Design Strategy
Metrics Pal — An AI Plugin for UX Analytics & Event Design
Introduction
Design teams are increasingly expected to make data-informed decisions, yet analytics setup remains one of the biggest friction points between design, product, and engineering.
To address this, I designed and built Metrics GPT, an AI-powered plugin that helps UX designers generate consistent, high-quality analytics event plans directly from design context — without needing deep analytics expertise.
The plugin integrates AI into designers’ existing workflows, turning analytics from a downstream task into a first-class design activity.
The Problem
Across large product organizations, analytics work often suffers from:
-
Inconsistent event naming and structure
-
Late involvement (analytics added after design or build)
-
High dependency on specialized analytics partners
-
Low designer confidence in defining “what to measure”
-
Designers want to design measurable experiences — but the tooling and knowledge are fragmented.
Designers don’t need to become analysts.
They need guidance at the moment of design.
My Role
-
Role: UX Design Lead & Builder
-
Timeline: 3 months
-
Tools: Figma, Chat GPT, internal analytics standards, Markdown, Cursor
-
Outcome: Faster event planning, improved consistency, and increased designer confidence in analytics
Goals
Clarity
Help designers define meaningful, well-structured analytics events without requiring deep analytics expertise.
Consistency
Enforce shared naming conventions and logical event structures to reduce fragmentation across teams and features.
Integration
Fit naturally into existing design workflows so analytics becomes part of the design process—not a downstream task.
Partnership
Scale across teams while supporting (not replacing) analytics experts through guided, standards-aligned output.
How it works
-
Design context is captured
The plugin automatically "reads" the screen to get context.
2. AI helps define goals by which all events will roll up to.
-
Event naming conventions
-
Required vs optional properties
-
State transitions
-
Success and failure signals
3. Structured output is generated The plugin produces a clean, standardized tracking plan designers can:
-
Review
-
Edit -- fully editable in figma
-
Share with product and engineering
Validation & Impact
While still an early-stage tool, initial usage showed:
-
⏱️ Significant time savings in creating tracking plans
-
📐 Improved consistency across features and teams
-
💬 Higher confidence from designers when discussing analytics
-
🤝 Better collaboration with analytics and engineering partners
Designers reported that analytics felt approachable rather than intimidating.
.png)