A/B Testing

Also called Split Testing

Comparing two versions of something, a page, a headline, a flow, by showing each to a random slice of users and measuring which performs better on a chosen metric.

Why it matters

A/B testing replaces opinion with evidence for decisions that get enough traffic to measure. It is powerful for optimization, but needs real volume and honest metrics, and cannot answer the big directional questions.

For example

A team shows half its visitors a green signup button and half a blue one, and keeps the blue after it converts 12 percent better.

Worth your time

A/B Testing Alternatives for Low-Traffic Websites CXL · article This is the honest answer most founders don't want to hear: with a few hundred visitors a month, an A/B test almost never reaches statistical significance, so the winner it declares is usually noise. CXL walks through why that happens and, more usefully, what to do instead, from qualitative user testing to focusing on bigger swings that don't need thousands of conversions to prove out. It reframes the whole question from 'which button color won' to 'what is actually confusing people'. Open cxl.com →

Related terms

Go deeper

See how founders actually handle this on Building the product, part of the Starting Up hub.

eChai Partner Brands