Step 1 — Pick your test type
A/B conversion test inputs
Your current CVR — what control performs at
Smallest relative lift you care about (10% = lift baseline by 10% relative)
95% standard; 90% acceptable for low-risk tests
80% standard; 90% for higher-stakes tests
Free training · Experimentation discipline
From hunches to validated learning
A good test design saves months. A bad one costs months. RGM training walks through hypothesis writing, prioritization (ICE/RICE), power analysis, and how to read results honestly. Free, no signup.
Start the training →Methodology & sources
- Lewis, R. A. & Rao, J. M. (2015). "The Unfavorable Economics of Measuring the Returns to Advertising." Microsoft Research / NBER.
- Standard sample-size formulas via Evan Miller's A/B test calculator methodology.
- Meta Conversion Lift documentation.
- Google Conversion Lift documentation.
- Common Thread Collective & Tinuiti incrementality testing case studies.
- Croll, A. & Yoskovitz, B. (2013). Lean Analytics. O'Reilly.