written case that was clear, logical, and persuasive.
None of those skills transferred to writing words meant to be spoken out loud.
I discovered this in the most painful way possible: by hearing my own words played back to me. And the exercise that led to that discovery is one I now consider essential for anyone who scripts a performance of any kind.
The Exercise
Ken Weber describes this process in Maximum Entertainment, and it is disarmingly simple. Write your script. The whole thing. Every word you plan to say on stage, plus notes on actions, movements, and transitions. Get it as complete as you can.
Then record yourself speaking it. Not reading it — speaking it. Deliver it the way you would on stage. Full energy. Full commitment. Full vocal delivery, as if the audience were right there.
Then listen back with your eyes closed.
That last part is critical. Close your eyes because you are no longer the performer. You are the audience. You cannot see the cards, the props, the gestures, the facial expressions. All you have are words and voice. And you need to hear — really hear — whether those words and that voice are doing the job.
I tried this for the first time in a hotel room in Vienna. It was late — maybe eleven at night after a long day of client meetings. I had my script printed out. I propped my phone against the bedside lamp, hit record, and delivered the script start to finish, about seven minutes of material.
Then I sat on the edge of the bed, put in my earbuds, closed my eyes, and pressed play.
What I heard was devastating.
The Gap
There is a gap between writing and speaking that is much wider than most people realize. Written language and spoken language are different animals. They follow different rhythms. They use different structures. They have different tolerances for complexity.
Written language can support long, compound sentences with multiple subordinate clauses because the reader can slow down, reread, and process at their own pace. Spoken language cannot. Spoken language needs to be understood in real time, at the speed of delivery, with no rewind button.
My first script was full of sentences that worked beautifully on paper and died in the air. Sentences like: “What I am about to show you is something that, when I first encountered it, fundamentally changed the way I thought about what is possible when two people focus their attention on the same object.”
Read that on a page, and it tracks. You can parse the clauses. You can follow the logic. You can appreciate the building rhythm.
Say that out loud to a human being standing three feet from you, and they are lost by the second comma. Their eyes glaze. Their attention wanders. By the time you reach “the same object,” they have already stopped listening and are thinking about what they want for dinner.
This was the first thing I heard when I listened back to my recording. My sentences were too long. My vocabulary was too formal. My structure was too nested. I was writing for readers, not listeners.
But that was only the beginning.
The Voice Problem
The second thing I noticed was my voice. Not the words — the voice itself. The way I sounded when I delivered scripted material was completely different from the way I sounded in conversation.
In conversation, I am relaxed. I vary my pace. I pause when I am thinking. I emphasize words naturally based on what I actually care about. My voice has texture.
On the recording, I sounded flat. Even. Metered. Like a newsreader. Every sentence had the same rhythm. Every phrase got the same emphasis. Every pause was the same length. I was reading, even though I thought I was speaking.
This is the trap that scripting sets for you if you are not careful. When you have words on a page, your brain defaults to reading mode. Your voice adopts the cadence of someone processing text rather than the cadence of someone sharing a thought. The result is technically competent and emotionally dead.
I listened to about ninety seconds before I pulled out the earbuds and stared at the ceiling. I was genuinely embarrassed, even though no one else was in the room. Because I realized that what I had been delivering on stage — what I thought was a polished, professional performance — probably sounded exactly like this. Flat. Formal. Lifeless.
The writing was not the problem. Or rather, the writing was only half the problem. The other half was the delivery. And I would never have known about either half if I had not recorded myself and listened back.
The Revision Process
I started the revision that same night. Here is what I did, and what I still do every time I script new material.
First, I went through the script and broke every long sentence into short ones. Not every sentence needs to be short. But the default should be short, and length should be a deliberate choice, not a habit. If a sentence has more than one idea in it, it is probably too long for spoken delivery.
Second, I eliminated formal vocabulary. Words like “fundamentally,” “encountered,” “demonstrate,” and “perspective” are consulting words. They are written words. People do not say “fundamentally” in conversation. They say “completely.” They do not say “encountered.” They say “found” or “came across.” Every word needed to pass the test: would I actually say this to a friend?
Third, I rewrote transitions. On paper, transitions can be implicit. You can end one paragraph and start the next, and the reader’s eye carries them across the gap. On stage, transitions need to be explicit. You need a bridge sentence. You need a phrase that tells the listener, “We are moving from this to that.” Without it, the audience feels a jolt, a moment of confusion, a tiny break in the flow that undermines their trust in you as a guide.
Fourth — and this was the hardest part — I threw out entire sections that sounded good on paper but added nothing to the experience. Passages that were there because they were well-written, not because they served the performance. Observations that I liked as a writer but that an audience would experience as filler. Every word in a performance must earn its place, and the recording made it brutally clear which words were earning and which were freeloading.
Then I recorded the revised script and listened back again.
Better. But not right yet.
So I revised again. And recorded again. And listened again.
I went through this cycle four times before the script sounded like me. Not the me who writes emails. Not the me who presents to boards of directors. The me who sits across from someone at a coffee shop and explains something I find genuinely fascinating.
The Closed-Eyes Principle
Closing your eyes while you listen is not a gimmick. It is doing something specific and important: it removes all the visual information that normally compensates for weak language.
When you are on stage, your physical presence covers a multitude of verbal sins. Your gestures add emphasis. Your facial expressions convey emotion. Your movement creates energy. The visual spectacle of the performance — the cards, the props, the effects — fills gaps that the words leave open.
When you close your eyes, all of that disappears. You are left with nothing but the words and the voice. And if the words are weak — if they are confusing, or flat, or boring — there is nothing to hide behind.
This is a humbling experience. It was humbling for me. But it is also one of the most useful things I have ever done for my performance. Because the words that survive the closed-eyes test are words that work. They are words that carry their own weight. They do not need visual support. They do not need gestures to land. They communicate on their own terms, and when you add back the visual elements of performance, they become even stronger.
Think of it this way: if your script works with eyes closed, it will be powerful with eyes open. If it only works with eyes open, it is fragile, dependent on your physical performance to compensate for verbal weakness. And on the nights when your energy is lower, when the room is tougher, when something goes sideways — those are the nights when the visual elements may not be enough. On those nights, you need words that carry themselves.
What I Hear Now
These days, the recording exercise is a standard part of my scripting process. I write, I record, I listen, I revise. Sometimes the cycle takes two passes. Sometimes it takes six. The number does not matter. What matters is that I do not consider a script finished until it sounds right in my ears with my eyes closed.
I listen for specific things now. I listen for sentences that make me lose the thread. I listen for words that sound borrowed rather than owned. I listen for stretches where my voice falls into a pattern — same rhythm, same pace, same emphasis — because patterns are where the audience starts to tune out. I listen for moments where I am telling the audience something instead of sharing something with them.
And I listen for the moments that work. The lines that make me lean forward even though I wrote them. The pauses that create anticipation even though I planned them. The phrases that sound so natural I almost forget I scripted them. Those are the moments I am building toward.
The Bridge to Your Work
This exercise is not specific to magic or mentalism. It works for any spoken performance. Keynote addresses. Sales presentations. Wedding toasts. Anything where you are going to stand in front of people and use words.
The principle is the same everywhere: written language and spoken language are different, and the only way to close the gap is to hear yourself speak your written words and revise until the gap disappears.
If you have never recorded yourself delivering your material, you are operating blind. You are assuming your words work because they look right on the page. That assumption is almost certainly wrong, at least in places. You will not know which places until you hear them.
Sit down. Write your script. Stand up. Record yourself delivering it. Sit back down. Close your eyes. Press play.
Then start revising.
The version that comes out the other side of this process will sound nothing like the version you started with. It will be shorter, sharper, simpler, and more human. It will sound like you. Not the you who writes. The you who talks.
And that is the version your audience deserves to hear.