A/B test
Conversion and UX
Conversion and UX
In an A/B test you show two versions of a page or element to comparable groups of visitors and measure which performs better on one predefined goal.
Decide up front what you measure, how many visitors you need and how long you run — at least one or two full weeks, so day-of-week differences are included. Don't stop the moment the result looks good.
Most tests end without a clear winner, and that's fine: 'no difference' also saves you a rebuild that wouldn't have paid off.
Related terms
- Statistical significanceStatistical significance indicates how likely it is that the difference between two variants isn't chance. In marketing we usually work at 95% confidence.
- CROCRO is conversion rate optimisation: getting more out of the traffic you already have through research and experiments.
- ConversionA conversion is an action you defined as valuable: an enquiry, purchase, download or phone call. The conversion rate is the share of visitors who take it.