What problem does it solve?
Design, plan, and execute statistically valid A/B tests and growth experiments to determine which approach drives meaningful improvement, while building a repeatable experimentation program that scales with your product.
Core Features & Use Cases
- Hypothesis framework: Structure tests with clear observations, proposed changes, expected outcomes, and measurable success criteria.
- Test type guidance: Supports A/B, A/B/n, MVT, and split URL tests across pages, features, or flows to match complexity and traffic.
- Sample size & duration guidance: Provides structured guidance on required samples, test duration, and when to stop or extend experiments.
- Metrics framework: Defines primary, secondary, and guardrail metrics to align experiments with business value and risk controls.
- Variant design & testing best practices: Offers guidance on what to vary (copy, layout, CTAs, sequencing) and how to allocate traffic for reliable results.
- Documentation & playbooks: Encourages thorough documentation of hypotheses, results, learnings, and reusable patterns for future tests.
- Templates & cadence: Includes templates and recommended cadences to sustain a continuous growth experimentation program.
Quick Start
Identify your test context and baseline metrics, then write a hypothesis using the Because [observation], we believe [change] will cause [outcome] format and define a primary metric to measure.
