Case Study · Brand Systems · AI Generative Workflow
The One Intelligence
Framework
What does it take to make a device ecosystem feel like a single, coherent mind? A deep dive into using AI generative workflows to unify creative output across a fragmented brand system.
A world-class ecosystem.
A fragmented creative reality.
Amazon's device portfolio — Echo, Fire, Ring, Alexa — represents one of the most sophisticated consumer technology ecosystems ever built. But behind the scenes, each product line was managed by separate teams and supported by different agencies. Every group had its own visual logic, its own tone, its own interpretation of the brand.
The result was a brand that looked and felt different depending on where you encountered it — on a shelf at Target, in a paid social ad, at a trade show, on the homepage. The devices were genuinely connected. The creative wasn't.
"The goal was never to make these products feel like they work together. The goal was to make them feel like they were never apart."
Core strategic premiseA traditional campaign solution — a unifying theme, a hero visual — wouldn't hold across teams and agencies operating independently. What was needed was something structural: a way to ensure that no matter who was producing the creative, the output carried one voice and one intelligence.
Fragmentation isn't a brand problem.
It's a workflow problem.
The instinct in this situation is to build a better style guide — clearer rules, stricter templates, more rigorous brand policing. But style guides don't scale across agencies. They get interpreted. They drift. They become the thing everyone says they follow and no one actually does.
The real issue was upstream. Without a shared intelligence informing every creative decision — regardless of who was making it — no amount of guidelines would produce consistency at the output level.
The shift was this: instead of trying to control creative output after the fact, I focused on engineering the input — building a generative framework that encoded the brand's intelligence directly into the workflow. Each team could keep their process. They just needed to be working from the same source.
I didn't design a campaign.
I engineered an intelligence layer.
Ecosystem Audit
Mapped every product line in the Amazon devices portfolio — its current creative positioning, the teams involved, the agencies producing work, and every touchpoint where it showed up publicly. Identified where the disconnects were sharpest: retail, social, experiential, and web were each telling a slightly different story.
Prompt DNA Development
Translated Amazon's brand language, product positioning hierarchy, visual identity, and tone into structured prompt engineering frameworks. Not a style guide that lives in a folder — a living creative brief that could be activated inside any generative AI tool and produce on-brand output without manual oversight on every asset.
Touchpoint Mapping
Defined the specific channels where creative inconsistency was doing the most damage — in-store, paid social, trade shows, product launches, digital. Each channel became an input variable in the workflow, so the generative system could adapt tone and format to context without drifting from the brand core.
Generative Workflow Build
Built and tested a multi-tool AI workflow that could produce on-brand creative across formats — copy, visual direction, campaign concepts — without requiring a central creative team to manually enforce consistency. The system was designed to be repeatable and transferable across any team working on the ecosystem.
Five components.
One coherent system.
This wasn't a pitch deck or a finished campaign — it was a working strategic framework developed to show what an AI-powered orchestration layer could look like in practice. Each component was designed to be immediately applicable if taken into production.
What this work
actually shows.
The Amazon case study demonstrates three things I think matter in brand systems work.
First, that creative fragmentation at ecosystem scale isn't a people problem or a budget problem — it's a workflow problem. When the intelligence informing creative decisions lives inside individual teams and individual agencies, consistency becomes impossible to enforce after the fact. It has to be built into the process.
Second, that prompt engineering is a strategic discipline, not a technical one. The decisions that matter — how a brand speaks, how products relate to each other, what a customer should feel at each touchpoint — are creative strategy decisions. AI is just the tool that makes them scalable.
Third, that the same framework applies to any organization managing multiple products, teams, or agencies at once. The problem isn't unique to Amazon. The solution isn't either.
"I don't just use AI tools — I build the systems that make them work together. The output is consistency. The work is in the architecture."
On AI generative workflow direction