How to write an explainer video script that teaches

Teach, don’t describe: the one rule of an explainer video script

Say you’re making a two-minute lesson on the heart. There’s a beautiful 3D heart on screen, and your first draft reads: “Here we see the heart. It has four chambers. The left ventricle is the largest.”

Every word is true, and almost nothing is taught. The viewer can see the heart already. The voice spends its seconds on what the eyes did for free, and never says why it matters.

Now the same picture with a different line: “Put your hand flat on the left of your chest. That thump is your heart’s left side squeezing, sending blood out to your whole body.” The viewer does something, feels something, and ties the picture to their own body.

That’s the whole rule. The picture shows; the voice teaches. A narration that teaches talks to “you”, gives cues (what to do, what to notice, what to avoid) and says what the picture means. It never reads the screen back.

There’s research behind the “you”, too. Richard E. Mayer, the psychologist behind much of what we know about learning from narrated animation, calls it the personalization principle: people learn better from a conversational style than from a formal one.

Before and after: describing vs teaching

Four lines from typical first drafts, each rewritten to teach. The second version always gives the viewer something to do, to notice or to answer.

• Before: “Here we see a diagram of the heart.” After: “Put your hand on your chest. That thump is the left side of your heart pushing blood out.”
• Before: “This chart shows the temperature of the water over time.” After: “Watch the line go flat at 100 degrees. The heat keeps coming in, so where does it go? Into turning water into steam.”
• Before: “In this clip, the athlete performs a squat.” After: “Push your knees out as you go down. Keep your weight in your heels, not your toes.”
• Before: “This screen shows the three steps of onboarding.” After: “Your new client signs on Monday. By Friday, they should have done three things. Here’s the first.”

The before lines name the picture; the after lines use it. If deleting a line would cost the video nothing, it was describing.

How to write a script for an explainer video, step by step

Write it the way you’d explain it to one person sitting next to you. Then check it line by line.

1. One idea per line. Each line gets one picture. If a line needs two pictures, it’s two lines.
2. Short spoken sentences: about 15 words at most, five or six seconds of voice. Read the line aloud. If you need a breath in the middle, cut it in two.
3. Say “you”. “You’ll feel it in the back of your thighs” teaches; “the hamstrings are engaged” describes.
4. Ask the question before the answer. “So where does all that wood come from?” makes the viewer guess, and a guess makes them want the answer.
5. Use concrete numbers. “The tree gained 74 kilos; the soil lost barely 60 grams” sticks. “The tree gained a significant amount of mass” slides right off.
6. Leave a pause before the key line. In Clipteach an empty line between two paragraphs is a short breath in the voice, and once the video is made you can add a pause after any scene.
7. End on the one line you want remembered, said plainly. One sentence, not a summary.

Write the [notes]: direct the scenes without saying a word

A video script has two layers: what the voice says and what the screen shows. In Clipteach you write both in one text. Anything in [square brackets] is a note: the voice never reads it, and the AI director takes it as the instruction for the lines that follow.

Put each note on its own line, just before the words it shows. Be concrete: what appears, in what order, and the exact words of any label, in quotes. Colours and positions help. [A leaf, drawn big, the sun’s rays on it, the word “sugar” written inside] tells the director far more than [photosynthesis].

This is where the describing goes. “A heart, its left side lit up” belongs in a note, for the director. Freed from describing, the voice can teach.

A [SCREEN] note marks where one of your screen recordings plays: the words after it are the slot for that recording, until the next note. Write those lines like a guide at the viewer’s side (“Click Create, then paste your lesson”), not like a tour of the interface.

Recording your own voice instead? Use the Script & notes box beside your recording to tell the director what to show and what to highlight.

A complete educational video script, with its notes

Here’s a short script written with every check above: six notes, six spoken lines, about half a minute in the Doodle style. Paste it into Clipteach as is, with Rewrite off: the voice reads the lines and the notes direct the drawings.

[A glass of water, drawn line by line, an ice cube floating at the top.]

Drop an ice cube in a glass of water. Why doesn’t it sink?

[Beside it, a pot of melted wax with a solid lump of wax lying at the bottom. A red question mark between the two.]

Most solids sink in their own liquid. Solid wax sinks in melted wax.

[Water molecules as small circles: packed and jumbled under “water”, spaced out in neat hexagons under “ice”.]

When water freezes, its molecules lock into a pattern that takes up more room.

[A big number, “9%”, with the words “more room” in blue.]

About nine percent more. Same water, lighter for its size. So it floats.

[An iceberg, drawn big, cut by the sea line, the part under the water shaded blue, with the label “90% under”.]

That’s why you only see an iceberg’s tip: about nine-tenths of it is underwater.

[A frozen lake: a layer of ice on top, fish swimming in the dark water below. Then the camera pulls back over the whole board.]

And it’s why lakes freeze from the top down, and fish survive the winter below.

The voice never says “here is a glass” or “this diagram shows”. The notes carry the pictures. Every spoken line asks a question, explains or points at something you can check yourself.

Rewrite on or off: who writes the narration?

With Rewrite on, you paste raw material such as a lesson, your notes or an article, and the AI writer turns it into a spoken narration at the length you choose. It’s the fast road when you have content but no script.

With Rewrite off, your text is the narration, read word for word, and your [notes] reach the director. With Rewrite on they don’t: the writer writes plain narration. So once you’ve written a script the way this post describes, keep Rewrite off and keep every word.

Either way, nothing is final: click any scene in the studio to change its words, or ask the video assistant in plain words. Rather speak than type? Record the lesson in your own voice: every scene comes in on the word it belongs to.

The same rule runs through Clipteach

We hold our own AI to it. Pick AI narrates on a silent clip, a gym set or a drill, and the narration coaches from what the clip shows: what to do, what to feel, what to avoid. See it on sport clips.

Whoever you teach, the voice has the same job, in a paid course as in a lesson for your class.

And if you wonder why a drawing that appears as the word is said works so well, that’s the next read: why whiteboard animation works.