What should you A/B test first on a landing page? The things a visitor sees in the first five seconds, and anything that makes the next step harder than it needs to be. These are the twelve patterns we reach for first, roughly in the order we test them.
Above the fold
- 01A headline that says what you do and who it’s for, not a slogan.
- 02One primary call to action. A second, softer one at most.
- 03Proof next to the promise: a number, a logo row or a short quote right under the headline.
- 04A real product image or screenshot instead of an abstract illustration.
The middle of the page
- 01Benefits written as outcomes (“get paid in two days”), not features (“instant payouts API”).
- 02Objections answered before they’re asked: price, time, risk and “will this work for us?”
- 03A short FAQ near the bottom, written from real sales calls.
- 04Case studies with one clear number each, not walls of text.
The ask
- 01Fewer form fields. Ask only what you need to reply, and the rest later.
- 02A button that says what happens next (“Book a 30-minute call”), not “Submit”.
- 03Message match: the page repeats the exact words of the ad or search that brought someone there.
- 04Speed. A page that loads in under two seconds on a phone, because slow pages lose people before they read anything.
“Don’t redesign a page you haven’t measured. Fix the tracking, then test.”
How to test without fooling yourself
Test one change at a time on pages with enough traffic, run each test for at least a full week so weekdays and weekends both count, and decide what “winning” means before you start. Most importantly, make sure the conversion you’re counting is real: a booked call or a purchase, tracked properly, not a button click.
If your tracking isn’t trustworthy yet, start there. Setting it up, then building and testing the pages, is our tracking and conversion work.
Growth lead at Rvysion. Runs paid ads, landing pages and conversion tests for startups and growing brands.