Imagine if Optimizely and Claude had a child.
Ask Flint to test a new version of your landing page. It builds the variant, splits your traffic, and tells you which page won. One Fortune 10 healthcare brand cut its CAC in half in 48 hours.
Why A/B tests
The only rule in performance marketing is ABT: always be testing. But each test can take weeks to coordinate across teams, and experiment systems are hard to maintain.
Flint already runs your Google Ads and builds the pages your ads point to. So it can run the test too.
What it does
Pages. Flint builds each variant as a page on your site, in your design system. Test a headline, a hero, a call to action, or a whole new page.
Traffic. Run up to four pages at once: your current page and three variants. Split traffic evenly, or keep most of it on the page you have today, like 70/15/15.
Goals. Pick the number your team reports on. A click on a button, a custom event, or a conversion further down the funnel, like a Shopify purchase or a HubSpot qualified lead. Flint tags each visitor with the version they saw, so the conversion counts toward the right page.
Results. Every experiment shows a preview of each version, then visitors, goal rate, lift and chance to win, plus how many days are left before there's enough data to call it.
How to start
If you're on Flint, ask in the app or in Slack. A few prompts to start with:
- @Flint set up a new variant for my landing page and run an A/B test against it
- @Flint test three new headlines on /pricing against the current one, with 70% of traffic on the current page
- @Flint how is my A/B test going?
You can view more at /docs/abtesting.
New to Flint? Book a demo.
