Vazen Software Launch Animation by Andrey Ramirez Vazen Software Launch Animation by Andrey Ramirez

Vazen Software Launch Animation

Portfolio project
Andrey Ramirez
Verified paid work

Overview

I designed and animated a one-minute explainer for Vazen, a retail data AI platform, that shows how brands and retailers grow a category together. The video moves from the question every category faces to Vazen's four products: Coplan, Studio, Testing and Ask Vazen.

The challenge

Vazen had to explain an abstract product, shared retail data put to work by AI, in about a minute. The film needed to open on something every brand and retailer recognises: the shelf, the categories on it and the question "How do we grow?"
From there it had to introduce a platform that brings brands and retailers into one room, then give each of the four products a clear moment. It also had to feel warm and human rather than technical, which shaped the decision to lean on old-fashioned brand imagery.

Approach

I built the story as a storyboard in Figma first, then animated it in After Effects scene by scene against the voiceover.
Opening. A fast collage of real shopping moments (milk, coffee, snacks, sauces, pet food, laundry) sets up the line "Every category has the same question."
Brands in one room. A grid of vintage-style TV commercials sits beside a black-and-white grocery shot, so the brand side and the retailer side share the frame.
The data. The grid breaks into data squares, then the words Retailer, Brand and Data build up and feed into the Vazen logo.
The products. Coplan, Studio, Testing and Ask Vazen each get their own scene, with product UI, planograms and a typed prompt.
The close. "The job is done when the shelf changes" leads to the shopper and the category growing, ending on the Vazen logo and "The Retail Data AI Platform."

Iterating with the client

The film went through several feedback rounds with Vazen's team. Each one pushed it simpler, warmer and easier to read at speed.

Solving the footage budget

The vintage look first came from real archive ads, and licensing them would have cost $5,531 for one year of web use. I rebuilt that footage for about $14 in AI generation, plus clips from an existing Envato subscription.
The problem. The edit used nine Getty clips and ads ($5,270) and eight iStock items ($261). The Getty material was also editorial-only, showed real third-party brands and would have needed extra permissions for a marketing film.
Category footage. I replaced the iStock shots with Envato Elements clips already covered by the subscription: dairy, coffee, pet food, laundry, sauces and snacks.
Brand commercials. Instead of real brands, I invented them. Using Flora, I generated frames for fictional products in the style of 1950s to 1970s TV ads, such as Koko-Roo chocolate milk, Dobson's Wheat Squares and Old Harbor coffee, plus modern ones like Mondo Crisps and Volto. I then animated the strongest frames into short commercials.
Testing models. I compared image models side by side, and for the hero shot animated the same frame in two video models, keeping the sharper one.
The hero shot. The large retailer tile went through a full-colour Marlow's Oat Flakes ad before landing on an AI-made black-and-white grocery scene that matches the newsreel feel of the original.
The result keeps the vintage charm the client asked for, with no third-party brands on screen and nothing left to license.

Sound

The mix pairs a new voiceover with an upbeat retro funk track and a full pass of UI sound design.
Sound design. Whooshes, pops, ticks and data sounds from the Ocular Sounds Vector library mark each tile, transition and logo reveal, timed so each sound's peak lands on the motion.
Music. The track sits under the voice and rises at the open and the close.
Levelling. Every voice line is matched to -16 LUFS, effects sit in a consistent range by type, and the full mix measures -16.7 LUFS with peaks at -1.6 dBFS, ready for web playback.

Outcome and what I learned

The final film turns an abstract data platform into a short, warm story, and it was delivered with no footage left to license.
Price the footage early. Archive ads looked perfect in the edit but cost far more than the budget, and their editorial-only licences ruled them out for marketing. Checking cost and rights at the storyboard stage would have saved a round.
Invented brands keep the charm without the risk. Fictional vintage products gave the client the nostalgic look they wanted while staying clear of real trademarks.
AI works best as a production tool. Generating many options cheaply, then choosing and grading them inside the edit, gave more control than searching stock libraries.
Small, frequent previews kept feedback concrete. Short renders of single scenes let the client react to specifics instead of the whole cut

0 likes · 5 views · Nov 16, 2026

Animated explainer for Vazen, a retail data AI platform: vintage-style invented brand ads, archive grocery scenes and clean UI show how brands and retailers gro

Timeline

Sep 16, 2026 - Nov 16, 2026

Tags