Growth Experiment Designer & Readout
Designs one A/B test with hypothesis, sample size, duration and decision rule, then reads out results with SRM and confidence intervals.
Before & after
Same prompt, same agent — the only difference is the skill.
Churn is a critical metric. There are multiple levers: product quality, onboarding experience, customer success outreach, pricing, competitive pressure. The best approach depends on understanding why customers are leaving. You'd want to segment churn by cohort, product area, or customer segment. What does your churn data show? Have you surveyed customers about their reasons?
Delivers a CHURN REVERSAL PLAYBOOK with ROOT CAUSE ANALYSIS framework: survey lapsed customers, analyze feature usage before churn, segment by cohort. INTERVENTION SEQUENCE: quickest win first, whether pricing adjustment, product stability fix, or onboarding redesign based on data. SUCCESS METRICS: target churn reduction 0.5-2 percent monthly, payback period 6-12 months. CUSTOMER SUCCESS TACTICS: re-engagement cadence, value reset calls, and PRODUCT ROADMAP IMPACT: how to prioritize features that stop churn. Returns a 90-DAY EXECUTION PLAN with owner and KPI per initiative. Chains with other skills. Structured workflow for consistent strategy.
About this skill
Stop shipping changes on noisy numbers. This skill designs one experiment with a sample size your traffic can support, then reads it out against a decision rule set before launch.
What it does
- Writes an evidence-based hypothesis with one primary metric and guardrails
- Calculates sample size and duration from baseline, MDE, alpha, and power
- Flags underpowered tests and offers alternatives
- Sets randomization, exposure, no-peeking, and Ship/Kill/Iterate rules before launch
- Reads out with sample ratio mismatch check, confidence intervals, and guardrail results
What you get
An experiment brief (hypothesis, metrics, sample size, duration, decision rule) and a readout with data quality checks, results table, decision, learnings, and the next test.
Best for
Growth, product, and lifecycle marketers running A/B tests on signup, onboarding, pricing, or email.
SKILL.md preview
---
name: growth-experiment-designer
description: Design one growth experiment properly before it runs - hypothesis, primary and guardrail metrics, minimum detectable effect, sample size and duration, randomization and exposure rules, and a decision rule - then read out the results with the right statistics and a ship / iterate / kill call. Use when the user says "set up an A/B test for X", "how long should this test run?", "is this result significant?", "read out our pricing page test", or wants to stop shipping changes based on noisy numbers. Also triggers on "how long should this test run?", "is this lift significant?", or "what sample size do I need?".
---
… (sign up to view the full skill)More marketing & growth skills
View all Marketing & Growth skills →Paid Ads Campaign Planning and ROAS/CPA Diagnosis
Plans paid campaigns on Google, Meta, LinkedIn and X from break-even math, writes ads, and finds the metric behind a ROAS or CPA drop.
Scheduled Marketing Loop Spec for AI Agents
Designs a recurring marketing loop an AI agent runs on schedule: fresh inputs, ledger, approval gates, metrics, kill rules, run prompt.
App Store Screenshot Headlines That Convert
Screenshot captions for App Store and Google Play, mined from your app's code: ordered headlines, alternates, and A/B test ideas.
Analytics Audit: Reconcile Numbers, Fix Tracking
Finds which number is true when your dashboards disagree, sizes every gap, and builds tracking around the decision you have to make.