CRO Toolbox

A/B switcher, Datalayer reader, test analyser, MDE calculator & SRM checker for Convert.com, VWO, Varify, Kameleoon, Optimizely

CRO Toolbox is a Chrome extension for CRO consultants, online marketers, and anyone involved in A/B testing. It opens as a side panel next to the website you're working on and gives you five tools to support your testing workflow — from QA on live sites to planning and analysing experiments. V1.0.0 New name: CRO Toolbox — previously known as A/B Experiment Variant Switcher, the extension has grown well beyond variant switching, so it has a new name to match. It now opens as a side panel that stays open while you browse, rather than a small popup that closes the moment you click away. A/B Switcher - rebuilt from the ground up — now shows experiment names and variant names instead of just technical IDs, making it much easier to know which test you're looking at. Also detects when something is a live deployment rather than an actual A/B test, so you won't accidentally try to switch a variant that doesn't exist. Now supports Convert.com, VWO, Varify, Kameleoon, and Optimizely DataLayer Reader — inspect the datalayer in real time. Live mode streams new events as they're pushed, with automatic tagging for GA4, ecommerce, and GTM events. Filter by keyword, group events by type, and keep a per-page history so you can see exactly what fired where — useful for QA'ing tracking before, during, and after a test. A/B Analyzer — enter your visitors and conversions and get a clear answer on whether your test result is significant. Choose between two analysis approaches depending on how you like to work. Shows the probability that your variant is winning, the expected uplift, and a plain-language interpretation of the result. MDE Calculator — tells you the smallest improvement your test would be able to reliably detect, week by week, based on your site traffic. Helps you decide how long to run a test before the results are meaningful. SRM Checker — checks whether your test traffic was split as intended. An uneven split can make results unreliable even if the numbers look promising. Paste in your visitor counts and get an instant verdict with a per-variant breakdown. Sample Size Calculator — figure out how many visitors you need per variant before you start, so you don't end your test too early or run it longer than necessary.