Blog
How to Write an Explainer Video Script: Structure, Word Counts and a Template
An explainer video script is where the whole video gets decided: what the viewer hears, in what order, and how many seconds each idea is allowed. Write it in six beats, give each beat a word allowance, and read it aloud before anyone draws a frame. Below is the structure we use for B2B technology products, a template you can copy and a worked example.
What an explainer video script decides
Most teams sit down to a script with one question: what should we say about the product? That question produces a feature list read aloud. The better question is what the viewer should be able to repeat to a colleague ten minutes after the video ends.
The script answers that question in writing, before anyone designs a single picture. It fixes the order of ideas, the words the voiceover will use, the moments the screen must show something specific, and the length of every section. The storyboard draws it, the voice artist records it and the animator times each movement to it. In our explainer video production process, the script is the first of three sign-offs, and the one that shapes the other two.
Three decisions before you write a line
A script written before these three decisions will be rewritten after them. Put each one in a single sentence at the top of the document, and get the person who signs off to agree all three.
- The viewer, named. A head of finance at a 200-person logistics firm in Kuala Lumpur watches differently from a developer evaluating an API. Name the role, the company size and what they already believe about the problem. The vocabulary, the pace and the examples all follow from that person.
- The one takeaway. Write the sentence the viewer should leave with, in their words. “It reads our invoices so the team only checks exceptions” is a takeaway. “An AI-powered end-to-end platform” is a tagline, and nobody repeats a tagline to a colleague.
- The next step. Book a walkthrough, start a trial, read the security page, forward it to the IT lead. Choose one. A video that ends with three options gets none of them.
Then decide where the video will play, because placement sets the runtime and runtime sets the word count. Our guide to explainer video length and word counts covers lengths by placement and the 140-words-a-minute planning figure used throughout this article.
Have the takeaway, but not the script?
Send us the viewer, the sentence you want them to repeat and the next step. Our scriptwriters will turn it into a timed first draft built on the structure below.
Start an explainer script →A six-beat structure for B2B explainers
Business buyers watch an explainer to answer three private questions: is this about my problem, does it actually work, and what do I do now? Six beats answer them in order. The shares below are for a 90-second video, our most common explainer length, and the words assume the 140-a-minute planning figure.
| Beat | Its job | Time in a 90-second video | Approximate words |
|---|---|---|---|
| 1. Hook | Name the viewer’s situation so they recognise themselves | 0–10 sec | 20–25 |
| 2. Stakes | Show what the situation does to their week, their team or their numbers | 10–25 sec | 30–35 |
| 3. Shift | Introduce the product as the change, in one sentence | 25–35 sec | 20–25 |
| 4. How it works | Walk through three steps the viewer can picture | 35–70 sec | 70–80 |
| 5. Proof | Give one reason to believe it: a named customer, an approved result, a certification | 70–82 sec | 25–30 |
| 6. Next step | Ask for one action and say where to take it | 82–90 sec | 15–20 |
These shares are our planning rule of thumb. They add up to slightly under the full 210 words on purpose: the viewer needs a beat of silence to watch the interface do something.
Why the hook names a situation
An opening that describes a moment the viewer lives through (“month-end, 400 invoices, one shared inbox”) earns the next ten seconds. Keep the brand out of the first line and bring it in at the shift.
Why how-it-works is three steps
Three steps is about as much process as a viewer can hold from one viewing. If the product genuinely needs five, merge the smallest two or move the detail into a product demo video, which has the runtime for it. The difference between the two formats is set out in explainer video vs product demo video.
Why proof comes late
Proof lands harder once the viewer understands what is being proved. A customer logo placed after the three steps answers the question the viewer has just started asking. Use only results your customer has approved.
Explainer video script template
Copy this explainer video script template into a document with two columns. The left column is what the viewer hears. The right column is a short note for the storyboard artist about what the viewer sees. Fill the left column first, and keep each beat inside its word allowance.
| Beat and allowance | Voiceover (what they hear) | Visual note (what they see) |
|---|---|---|
| Header | Viewer: [role, company type, market]. Takeaway: [one sentence in the viewer’s words]. Next step: [one action]. Runtime: [seconds]. Placement: [where it plays]. | Brand guidelines, product screens to use, anything that must not appear |
| 1. Hook 20–25 words |
[A moment the viewer recognises, told in second person. Time, place, a number.] | [The situation, shown plainly. No logo yet.] |
| 2. Stakes 30–35 words |
[What that moment does to their team, their customers or their week. One concrete consequence.] | [The consequence made visible.] |
| 3. Shift 20–25 words |
[Product name] [does the one thing], so [the takeaway]. | [First appearance of the brand and product.] |
| 4. How it works 70–80 words |
First, [step one]. Next, [step two]. Then, [step three and its result]. | [One screen or diagram per step. Mark the element to highlight on each.] |
| 5. Proof 25–30 words |
[Named customer, approved result or certification, stated plainly.] | [Logo, quote card or badge.] |
| 6. Next step 15–20 words |
[One action], at [where]. | [End card: action, URL, brand.] |
The visual-note column is deliberately short. It tells the storyboard artist what must appear, and leaves how it appears to them. Our guide to the video storyboard picks up where this column ends, turning each row into frames with timing and on-screen text.
An explainer video script example
Here is the template filled in. This is an illustrative example for a fictional product. “Tallyroom” and its features are invented to show the structure at work; they describe no real company or customer.
Header. Viewer: finance manager at a mid-sized distributor paying suppliers in several Southeast Asian currencies. Takeaway: “It reads our invoices, so the team only checks the ones that don’t match.” Next step: book a walkthrough. Runtime: 90 seconds. Placement: product page and sales follow-up email.
| Beat | Voiceover | Visual note |
|---|---|---|
| 1. Hook | It’s the last week of the month. Your team has four hundred supplier invoices, three currencies and one shared inbox. | An inbox counter ticking up; three currency symbols stacking on a desk. |
| 2. Stakes | Every invoice gets opened, typed in and chased for approval by hand. One wrong digit in a bank detail, and a supplier waits another month to be paid. | A hand keying numbers; one digit flips red; a calendar page turns. |
| 3. Shift | Tallyroom reads each invoice the moment it arrives, so your team starts the day with the checking already done. | Brand appears; the inbox counter drops to a short list. |
| 4. How it works | First, forward your invoices to one address. Tallyroom pulls out the supplier, amount, currency and due date, and flags anything that doesn’t match the purchase order. Next, each invoice goes to the right approver, by email or on their phone. Then, once it’s approved, it’s queued for payment and posted to your accounting system. | Step 1: invoice fields highlighted one by one; a mismatch outlined in amber. Step 2: approval card on a phone, tap to approve. Step 3: payment queue and a ledger entry appearing. |
| 5. Proof | Tallyroom connects to the accounting software your team already uses, and keeps a full audit trail of every approval. | Integration icons; an audit log scrolling. |
| 6. Next step | See your own invoices go through it. Book a twenty-minute walkthrough at tallyroom.example. | End card with the action, the URL and the brand. |
The voiceover runs to about 170 words, comfortably inside a 90-second allowance with room for the screen to breathe. In a real script, the proof beat would carry a named customer and a result they have approved for use. We kept it to product facts here because the customer would be invented.
Notice what the example leaves out: the plan options, the mobile app’s settings screen, the company’s founding story. Each is true, and each would take attention the three steps need. For finished videos that follow a similar shape, browse our B2B SaaS explainer video examples.
Writing lines that work out loud
A script is read once by a reviewer and heard hundreds of times by viewers. Knowing how to write an explainer video script mostly means writing for that second audience. Four habits do most of the work, and each has research behind it; the full evidence sits in our piece on explainer video techniques backed by research.
Talk to one person, in second person
Write “you forward your invoices”, and let the voiceover sound like a colleague explaining something at a desk. A meta-analysis of conversational-style instruction found it improved recall (d = 0.30) and, more strongly, the ability to apply what was learned (d = 0.54).[1] Those studies measured instructional material. The mechanism carries straight into an explainer, whose job is also to be understood.
Cut the interesting detour
The founding anecdote, the clever aside, the market statistic that does not bear on the three steps: each feels like colour in a review. A meta-analysis of 39 experimental effects found that interesting but irrelevant material reduced learning, with small to medium effects on retention and medium effects on transfer.[2] If a line does not serve the takeaway, it goes.
Write the cue into the script
A cue only works if it lands on the word it points to. So write it into the visual-note column: “highlight the due-date field” next to the line that says “due date”. The storyboard artist then times the highlight to the voiceover, and nobody has to guess.
Let the voice carry the sentences
Across 61 studies with 7,036 participants, animation outperformed static graphics overall (g = 0.226), and the advantage grew when the animation came with spoken commentary (g = 0.336).[4] For a script, that means sentences belong in the voiceover and the screen carries pictures and short labels. A voiceover that reads out a paragraph printed on screen asks the viewer to do the same job twice.
How to review a script without rewriting it
Script reviews stall when five people each rewrite the lines they would have written. Give every reviewer the same five checks.
- Read it aloud, against a timer. If it overruns, cut a sentence; speaking faster only moves the problem to the voice session.
- Check it against the header. Does the hook describe the named viewer? Could they repeat the takeaway after one listen? Is there one next step?
- Check the facts with product. Product managers confirm each step is accurate; the scriptwriter keeps the wording.
- Clear the proof. Any customer name, logo or result needs written approval before the script locks.
- Sign off once. One named owner merges comments into a single set, approves, and sends the script to storyboarding.
Review the script for truth and timing. Leave the wording to the person who will have to hear it three hundred times.
Terms that come up in these reviews, from scratch voiceover to lower thirds, are defined in our video production glossary.
Scripting for viewers across APAC
A Singapore company’s explainer usually plays well beyond Singapore: in a sales email to Jakarta, on a partner portal in Sydney, at a trade stand in Bangkok. Many of those viewers watch in their second or third language, so the English master needs to be easy to follow and easy to translate.
- Plain sentences, no idioms. “Hit the ground running” survives neither translation nor a non-native listener. Say the literal thing.
- Slack in the timing. Translated voiceover often runs longer than English. A script timed to the second in English will overrun in Bahasa Indonesia or Thai, so leave pauses you can give back.
- Captions on every version. A review of more than 100 empirical studies found that captions improve comprehension of, attention to and memory for video.[5] They are no substitute for clear language: one experiment with online learners found no main effect of subtitles, while proficiency and video complexity both shaped learning.[6]
- Labels you can swap. Keep on-screen words short so they can be reset in another language without redrawing the scene.
Which markets get subtitles and which get a local voice is a separate decision, covered in our guide to multilingual video localisation in Southeast Asia.
Want a script that survives the storyboard?
See how our scripts turn into finished explainers in the portfolio, then tell us about your viewer and your markets. We will draft, time and storyboard it with you.
Talk to our scriptwriters →The short version
An explainer video script decides the video before any pictures exist, so start with three sentences: who the viewer is, what they should repeat afterwards and the one action you want. Build the voiceover in six beats (hook, stakes, shift, how it works, proof, next step) with how-it-works taking about 40 per cent of the runtime. Write in second person, cut every interesting detour, put the on-screen cue next to the line it supports, and let the voice carry the sentences. Read it aloud against a timer, clear the proof, sign it off once, and hand it to the storyboard.
Frequently asked questions
How many words should an explainer video script have?
Plan on roughly 140 spoken words for every minute of runtime, so a 90-second explainer carries about 210 words at most. Most good scripts land a little under that, because the viewer needs short pauses to watch the product on screen. Technical vocabulary and translated versions push the number lower again, so read every draft aloud against a timer.
What is the best structure for an explainer video script?
For B2B technology products, six beats work reliably: a hook that names the viewer's situation, the stakes of leaving it alone, the shift your product makes, how it works in three steps, one line of proof, and a single next step. The how-it-works beat takes the largest share of the runtime, usually around 40 per cent.
Is there an explainer video script template we can use?
Yes. The template in this guide is a two-column table with one row per beat: the voiceover you will hear on the left, and a short visual note on the right. Fill in the voiceover first, keep each beat inside its word allowance, and write the visual note as an instruction for the storyboard artist.
Should the voiceover repeat the words shown on screen?
Keep them apart for most of the video. Let the voiceover carry the sentences and let the screen carry pictures, product views and short labels of two to four words. A meta-analysis of animation research found larger learning gains when animation came with spoken commentary. Reserve full on-screen sentences for captions, the headline claim and the call to action.
Should our team write the script, or the video studio?
Your team owns the facts and the studio should own the words. You supply the audience, the one takeaway, the three steps and the approved proof. A studio scriptwriter turns that into lines written for the ear and timed to the runtime. Nominate one person to consolidate feedback and sign off, so the script arrives at storyboarding with one voice.
How do we write one explainer script for viewers in Singapore, Malaysia, Indonesia and Australia?
Write the English master in short, plain sentences, avoid idioms and local jokes, and leave the timing some slack. Translated voiceover often runs longer than English, so a script timed exactly to 90 seconds will overrun in Bahasa Indonesia or Thai. Keep on-screen text as editable labels, and decide which markets get subtitles and which get a local voiceover.
References
- Ginns, P., Martin, A. J., & Marsh, H. W. (2013). Designing instructional text in a conversational style: A meta-analysis. Educational Psychology Review, 25(4), 445–472. doi:10.1007/s10648-013-9228-0
- Rey, G. D. (2012). A review of research and a meta-analysis of the seductive detail effect. Educational Research Review, 7(3), 216–237. doi:10.1016/j.edurev.2012.05.003
- Schneider, S., Beege, M., Nebel, S., & Rey, G. D. (2018). A meta-analysis of how signaling affects learning with media. Educational Research Review, 23, 1–24. doi:10.1016/j.edurev.2017.11.001
- Berney, S., & Bétrancourt, M. (2016). Does animation enhance learning? A meta-analysis. Computers & Education, 101, 150–167. doi:10.1016/j.compedu.2016.06.005
- Gernsbacher, M. A. (2015). Video captions benefit everyone. Policy Insights from the Behavioral and Brain Sciences, 2(1), 195–202. doi:10.1177/2372732215602130
- van der Zee, T., Admiraal, W., Paas, F., Saab, N., & Giesbers, B. (2017). Effects of subtitles, complexity, and language proficiency on learning from online education videos. Journal of Media Psychology, 29(1), 18–30. doi:10.1027/1864-1105/a000208