Video, without a shoot

A script, three shots, and a voice that reads it.

The creator-style video ad is the format everyone wants and almost nobody can produce weekly. NEXA writes the spot before it films it: the concept, the shot list, the spoken script and the end card are all written in one pass, then the footage is rendered and cut together with the voiceover and burned-in captions.

You need vertical video every week. Hiring a creator per ad does not scale, and a still image is not what the feed rewards.

1. 1

   The director writes the spot first

   Concept, three shots with their own prompts and diegetic audio, a caption per shot, the voiceover script and an end card — written together, so the words and the pictures were planned as one thing.

2. 2

   Your product anchors the opening shot

   The product photo you upload seeds the first shot and is re-framed to the clip's own canvas first, so a square packshot does not open a vertical ad with black bars. Later shots deliberately crop or soft-focus the product so the renderer never has to guess a label it cannot read.

3. 3

   Voice, then captions timed to it

   The voiceover is synthesised with word-level timings, and each phrase's caption is shown for exactly the window in which it is spoken. Captions carry every load-bearing fact, because the feed plays silent.

4. 4

   The format picks the renderer

   Every format renders as one continuous piece, so the same physical product survives from the first frame to the last. The formats built around a hard cut — before and after, problem then solution, a tutorial's steps — keep their cut: it is written into the piece, with the same product, person and light on both sides of it, rather than rendered as separate shots that cannot be trusted to show the same bottle twice.

5. 5

   It is assembled, not stitched by you

   The shots, the caption frames, the end card and the voice track go to a render worker that cuts them and ducks the audio. It arrives as one file, in the format you asked for.

Not an estimate written for this page. Every number below is read from the credit costs the generator itself uses, so this page cannot drift away from the product.

The cheapest plan that covers that month is Business at $129 a month, with 9,900 credits. Nothing on top of it: there is no fee on what you spend on ads.

• There is no avatar library. Three of the twelve formats do put a person on camera, lip-synced in a single pass, but that face is generated fresh for each brief — the same presenter does not come back across your ads, and you cannot upload a creator of your own.
• A generated presenter is a generated person. Meta, TikTok and the FTC all have rules about disclosing that. NEXA can draw an "AI-generated" label on the ad — one checkbox, off by default — but whether and where a platform requires it is still yours to know.
• There is no music library. The audio is the diegetic sound the director wrote and the voiceover it scripted, and licensing a music bed is not something we can do on your behalf yet.
• A directed video is the most expensive thing NEXA makes, by a wide margin. The cost is on the button before you press it.