15, 30 or 60 seconds: how long should your AI brand film be?

One wheel, three different films

Imagine a small bicycle workshop that wants a film. A front wheel turns on a stand, a hand checks its movement, and the bicycle waits by the door. We could make those images into a quick invitation to visit. We could also explain how the workshop approaches a repair, or follow the owner through a working morning. Those are three separate promises to a viewer, even when the location stays the same.

The first question I ask is what must still be clear after the screen goes dark. Perhaps it is simply that the workshop welcomes everyday riders. Perhaps a customer needs to see who handles their bicycle. Writing that answer in one sentence gives us something more useful than a request to make the film as long as possible within a budget.

Choose the question before the number

Fifteen, thirty and sixty seconds are useful lengths to discuss, not instructions from an algorithm. A visitor on a service page may want an answer that a passing social viewer has not asked yet. I would check the actual placement and its current technical limits separately, then choose what this particular film is responsible for explaining.

Here are three possible briefs for our invented workshop. They are alternatives for discussion. A short version does not automatically sell better, and a longer version does not automatically earn trust.

Spend seconds on understanding

For a first thirty-second sketch I might allow six seconds to notice the wheel, twelve for the work, eight for the result and four for an invitation. These are made-up planning values. They tell us what to try, not how a finished film must be edited. The turning wheel may need less time; the hand action may need more. A still picture cannot prove that either movement will work.

Use the worksheet below to change that allocation. If the total exceeds the selected length, something needs a decision. Remove an idea, shorten a moment that is already understood, or choose a longer version. Speeding up every action can hide the very care that the workshop wanted to show. If the total is short, the unused seconds can remain unused; the film does not owe anyone filler.

Watch for the moment you stop understanding

I would test a rough sequence before asking for polished shots. Put simple frames in order and let each stay for its proposed duration. Ask someone who has not heard the brief what happened and what the workshop is inviting them to do. Record the question they actually ask. A repeated view from the person making the film is a different observation from a first view by a possible customer.

Listen with the picture, too. A narrator may finish a sentence while the viewer is still trying to identify a tool. Another language can change the timing. I would read each translated line aloud and check it with its own captions instead of assuming that the English cut sets the pace for everyone. An image, a voice and a caption can share a second without sharing the same amount of attention.

• Mark where an action becomes understandable, rather than counting cuts
• Read important on-screen words on an ordinary phone
• Check the ending at normal playback speed, without explaining it first

A longer file is not a larger story

AI generation adds a practical distinction: a clip that lasts eight seconds does not necessarily contain eight seconds I want to use. The useful action might happen near its middle. I plan the edit around accepted moments and discuss how to obtain them; I do not treat a generator duration setting as the length of the final film.

If the workshop needs to demonstrate a real repair, the factual action deserves a suitable recording or another verified source. An imagined hand and wheel cannot certify that a service was performed correctly. The AI scene here helps discuss an invitation and mood. That boundary matters at fifteen seconds just as much as at sixty.

Bring one uncertainty to the brief

For SOLONSKY Films, a useful starting note would say: we want a customer to understand what happens at drop-off, the main version will live on our service page, and we are unsure whether the explanation needs thirty or sixty seconds. Add what must remain real, who approves the film and which shorter version is actually needed. That gives the production conversation a concrete problem to solve.

Keep the timing note as a draft and revise it after the rough sequence. If viewers understand the process but cannot find the next step, work on the ending. If they remember only the attractive wheel, work on the connection between the picture and the message. Choosing the runtime is one decision in that work, not a promise that attention, bookings or sales will follow.