Turning Ideas into Test Plans in 10 Minutes
Reduced experiment planning time from 1–2 hours to 10 minutes by building an AI-powered workflow that translates stakeholder ideas into structured test plans.
THE PROBLEM
Ideas for A/B tests were:
Loosely defined
Missing key details (like hypotheses or goals)
Difficult to implement
Stuck in people’s brains without a place to document them
We needed a way to:
Standardize and scale A/B testing
Ensure each idea was actionable and implementation-ready
Reduce back-and-forth between stakeholders and execution teams
Capture ideas for future implementation
My Role
I owned the experimentation process, so I…
Identified workflow bottleneck
Defined structure of a “good test plan”
Designed input schema for AI
Built and iterated on my agent using AI tools
Operationalized workflow across team
NEW WORKFLOW LOADING
*
NEW WORKFLOW LOADING
*
NEW WORKFLOW LOADING
*
NEW WORKFLOW LOADING
*
NEW WORKFLOW LOADING
NEW WORKFLOW LOADING * NEW WORKFLOW LOADING * NEW WORKFLOW LOADING * NEW WORKFLOW LOADING * NEW WORKFLOW LOADING
The Solution
STakeholder has idea
1
“I wonder if we’d get more clickthroughs if we change the wording of that button?”
Idea captured in asana
2
Stakeholder either fills out Asana form or stakeholder shares idea directly via message or call.
Inputs fed into Agent
3
“We want to A/B test the microcopy in the hero of www.URL.com.”
Agent generates Test plan
4
Required components of the test plan:
Hypothesis
Variants
Target audience
Success metrics
Implementation notes
THE IMPACT
Faster experimentation velocity
Reduced cognitive load for stakeholders
More tests launched (increase in throughput)
Improved cross-functional collaboration
Final thoughts
This wasn’t just a UX project to me.
While wireframes and user flows are important, I didn’t just design a screen—I redesigned how work gets done at my company. This was the first time I felt like I made an impact at home.