User Archetype Framework for Google Merchant Center
A foundational research synthesis that mapped every user type in Google's business tools ecosystem, including AI agents, giving cross-functional teams a shared language for product decisions that is actively used across Google today.
The Challenge
Google's business tools teams lacked a unified, comprehensive understanding of who they were actually building for. Different product teams held different mental models of their users, making cross-functional alignment difficult and leading to product decisions that optimized for some user types at the expense of others.
The challenge was compounded by the emergence of AI agents as active participants in Google's business tool workflows, a user type that existing frameworks didn't account for, but that would increasingly shape how the products needed to be designed.
Project Goals
- arrow_rightBuild a comprehensive, evidence-based map of the full user landscape
- arrow_rightCreate shared language that could align product teams across Google
- arrow_rightExplicitly account for AI agents as a distinct user type alongside humans
- arrow_rightDeliver a durable framework, not a snapshot, but a living tool
Methodology
Rather than conducting new primary research, this project synthesized the extensive body of existing research across Google's business tools teams, ~55 reports spanning multiple products, time periods, and research methodologies, into a single coherent framework.
Research Audit
Systematically reviewed ~55 existing research reports across Google Merchant Center and adjacent business tools, cataloging user types, behaviors, needs, and gaps.
Pattern Synthesis
Applied thematic analysis across the corpus to identify recurring user types, behavioral patterns, and mental models that appeared consistently across multiple studies and product areas.
AI Agent Modeling
Developed a framework approach that explicitly modeled AI agents as distinct user types, anticipating how agentic AI would change the product design requirements before that shift became the norm.
Stakeholder Validation
Collaborated closely with a Google Staff UXR to validate the framework's structure and ensure it mapped accurately to the lived experience of Google's product teams.
Key Learning
A framework's real value is the shared language it creates, not the archetypes themselves.
I came into this project knowing that archetypes are tools for alignment. I left with a much sharper understanding of what that actually means in practice.
Synthesizing roughly 55 existing research reports required finding signal across wildly different methodologies, time periods, and product contexts. That process pushed me to develop a more disciplined approach to secondary research synthesis, and to get comfortable making judgment calls about what patterns were robust versus what reflected a single study's limitations.
Collaborating closely with a Google Staff UXR also pushed my thinking in new directions. Having a senior partner who knew the product space deeply meant I had to articulate my reasoning clearly and defend my synthesis choices, which made the framework stronger.
The most forward-looking part of the work was modeling AI agents as distinct user types alongside humans. At the time, that was an unusual design choice. It required reasoning about how agentic AI would change product design requirements before that shift became the norm. The fact that the framework is still actively used across Google teams today suggests that framing held up.
Outcome
The framework was adopted across multiple Google product teams and is actively used to guide product decisions and align cross-functional teams. It gave teams a shared language for reasoning about their users, including AI agents, and directly influenced product roadmap decisions across Google's business tools ecosystem.