We created a full product campaign using AI
One phone photo of a can. That is all we had to build a full launch campaign for Pulpa, an energy drink: product shots, lifestyle images, posters and a video ad. Here is how we did it, and what makes it work.
What we started with
One photo, taken on a phone: the can in a hand, on a plain floor. No studio, no photographer, no props.
What came out of it
- Product shots: the can on its own, in any setting, light or angle.
- Lifestyle images: people enjoying the drink, in places that were never shot.
- Posters: ready-to-post ads with the headline built in.
- A launch ad: a video generated shot by shot.
How it works, step by step
- Lock the product. The real photo becomes the reference. The label, the colours and the proportions of the can are fixed before anything else is made.
- Design the pieces. The people, the places and the props are created as stills first, so a face or a location stays the same from one image to the next.
- Generate and art direct. Each image is generated against those references, then selected and refined the way a photographer would edit a shoot.
- Animate. For the ad, each shot is generated from the same pieces, then cut together with sound and pacing.
When it makes sense
- You are launching a product and need visuals before a shoot is possible.
- You need many formats at once: social posts, stories, ads, posters.
- You want to test several directions before committing to one.
- You need new seasonal or flavour variations without shooting again.
Limits to keep in mind
AI does not remove the need for judgement. Small text on packaging still has to be checked image by image, and not every result is usable: the work is in choosing, correcting and keeping the brand consistent. A clear art direction matters as much as with a traditional shoot.