The A/B testing and conversion rate optimization platform for agencies. Split-test any page, track Stripe-verified conversions and revenue by variant, and know how much traffic you need before you look at results. Every client keeps its own project, so their experiments, revenue, and Search Console data never mix with anyone else's.
No credit card required · or see a live split test in action →
How it works
Three steps, no separate CRO platform, analytics tool, and spreadsheet to keep in sync.
01
Drop a single script tag into the page you want to test. No redirect-based split URLs, no separate hosting for variants: Kitchen Lab swaps content in place on your existing page.
02
Describe what changes between A and B (and C, if you're testing more than one idea), set a conversion goal, and Kitchen Lab handles the traffic split automatically.
03
Watch conversions, revenue, and statistical significance update live, and stop before you're tempted to call a winner too early, because the calculator already told you how long to wait.
Everything in one place
Kitchen Lab pairs the A/B test itself with the diagnostic and reporting data that usually lives in three or four other tools.
Testing
Drop in one snippet, define your variants, and Kitchen Lab handles the split: no separate A/B testing or split-testing platform to learn.
Revenue
Conversions get matched to real Stripe charges, so "B won" is backed by actual payments: not just a click on a button, segmented by device, source, and new vs. returning.
Confidence
Statistical significance built into every running test, with a sample-size calculator so you know how many visitors (and roughly how many days) before you're allowed to peek.
Heatmaps
Click heatmaps, scroll-depth charts, and rage- and dead-click detection per variant, so a losing test comes with an answer for why, not just a number.
Search
Organic clicks, impressions, CTR, and average ranking position for the page you're testing: pulled straight from Search Console, no manual exports or a second tab.
Speed
Largest Contentful Paint, Interaction to Next Paint, and Cumulative Layout Shift from Google's Chrome UX Report, so a "faster" variant isn't just a guess either.
Built for agencies
Kitchen Lab was built around the way CRO and SEO agencies actually work: many client sites, many people asking "what did we test last month," and zero tolerance for one client's revenue numbers ending up in another client's report.
Questions
A/B testing (also called split testing) shows two or more versions of a page to different visitors and measures which one converts better. It's the core method behind conversion rate optimization (CRO): instead of guessing which headline, layout, or offer performs best, you let real traffic decide, and you don't ship a change to everyone until the data backs it.
Most heatmap and session-recording tools stop at "here's what people did." Kitchen Lab starts there too (click heatmaps, rage clicks, dead clicks, and scroll depth are built in) but it also runs the actual split test, tracks conversions and Stripe-verified revenue per variant, and tells you when a result is statistically significant. It's the testing and the diagnosis in one place.
You can log conversions with a simple event call, but if a project connects Stripe, Kitchen Lab matches each conversion to the closest real Stripe charge in the same window and reports that as verified revenue per variant, so the number behind a "B won" call is an actual payment, not just a click on a button.
Kitchen Lab checks statistical significance on every running test and shows a built-in sample-size calculator, so before you start you can see how many visitors per variant you need to reliably detect a given lift at 95% confidence and 80% power, and roughly how many days that'll take at your current traffic.
Yes. Every site or client lives in its own project, with its own tests, ideas, Search Console connection, and Stripe account. Nothing from one project ever shows up in another, which is the main reason agencies pick Kitchen Lab over a single-site CRO tool.
Yes. Each project's Search & Speed tab pulls real Core Web Vitals (Largest Contentful Paint, Interaction to Next Paint, and Cumulative Layout Shift) from Google's Chrome UX Report, and organic clicks, impressions, CTR, and average position straight from Google Search Console, so page speed and search performance sit next to your test results instead of in a separate tool.
Free to start: no credit card, one snippet, your first test running today.