Short answer
A product demo script template has two columns, what the viewer hears and what they see, and one row per scene with a start and end time. A 60-second demo usually runs six scenes: hook, problem, two or three steps of a real task, the result and one call to action. At roughly 140 to 160 spoken words a minute, plan about 110 to 130 words.
Demo videos go wrong at the script stage far more often than in the edit. The usual cause is a script written as a paragraph of feature claims, with the screens worked out later. This product demo script template forces the opposite: every line of voiceover has a screen next to it, and every screen has a time.
Below you'll find the template itself, a word budget for each video length, a filled example for a made-up invoicing app called Ledgerly, and the writing habits that keep narration and picture in step.
How to use this product demo script template
Write in two columns. The left is the voiceover, exactly as it will be spoken. The right is what's on screen during that line: which page, what the cursor does, what gets zoomed, any text that appears. If you can't fill the right-hand cell, the line is probably a claim with nothing to show, and it should go.
Give every row a start and end time, then read the whole thing out loud at the pace you'll record it. Your first read will run long. Everyone's does.
- One row per scene, and one idea per row.
- Write the voiceover the way you talk: contractions, short sentences, no slashes or brackets left in.
- Name the sample data in the on-screen column (customer names, amounts, dates) so the screens get built to match.
- Leave a beat of silence after the key action. The screen needs a second to land.
The template: copy and fill in
This is the 60-second version. For a 90-second demo, add one or two task steps; for a 30-second clip, drop the problem row and keep one step.
- Hook (0:00–0:05). Voiceover: [the result, in one line the viewer wants]. On screen: [the finished result in the real UI, already full of data].
- Problem (0:05–0:13). Voiceover: [the pain in the viewer's words, one or two sentences]. On screen: [where the work starts today: the cluttered list, the empty form, the spreadsheet].
- Step one (0:13–0:25). Voiceover: [what the viewer does first, and why it's easy]. On screen: [the first action, cursor and zoom on the one field or button that matters].
- Step two (0:25–0:37). Voiceover: [the next action and what it saves them]. On screen: [the second action; show the product doing the work].
- Step three, optional (0:37–0:47). Voiceover: [the step that makes this product different]. On screen: [the distinctive feature in use, not a settings page].
- Result (0:47–0:55). Voiceover: [what they end up with, in concrete terms]. On screen: [the outcome held still long enough to read].
- Call to action (0:55–1:00). Voiceover: [one action, e.g. start a free trial]. On screen: [logo, URL or button, nothing else].
If you'd rather start from a generated draft, the free demo script generator fills this structure from a short description of your product. Edit what it gives you; you know which steps your viewers actually take.
Word budget for each video length
Most narrators land somewhere around 140 to 160 words a minute when reading a script naturally. Treat that as an approximation: some people speak much faster, non-native narrators and AI voices vary, and demo narration needs pauses while the screen does the work. Time your own read rather than trusting the arithmetic.
| Video length | Words at full pace | Words to aim for (with pauses) | Scenes | Typical use |
|---|---|---|---|---|
| 15 seconds | 35–40 | 25–30 | 2–3 | Social teaser, homepage loop captions |
| 30 seconds | 70–80 | 55–65 | 3–4 | Feature clip, ad |
| 60 seconds | 140–160 | 110–130 | 5–6 | Core demo, launch post |
| 90 seconds | 210–240 | 170–190 | 6–8 | Landing page tour |
| 2 minutes | 280–320 | 230–260 | 8–10 | Product tour on YouTube, sales email |
| 3 minutes | 420–480 | 350–390 | 10–14 | Tutorial, founder walkthrough |
If your draft is well over the right-hand number, cut a scene rather than speeding up the voice. Fast narration over fast cuts is the most common reason viewers say a demo felt rushed. And put the key moment early: Wistia reports that viewers tend to drop off around the halfway mark of a video (Wistia). For which length suits which channel, see how long a demo video should be.
A filled product demo script example: Ledgerly in 60 seconds
Ledgerly is a made-up invoicing app for freelancers. The viewer is a freelance designer who has just landed on the homepage. The one job: see that chasing late invoices can be automatic. The voiceover below is 116 words.
- Hook (0:00–0:05). Voiceover: "Get paid on time, without writing a single reminder email yourself." On screen: the Ledgerly dashboard, three invoices marked Paid in green.
- Problem (0:05–0:13). Voiceover: "Late invoices cost you twice: once in cash, and again in the awkward emails you keep putting off until next week." On screen: an invoice list, two rows overdue in red, "Harbor & Pine, $2,400, 18 days late".
- Step one (0:13–0:25). Voiceover: "Open any invoice and switch on auto-reminders. Pick a schedule that suits your clients, or just keep the default." On screen: the invoice opens; zoom on the Auto-reminders toggle as it turns on; schedule reads 3, 7 and 14 days.
- Step two (0:25–0:37). Voiceover: "Ledgerly writes each reminder in your tone and sends it from your own email address, so it reads like you wrote it." On screen: email preview addressed to Harbor & Pine, signed by the freelancer.
- Step three (0:37–0:47). Voiceover: "When the client pays, the reminders stop on their own, and your books update without you touching them." On screen: a payment notification; status flips to Paid; the reminder queue empties.
- Result (0:47–0:55). Voiceover: "No spreadsheets, no sticky notes and no awkward follow-ups. You do the work, and you get paid for it." On screen: dashboard, outstanding total dropping from $5,150 to $2,750.
- Call to action (0:55–1:00). Voiceover: "Try Ledgerly free for 14 days." On screen: logo and the sign-up button.
Notice what's missing: no feature list, no company history, no "in this video we'll show you". Every line has a screen, and the numbers on screen add up ($2,400 paid off $5,150 leaves $2,750).
Writing voiceover that matches the screen
The viewer should never hear about something before they can see it. Write each line so its key noun appears on screen as it's spoken, and cut the picture to the voice, not the other way round. Tools that build the video from the script can do the timing for you: DemoVexa, for example, generates the voiceover first and times each animated scene to the narration's words. If you're screen recording, record the voice first and edit to it.
Say what matters on screen out loud. WCAG 2.2's audio description criterion (Level AA) notes that when the important visual information is already in the main audio, no separate description track is needed (W3C). Narration like "the status flips to Paid" serves blind viewers and people half-watching at the same time.
- Don't narrate clicks ("now I click here"). Narrate why the click matters.
- Write numbers the way they'll be said: "two thousand four hundred dollars" reads differently from "$2,400" to an AI voice.
- Spell product names phonetically in a note for the narrator or voice settings.
- Read the script aloud once more after every edit. Lines that look fine on paper often trip the tongue.
- Keep the final script: it's the first draft of your captions. Captions for prerecorded video are a Level A requirement in WCAG 2.2.
Questions
How many words should a product demo script have?
About 110 to 130 words for a 60-second demo and 170 to 190 for 90 seconds, based on a rough speaking pace of 140 to 160 words a minute with room for pauses. Read it out loud and time it, because your narrator's real pace is what counts.
What's the difference between a demo video script and a live sales demo talk track?
A video script is fixed and timed to the second, with every line matched to a screen. A live talk track is a flexible outline the rep adapts to the buyer's questions, so it lists topics and proof points rather than exact sentences.
Should I write the script before building the screens?
Yes. The script decides which screens, data and actions you need, so writing it first stops you recording or building scenes that get cut. Our [how to make a product demo video](/how-to-make-a-product-demo-video) guide puts the script at step two for that reason.
Make product videos from your code
DemoVexa reads your app's code, writes the script, adds a voiceover and animates your real screens in 16:9, 9:16 and 1:1. Try it on your own product: 14 days free, no card.
Sources
- W3C: Understanding WCAG 2.2 Success Criterion 1.2.5, Audio Description (Prerecorded) (checked 29 October 2026)
- W3C: Understanding WCAG 2.2 Success Criterion 1.2.2, Captions (Prerecorded) (checked 29 October 2026)
- Wistia: How to choose the right marketing video length for any goal (checked 29 October 2026)
DemoVexa publishes this guide and makes a product video app, so we have a stake in the topic. Facts about other products and platforms come from the sources listed, on the dates shown, and can change. Corrections are welcome through our contact page.