Free A/B Testing Tools
Ten calculators and checklists for planning, running and reading experiments. Everything works in the browser with no signup and no email wall. If you use Optimizely, you can optionally connect your account to auto-verify experiment settings and import live traffic.
Experiment QA Checklist
Pre-launch checklist with 27 validation items across 7 categories. Connect Optimizely to auto-verify experiment settings.
A/B Test Sample Size Calculator
Calculate required sample size and estimate how long your experiment needs to run. Import live traffic data from running experiments.
Statistical Significance Calculator
Enter visitors and conversions per variation to get a two-tailed p-value, a confidence interval on the lift, and a plain verdict. Corrects for multiple variations and flags samples too small to trust.
Experimentation Velocity Calculator
Find out how many A/B tests your team can realistically run per year. Enter traffic, conversion rate, and MDE to get your velocity benchmark.
Stats Method Picker
Not sure which Optimizely stats engine to use? Answer 5 questions and get a personalized recommendation.
Sample Ratio Mismatch (SRM) Checker
Enter observed visitors per variation and get a chi-square p-value that tells you whether your traffic split is broken — the essential pre-analysis validity check.
CUPED Variance Reduction Estimator
Estimate how much traffic and time CUPED saves your experiment from the correlation between pre- and in-experiment metrics. Variance reduction = correlation squared.
Feature Flag Naming Convention Generator
Describe a flag and get a clean, consistent, searchable key in your chosen convention — across dot, kebab, snake, and screaming styles — plus naming discipline that scales.
A/B Test Prioritization Scorer
Rank your test backlog with the ICE or PIE framework. Score each idea on impact, confidence, and ease and get an ordered list, so you run the highest-leverage tests first.
Bulk Targeting Simulator
Run thousands of user profiles through an Optimizely datafile and inspect variation splits, audience matches, and rule funnels without firing impressions.