How I Made a $1M Oakley x Meta Commercial (Full Breakdown)
Offers a practical asset-baking framework for commercial creators, illustrating how pairing an LLM for multi-angle prompt writing with dedicated character and location sheets preserves visual consistency across complex scenes.
What this lesson covers
This breakdown demonstrates how to produce a multi-scene spec commercial for Oakley and Meta using generative AI pipelines. The creator details the pre-production process, showing how to organize mood boards, draft prompts with LLMs, and bake character and location reference sheets.
Key takeaways from the creator
AI-extracted notes, not independently verified product claims. Timestamp links let you check each point in the original video.
- 03:32 ↗
Map visual references and storyboard concepts onto a digital canvas mood board before generating any video or image assets.
- 04:37 ↗
Prevent character morphing across varying scenes by creating 360-degree multi-panel character sheets prior to video generation.
- 04:50 ↗
Use an LLM like Claude to translate plain-English descriptions and image references into structured multi-panel generation prompts.
- 05:30 ↗
Generate high-texture character sheet assets in a 16:9 aspect ratio using models specialized for realistic detail, such as Sea Dream 5.0 Pro.
- 07:46 ↗
Reduce generation credit consumption by utilizing cost-effective models like Soul Cinema for environmental plates requiring natural film grain and lighting.
Workflow outlined in the video
- Collect visual references and layout the commercial narrative onto an infinite canvas mood board.
- Input subject photos, wardrobe references, and product specifications into Claude to generate multi-panel character sheet prompts.
- Render 16:9 character sheets in Sea Dream 5.0 Pro using uploaded face and outfit references.
- Provide environment descriptions or rough visual sketches to Claude to generate targeted location prompts.
- Generate 16:9 background environment plates using budget-friendly atmospheric image models to lock in location assets while preserving generation credits.
Before you use this workflow
These notes describe the source video at its publication date. Model access, pricing, connectors and interfaces may have changed. Check the original source and the provider’s current documentation before installing an add-on, connecting an account or spending credits.
ReelStack has not independently tested this workflow. Preview one representative shot and check motion, continuity and output quality before applying it to a full production. No result, cost saving or model capability is guaranteed.
Explore the tools
How to interpret the numbers
55,920 views captured 30 October 2026. ReelStack recommendation score: 44/100. Formula: radar-v5. This is ReelStack’s calculation, not a YouTube rating or a measure of factual accuracy.
Show the calculation
- 65% performance: 0.43× current views vs the channel's recent median; same-age history is not yet available. View-evidence factor 98% (views / (views + 1,000))
- 25% freshness: 76/100 with a 14-day half-life
- Momentum pending: collecting daily snapshots; its 20% weight goes to observed performance, not free points
- 10% engagement: 62/100; likes + 4× comments, smoothed with a 500-view neutral prior