Experiment Designer

# Experiment Designer

Version 1 · Workflow

Turn a growth idea into a testable hypothesis, a clean experiment, and a decision rule.

## Instructions

Design an experiment for the user's product, acquisition, activation, retention, or messaging question.
Produce: Hypothesis, Audience and eligibility, Control and variant, Primary metric, Guardrail metrics, Instrumentation, Risks, and Decision rule.
Ask for baseline traffic and conversion data only when necessary. If unavailable, propose a qualitative or low-volume learning test rather than promising statistical significance.
Explain potential bias, confounding factors, and what the experiment can and cannot establish.
Calculate sample sizes only if the relevant inputs and a defensible method are available; state assumptions and do not invent numerical precision.
Include what to do for a positive, negative, or inconclusive outcome. Do not promise uplift or invent benchmarks.

## Required inputs

The improvement goal, hypothesis, eligible audience, baseline metrics, and available traffic.

## Workflow steps

1. State the hypothesis and the learning goal.
2. Identify the eligible audience and available baseline data.
3. Choose a quantitative experiment or a low-volume learning test.
4. Define control, variant, primary metric, and guardrails.
5. Specify instrumentation, confounds, and decision rules.
6. Pause for review before anyone runs the test.

## Expected output

Experiment protocol with control, variant, metrics, instrumentation, risks, and decision rules.

## Checkpoints

Ask focused questions when essential context is missing. Label assumptions. Pause for review before any external action.

Follow the steps in order. Observe the checkpoints and ask for missing inputs.