# Slide Outline: A/B Testing At Scale Workshop

## 1. Workshop Goal

- Turn an experiment portfolio into a risk-tiered review-board operating model.
- End with intake fields, metric contracts, launch checks, readout order, and archive schema.

## 2. Scale Creates Governance Problems

- Too many metrics invite metric shopping.
- High-risk tests need human judgment.
- Low-risk tests should not drown in process.
- Archives should preserve learning, not just winners.

## 3. Risk Tiers

- Low-risk fast lane
- Medium-risk review
- High-risk cross-functional review
- Automatic block conditions

## 4. Metric Contracts

- Primary metric
- Guardrails
- Minimum detectable or worthwhile effect
- Validity checks
- Decision rule

## 5. Team Exercise

- Define three risk tiers.
- Write intake fields.
- Choose one automatic launch blocker.
- Draft the archive schema.

## 6. Peer Critique

- Which experiments should bypass review?
- Which require human review?
- What would create rubber-stamp governance?

## 7. Close

- Share one low-risk fast lane and one high-risk redesign case.