UXY1 is the internal operating system for how we diagnose ecommerce revenue problems, decide what matters, run experiments, and turn outcomes into reusable learning.
It is built for weekly execution, not for passive documentation.
The repo exists to keep revenue work structured:
DATA -> DIAGNOSIS -> MATRIX -> PRIORITY -> EXPERIMENT -> VERDICT -> LEARNING -> NEXT
What It Is Not
This is not an analytics tool.
This is not an A/B testing engine.
This is not a BI warehouse.
This is not a random ideas backlog.
We use existing best-in-class tools for data collection, behavior analysis, and experiment delivery. Our core value is diagnosis, prioritization, experimentation logic, and learning.
Core Operating Loop
1. Data
We review the current data layer to understand where revenue is leaking.
2. Diagnosis
We turn raw observations into a short list of actual revenue problems, not vague opinions.
3. Matrix
Every idea, finding, and expert intuition becomes a structured matrix node.
4. Priority
We rank nodes by impact, confidence, and effort so the team works on the right things first.
5. Experiment
Selected nodes become formal experiments with a clear hypothesis, metric, guardrails, owner, and rollout plan.
6. Verdict
Every test ends with a decision: KEEP, KILL, ITERATE, or INCONCLUSIVE.
7. Learning
Each finished experiment creates reusable learning, not just a local outcome.
8. Next
The learning feeds the next diagnosis and next set of priorities.
What We Use vs What We Build
What we use
- GA4 for traffic, funnel, and conversion signals
- Shopify for commercial truth, product performance, cart, checkout, and repeat purchase data
- Meta Ads for acquisition quality and spend context
- Microsoft Clarity and Hotjar for behavior evidence
- Optimizely and Statsig for experiment execution
What we build
- diagnosis logic
- revenue matrix
- priority engine
- experiment decision rules
- learning system
- structured human and AI input flow
We do not rebuild commodity tooling.
We use best-in-class external tools for analytics and experimentation.
We build the layer that decides what to test, when to test it, and what to learn from it.
Repository Structure
uxy1-revenue-os/
├── README.md
├── docs/
├── sop/
├── matrix/
├── experiments/
├── learning/
├── templates/
├── data/
└── automation/docs/
System framing, philosophy, decision logic, and stack rules.
sop/
Operational procedures the team can use tomorrow.
matrix/
The live decision layer: active nodes, backlog, and archive.
experiments/
Experiment specs in progress and completed experiment records.
learning/
Structured wins, losses, and recurring patterns.
templates/
Copy-ready working templates for diagnosis, matrix nodes, experiments, learnings, and weekly reviews.
data/
Structured notes and insight summaries from GA4, Shopify, and ads platforms.
automation/
Manual-first placeholder for future automation.
How the Team Should Use It Weekly
Monday or start of week
- review data notes
- refresh diagnosis
- update matrix
- rank top priorities
- confirm the next 3 to 5 active items
During the week
- launch new experiments from selected nodes
- update active experiment status
- log anything that changes confidence or expected impact
- prevent new ideas from living outside the matrix
End of week
- make verdicts on finished tests
- capture wins, losses, and patterns
- archive completed nodes
- set the next priorities using learning, not memory
Working Rules
- Revenue first, not vanity metrics
- One leak at a time
- No redesign-first thinking
- No idea exists outside the system
- Human intuition is allowed, but it must be formalized
- AI helps generate options and structure thinking, but the system controls decisions
Why This Repo Matters
The moat is not having more tools.
The moat is knowing which leak matters, how to rank work, when to test, and how to compound learning across clients and properties.