A/B Testing on a Smartphone in Person's Hand
ane photo/Shutterstock.com

If you’ve researched web design, UX/UI pattern, or marketing, chances are you’ve heard the term A/B testing. But what does A/B testing actually hateful? Today nosotros’ll accept a closer look to observe out what it’s all about.

What Is A/B Testing?

Put only, information technology means comparing two versions of a production to see which one performs ameliorate. A/B testing is also chosen “dissever testing” or “saucepan testing,” equally in, “putting things into two different buckets.” And it tin can exist really useful in refining your design.

Why Use It?

A/B testing lets y’all test out a hypothesis and get together data before committing to a change, instead of doing it and just hoping for the best. On a large-scale site design or marketing project, that tin can save a massive amount of time and coin.

How Does it Piece of work?

The concept of A/B testing was really refined dorsum in the 1920s by a statistician and biologist named Ronald Fisher, who kickoff used it with agronomical experiments. Information technology apace went from “what happens if I use dissimilar fertilizer on this plot of land,” to clinical trials in medicine, and to web blueprint and marketing today.

Say yous’re designing a website, and you want to see which pattern tweaks will make people stay longer. You’d create 2 versions of the page, ane with the changes and one without — version A and version B. One version serves as the control, with no changes, and the other is the variation.

It commonly works like this:

  1. Choose what you lot want to examination.
  2. Show the control and variation versions to groups of people randomly.
  3. Track the data to testify which version influenced your results the most.

Randomization is critical to this testing process, as information technology helps remove other variables from the equation. If yous want to test the size of the subscribe button for your newsletter, for example, y’all’d show people the control and variation pages randomly on both desktop and mobile to keep that variable from skewing the information.

A/B testing can be done with more than than 2 pages, but you ordinarily use ii products to beginning. How many people y’all bear witness each version varies based on whether both versions are new, or the new version is competing against an established web folio. If both are new, you’ll probably split traffic fifty/fifty. If yous’re introducing changes against an established folio, it might be 60/twoscore.

Regardless of how y’all decide to distribute traffic to the pages, you lot always testify returning users the same version to maintain the integrity of the test. The test needs to run long enough to gather enough data to be statistically significant before a decision can be made. This sounds complicated, but there are complimentary tools out at that place to help you plot this out.

Any element of any page can exist A/B tested. Trying to get more than clicks through from Google? Test multiple headlines. Trying to get people to navigate through to other pages on your site? A/B test different card options and layouts.

Common page elements that become A/B tested are:

  • Call to activeness (CTA) buttons like Subscribe, Sign Up, etc.
  • Headlines
  • Landing pages
  • Images

Web designers can literally alter 1 affair on a folio, run an A/B test, and rail the results. If something changes, they can be reasonably certain it was considering of the tweak they made to the blueprint.

Again, this concept isn’t sectional to web blueprint. You can A/B test different marketing emails against each other, different medications, and then on. An A/B test is the nearly bones kind of randomized command trial and you tin can apply it to continuously better the user experience. If you lot’re interested in learning more and possibly implementing it in your projects, get further with a deep-dive on A/B testing.