Automating Production for Brand Assets

At the Premium Gruppe, I’m currently designing and building an automation platform that connects brand design, production and distribution into one coherent workflow. The goal is straightforward: make high-quality brand assets for streaming platforms faster, more reliable, and easier to scale — without losing craft or consistency.

This project lives exactly where my work tends to end up: between brand design, automation and real-world production pressure.

We produce and maintain hundreds of key visuals across multiple streaming platforms:

• JOYN
• Pluto TV
• Amazon Prime Video
• KDG / Vodafone
• client platforms (welt.de, WELT mediathek and apps)

Each platform comes with its own reality:

• different aspect ratios
• different technical specs
• different deadlines
• constantly shifting program schedules

None of this is conceptually complex — but at scale, it becomes fragile.

Traditionally, a lot of time went into manual checks: availability, resizing, gap tracking, deadline coordination across teams.

The bottleneck wasn’t creative quality. It was orchestration.

I designed a workflow that connects data, logic and creative production into a single pipeline, using n8n as the orchestration layer.

The system is built to:

• Detect missing assets automatically per platform and format
• Generate structured tasks with clear ownership and timing
• Scale, crop and validate existing key visuals based on platform rules
• Trigger fast-turnaround production for missing or breaking content
• Keep delivery states in sync across internal tools and endpoints

Instead of reacting late, the workflow now anticipates gaps before they surface downstream.

Editorial planning, brand design and delivery no longer operate as separate silos — they’re part of the same system.

The tools matter, but only insofar as they stay out of the way.

For orchestration, I use n8n because it allows complex logic to remain readable, adaptable and debuggable — even as workflows grow. For generative visuals and explorations, I integrate APIs like OpenAI / Gemini, combined with internal design systems and validation steps.

The guiding principle is simple: automation should be transparent, not magical.

Especially when brand consistency is involved, teams need to understand why something happens — not just that it does.

This project isn’t about replacing designers. It’s about protecting design quality under scale.

By automating repetitive and error-prone steps, the system creates space for what still requires judgment: concept, typography, composition and storytelling.

Brand systems don’t live in Figma alone. They live in processes, decisions and tools that support those decisions.

In that sense, this workflow treats brand design as a living system — not a static set of rules.

• Brand consistency scales best when logic is explicit
• Designers are well positioned to shape the systems they work with
• Automation isn’t the opposite of creativity — it’s a multiplier
• The best systems fade into the background while raising overall quality

Most importantly: Strong brands aren’t defined only by how they look, but by how smoothly they operate.

I’m deeply interested in how brand design evolves in an automated, AI-assisted production landscape.

The most meaningful work, for me, happens one level above individual assets — where systems, tools and design intent meet. Building this workflow didn’t feel like “process work”. It felt like brand design at a different level of abstraction.

And that’s the direction I want to keep exploring.