How to Use ChatGPT to Create a SaaS Demo Video

How to use ChatGPT to create a SaaS demo video is a specific workflow question -- and the answer is more concrete than most guides admit. ChatGPT's role is to generate a structured scene-by-scene script from your product's one-liner. SlickVid's role is to turn that script into an animated 1080p video with auto-zoom, scene transitions, and stat panels. Neither tool replaces the other, and the handoff between them is a single paste step. This guide shows the exact prompt, a worked example, and the copy-paste scene syntax you need.

If you are still deciding which tool to use for the video production side, see the best demo video maker comparison for the full breakdown.

What ChatGPT actually does in this workflow

ChatGPT is good at one thing here: taking a one-sentence product description and expanding it into a structured script with discrete scenes, clear headings, and short punchy copy. It removes the blank-page problem. What it cannot do is record your screen, animate a zoom on a click, render a video file, or add a progress bar scene. Those are all SlickVid's job.

The specific output you want from ChatGPT is a list of scenes with: a scene type (title card, screen recording, stat panel, CTA), a heading, and one or two sentences of narration or body text. That maps directly to SlickVid's markdown editor format.

The exact prompt to use

Paste this into ChatGPT, filling in the two brackets:

Write a 90-second SaaS demo video script for [your product name].
The product does: [one sentence describing what it does and who it's for].

Structure it as four scenes:
1. Title card -- state the problem the viewer has right now (1-2 sentences)
2. Screen recording -- describe what the viewer will see on screen (2-3 sentences of narration)
3. Stat panel -- show one key result or metric with a label and value
4. CTA -- what to do next (one short sentence)

Format each scene with a heading on its own line followed by the body text.

This prompt works because it constrains ChatGPT to a format that maps directly to SlickVid scenes. The four-scene structure covers the essential demo arc: problem, product, proof, action.

Real example: one-liner in, script out

Here is a real ChatGPT run with a fictional invoicing SaaS:

Input one-liner: "QuickInvoice lets freelancers send branded invoices and collect payment in under 60 seconds, with no accounting software required."

ChatGPT output (condensed):

Scene 1 -- Title card
# You're still chasing invoices manually
Most freelancers spend 20 minutes per client just on invoicing.
QuickInvoice cuts that to under 60 seconds.

Scene 2 -- Screen recording
# Create and send an invoice in one step
Select a client, enter the amount, and hit send.
QuickInvoice generates a branded PDF and a payment link automatically.
No templates to set up. No accounting software required.

Scene 3 -- Stat panel
# 847 freelancers
Get paid 3x faster on average

Scene 4 -- CTA
# Start free today
No credit card required.

That is paste-ready into SlickVid with one structural change: swap the "Scene X" labels for SlickVid's scene separator syntax (covered in the next section). The headings and body copy stay exactly as written.

Pasting the script into SlickVid

Open SlickVid's editor and replace the placeholder markdown with your script. The only change you need to make is replacing ChatGPT's "Scene X --" labels with SlickVid's separator lines. The headings and copy stay exactly as written.

For the screen recording scene, record your screen using SlickVid's built-in recorder first, then reference the recording URL in the --- video --- block. Every click in the recording auto-zooms -- no manual keyframing required.

SlickVid scene syntax reference

Here is the full script from the example above, formatted for SlickVid's markdown editor:

--- stage ---
size: 1920x1080

--- text ---
# You're still chasing invoices manually
Most freelancers spend 20 minutes per client just on invoicing.
QuickInvoice cuts that to under 60 seconds.

--- video ---
duration: 20s

![QuickInvoice demo](https://your-recording-url-here.mp4)

--- panel ---
variant: stats

# 847 freelancers
Get paid 3x faster on average

--- text ---
variant: bounce

# Start free today
No credit card required.

The four separator types used here -- --- text ---, --- video ---, --- panel --- -- cover the most common demo structure. For animated text reveals use variant: typewriter or variant: bounce under the separator. For a progress bar scene (useful for showing an onboarding funnel or a percentage metric) use --- progress ---. SlickVid parses the markdown in real time so you see a live preview as you type.

Common questions