A/B Test Setup & Experimentation

•
642
•

Designs, calculates, and evaluates rigorous A/B and split tests for ad creatives, landing pages, and marketing campaigns.

The 19-ab-test-setup skill equips performance marketers and growth teams with a disciplined framework to execute high-confidence split tests across landing pages, ad creatives, pricing models, and email copy. By calculating sample sizes, enforcing single-variable testing, establishing test duration guardrails (e.g., minimum 7-day cycles to avoid weekly seasonality bias), and setting statistical significance thresholds (p-value < 0.05), this skill helps eliminate false positives and gut-based decision-making. It guides you from hypothesis creation and tracking setup (using Meta, GA4, or PostHog) to post-test analysis and cumulative logging, ensuring you test the highest-impact elements first.

Key Features

01Statistical significance and decision-tree guidelines to prevent premature peaking and false positives
02642 GitHub stars
03Sample size and duration calculator based on baseline conversion rate and Minimum Detectable Effect (MDE)
04Cumulative test log structure to document learnings, winners, and baseline iterations over time
05Creative testing roadmaps for ad hooks (0-3s), formats, angles, and audience segments
06Rigorous single-variable testing rules and variable prioritization matrix (hook vs. LP vs. CTA)

Use Cases

01Calculating sample size and test duration for landing page redesigns and CTA modifications
02Analyzing A/B test results to verify statistical confidence before committing marketing budget
03Setting up ad creative split tests to identify top-performing 3-second hooks before scaling spend