A/B Testing – and Your High School Science Class

Written by: Gary Horsman

A/B Testing

and Your High School Science Class

At its core, A/B Testing is exactly what it sounds like – it is an experiment. You have two versions of an item (A and B) and a metric that you define that defines success of the experiment. To determine which version is better, you subject both versions to experimentation simultaneously. You measure which version was more successful versus your success metric and select the better version for real-world use. You can do this with pages on your website. You can even push this concept further and apply it to an entire contact path.

This is similar to the experiments you did in your high school science classes. You may remember the experiments in which you tested various substances to see which supports plant growth and which suppresses it. At different intervals, you measured the growth of plants as they were subjected to different conditions. You measured and tallied the height of the different plants which were subjected to the variations in conditions. Your success metric was height but it could have been another metric (color, number of leaves, seed production, etc.). You defined a metric and then measured against it. The results told you which conditions were optimal for performance versus your success metric.

In the case of a website page, you define a goal metric (a sale, a booking, a new subscriber . . .) and then measure how well the A version of the page performs versus the B version. You let the data define the best performance.

Consider how A/B Testing can help the performance of your website pages.

A/B Testing – What to Test