How we plan an AI commercial, from brief to final cut
- Production
- Process

Every commercial we make starts the same way: with a conversation about what the film needs to do, not how it should look. The look comes later, and it comes faster when the goal is clear.
Start with the brief
A good brief fits on one page. It names the audience, the one thing they should remember, where the film will run and when it needs to be live. If you have reference films, send them, even if they are from a different category.
We read every brief against three questions before we quote:
- Who is watching? A film for a trade show screen is built differently from one for a phone in a feed.
- What must stay exact? Products, logos and faces need the most care, so we plan them first.
- What can be loose? Backgrounds and atmosphere are where AI production saves the most time.
Building the look
Look development is where the film becomes real. We set the palette, the lenses and the light, then test them on a few key frames before anything moves. You can see the kind of motion study we share at this stage below.
Approvals that keep their word
Each stage ends with a clear approval: key frames, then an animatic, then the first cut. Once a stage is approved we do not reopen it, which keeps the schedule honest for everyone. More on how we review in Trust & Rights.
Timelines and deliverables
Most projects follow one of three shapes. The timings below are typical, from approved brief to final delivery.
| Package | Length | Typical timeline | Rounds of changes |
|---|---|---|---|
| Essential | Up to 60 sec | 2 to 3 weeks | 2 |
| Extended | Up to 180 sec | 3 to 5 weeks | 3 |
| Series | Up to 400 sec | 5 to 8 weeks | 3 per film |
Cutting for social
A commercial rarely lives in one place. We plan the vertical cuts from the start, so framing and type work at 9:16 as well as 16:9.
Review and delivery
Final files go out in every format you need, with captions and safe areas checked. Every frame has been reviewed by a person before it leaves the studio.


