A/B Testing Your Landing Pages: A Complete Guide

A/B testing is easy to start and easy to fool yourself with. The mechanics - show half your traffic one version, half another, compare - are simple. The discipline is in deciding what counts as a result before you look at one.

Before you start

• Write down what you expect to happen and why. A test without a hypothesis produces a number you can interpret any way you like.
• Change one thing. If you change the headline and the form and the image, a win tells you nothing about which change won.
• Decide in advance how long the test runs, and do not stop it early because the numbers look good.
• Make sure you have enough traffic for the difference you care about. A page with a handful of conversions a week cannot detect a small improvement, no matter how long you wait.

Reading the result honestly

Early in a test the two versions will swap places repeatedly. That is noise, not a trend. Watching a live dashboard and stopping the moment your preferred version is ahead is the most common way to produce a confident wrong answer.

It is also worth accepting the outcome where neither version wins. That is a real finding: whatever you changed was not what was holding the page back, and the next test should look somewhere else - usually at the offer rather than the page.

What to test first

Start with the things a visitor sees before deciding to stay: the headline, the first image, and what the form asks for. Button colours and word tweaks are easy to test and rarely the reason a page underperforms.