Explainer video script the structure, and 3 annotated samples
The six beats of an explainer script with timings and word budgets, the two-column format studios use, and three complete sample scripts for fictional products, with a note on every line.
Updated November 10, 2026
An explainer video script is the narration and the matching visuals for a short explainer, written as six beats: hook, problem, solution, how it works, proof and call to action. A 60-second script runs about 130 to 160 words, and the product should be named by about 15 seconds. Write it in two columns, said and shown.
What is the structure of an explainer video script?
Most studios use a version of problem, agitate, solve (Copyblogger); VideoScribe's flow is hook, problem, solution, how it works, social proof and call to action (VideoScribe). For 60 seconds at 150 words a minute:
Beat
Time
Words
Its job
Hook
0–5 s
~13
Open on the viewer's problem or the most striking moment. Not your logo.
Problem
5–15 s
~25
Make the problem specific and familiar, in the viewer's own words.
Solution
15–25 s
~25
Name the product as the answer, in one sentence.
How it works
25–45 s
~50
Two or three moments of the product doing the job, each with a picture.
Proof
45–52 s
~18
One reason to believe: a number, a customer, a result.
Call to action
52–60 s
~20
Name, where to go, one thing to do. Hold it long enough to read.
Word counts are the beat's seconds × 150 words per minute. Timings follow TechSmith's one-minute template (hook in the first 10 seconds, call to action in the last 10) (TechSmith).
How should you format an explainer video script?
In two columns: the voiceover on the left, what's on screen on the right, one row per shot. Vimeo and HubSpot both recommend it (Vimeo, HubSpot). It forces a picture for every line, shows where the visuals can say something so the voice doesn't have to, and is the document an animator or director works from. Keep each on-screen line up for at least three seconds (TechSmith), and make the video work with the sound off.
Each is for a fictional product, written to the structure above. The word count and runtime are computed from the voiceover at 150 words a minute.
Sample 1: Ledgerbird, an invoicing app for freelancers (60 seconds, motion graphics)
142 words of voiceover ≈ 57 seconds at 150 wpm; the rest is held visuals and the end card.
Beat
Voiceover
On screen
Why it's there
Hook
You finished the job three weeks ago. You're still waiting to get paid.
An invoice card slides in; a red "Overdue: 21 days" tag stamps onto it.
Starts inside the viewer's problem, with a specific number they recognise.
Problem
Chasing invoices means awkward emails, a spreadsheet you never quite update, and guessing who has paid and who has forgotten. It eats the hours you meant to spend on paid work.
Three quick cuts: a half-written reminder email, a messy spreadsheet, a phone with no new notifications.
Three concrete pains, one picture each. No statistics we can't back up.
Solution
Ledgerbird sends your invoices, tracks every one, and follows up for you, so getting paid stops being a second job.
Logo resolves out of the invoice card; the product screen fades in behind it.
One sentence, three verbs and the benefit. The product name lands at 15 seconds, not 0.
How it works
Add a client once, with their rates and payment terms. Send an invoice in two taps from your phone or laptop. If it's late, Ledgerbird sends a polite reminder on the schedule you choose, in your own words, and tells you the moment the money lands.
The real UI, simplified: client form, the send button pressed, a reminder timeline, a "Paid" notification.
Each clause matches one screen. Real interface, cleaned up, beats stock metaphors.
Proof
You can see when each invoice was opened, so you know exactly where things stand before you ask.
Close-up on an "Opened 9:41" status next to the invoice.
In a real script, put a true customer result or number here. This is a feature-proof stand-in.
Call to action
Ledgerbird. Get paid without the chasing. Try it free on your next three invoices.
Logo, tagline and "Try it free" button, held for three seconds.
Name, promise, one action. Text stays up long enough to read with the sound off.
64 words of voiceover ≈ 26 seconds at 150 wpm; the rest is held visuals and the end card.
Beat
Voiceover
On screen
Why it's there
Hook
Nobody remembers what was decided in that meeting. Not even the person who called it.
A calendar block labelled "Sync" dissolves into scribbles.
Short and relatable enough to stop a scroll.
Solution
Recapo joins your video call, listens, and writes the recap for you.
A small Recapo bot icon pops into a video-call grid.
At 30 seconds, problem and solution share the first 10 seconds.
How it works
Every decision, every owner and every deadline, sent to the whole team in the tool you already use, five minutes after you hang up. No more notes nobody reads.
A recap card builds line by line: three decisions, two names, one date; then an email lands.
One moment, shown completely, instead of three half-shown features.
Call to action
Recapo. Try it free on your next meeting.
Logo and URL on screen; captions burned in throughout.
Feeds autoplay muted, so every line is also on screen as text.
Sample 3: LeakLatch, a smart water-leak sensor (90 seconds, live action)
182 words of voiceover ≈ 73 seconds at 150 wpm; the rest is held visuals and the end card.
Beat
Voiceover
On screen
Why it's there
Hook
Most leaks start where nobody is looking. Under a sink, behind a machine, in a corner of the basement.
Slow push-in under a kitchen sink; a single drip falls in the dark.
A physical product earns the opening shot; live action makes it real.
Problem
A slow drip can run for weeks before anyone notices. By the time you see the stain on the ceiling, the floor underneath is already soaked, and the repair bill is already growing.
Three locations, handheld, a homeowner's torch finding a damp patch; then a ruined floorboard lifted out.
Turns a vague worry into a picture of their own floor.
Solution
LeakLatch is a water sensor the size of a coin. Put it where water could escape, and it tells your phone the second it gets wet.
Close-up: a hand places the sensor on the floor; it clicks into place.
Size and the core promise in one line, with the product in a hand for scale.
How it works
There's nothing to wire and no hub to set up: peel, stick, and pair it with the app. When a single drop touches it, you get an alert wherever you are, and it chirps loudly at home so whoever is in the house hears it too. If your Wi-Fi drops, it still sounds the alarm.
Montage of placing sensors; a phone on a café table lights up with "Leak detected: Kitchen"; the sensor chirps in an empty house.
Answers the buyer's three doubts in order: installation, setup, reliability.
Proof
Start with the three spots you worry about most: the kitchen sink, the washing machine and the water heater. One sensor covers each.
The three locations from the problem beat, now each with a sensor and a green tick.
Calls back to the opening shots. With a real product, use a real customer or test result here.
Call to action
LeakLatch. Know before the floor does. Get a three-pack today at your hardware store, or order online and have them under your sinks by the weekend.
Product pack shot on the kitchen counter; logo and "Available now" super.
Tells people where to buy, not just that it exists.
What do the three samples have in common?
The product name arrives after the problem, never in the first line.
Every voiceover line has its own picture, and the picture adds something the line doesn't say.
One idea per beat. The 30-second cut drops the problem and proof beats instead of squeezing them.
The call to action names one action, not three.
How does the script change by style?
Animated and 2D: the "on screen" column becomes the storyboard, so describe each scene precisely; for characters, name them and their problem early.
Live action: write for a speaker, read it aloud, and add locations and props to the visual column; that's the shot list.
Pick one viewer and one message, then write six beats: hook, problem, solution, how it works, proof and call to action. Write it in two columns, what is said and what is shown, so every line has a picture, and keep it to about 150 words per minute of video.
How many words should an explainer video script be?+
About 130 to 160 words for 60 seconds, 65 to 80 for 30 seconds and 195 to 240 for 90 seconds. Studios write to 125 to 160 words per minute, and a measured study of 862 online videos found speakers averaged 156 (Guo et al., 2014).
What is the structure of an explainer video script?+
Hook (0 to 5 seconds), problem (5 to 15), solution (15 to 25), how it works (25 to 45), proof (45 to 52) and call to action (52 to 60). It is an extended version of problem, agitate, solve.
Is there an explainer video script template?+
Yes: the video script template has two-column 30, 60 and 90 second templates with a word budget per beat, and the script generator drafts a scene-by-scene script from your product description. Both are free.
What is a good explainer video script example?+
The three samples on this page: a 60-second motion-graphics script for an invoicing app (142 words), a 30-second social cut for an AI notes tool (64 words) and a 90-second live-action script for a leak sensor (182 words), each annotated line by line.
Skip the blank page
Paste your product link. nod writes the script, then builds the video with motion graphics, voiceover and music.