A/B testing, also called split testing, compares two versions of something, such as a web page, email subject line or offer, by showing each to a different random group of people and measuring which achieves more of a chosen goal, such as clicks or sales.
A tool splits the audience, serves version A and version B, and records results. A statistical test then indicates whether the difference is likely real or just chance. Testing one change at a time makes the result easier to interpret, while multivariate tests combine several changes and need far more traffic.
Online retailers test product page layouts, button wording, delivery messages and checkout steps. Email teams test subject lines and send times. Hospitality venues can test menu descriptions, booking page layouts or promotional offers in digital channels.
The biggest pitfall is stopping early or testing with too little traffic, which produces misleading winners. Decide the goal and run time before starting, run tests across full weekly cycles, and avoid changing other things mid-test. Small shops may lack the traffic to test minor tweaks, so focus on bigger changes. Not every winner holds forever.
Testing replaces opinion with evidence, a core part of conversion rate optimisation. RetailCare's eCommerce and marketing work includes testing and measurement.