- Cory Checketts
- On April 26, 2016
- 0 Comments
A/B testing has been around long before the internet. However it has only been in the past decade that it has become a staple of online marketing. Internet giants like Google, Ebay, and Amazon are known to use A/B testing to perfect their web pages. Let’s break down what A/B testing is and exactly how you can use it to build a better business.
What is A/B Testing?
A/B testing is showing two different versions of a variable to a percentage of your customers and seeing which version converts better. To give a simple example: on your website’s homepage you could show 50% of the visitors a red download link, and the other 50% a green download link. After a month you check the data and see the red converted at a 23% rate, and the green at 9% rate. You now know the red download link converts at a much higher rate.
Can Details So Small Really Matter?
It may seem the difference between red and green would be irrelevant, but surprisingly small details like this could be the difference of an email campaign converting highly, or falling flat on its face. This article shows an A/B test that saved an app from cutting out almost half of its viewership. The short version of the article is that they changed app navigation from buttons on the front page, to a side menu.
Many people liked the design change, but their engagement time was halved! They changed it back to the original button menu, and created an option to make it a side menu if the user chose to. This just goes to show that no detail is too small.
How Can You Use A/B Testing?
Obviously you cannot A/B test everything. If you tried that you’d constantly be changing every detail until no two products were alike. If you come across something that you have two ideas for, and are not sure which is best, then you should run an A/B test. If something is not converting as well as it you feel it should then run an A/B test on a variable that could be culprit. Common variables that are usually A/B tested are:
- Calls to action
- Any graphic you use in direct correlation to your sales efforts
- Sales copy or product descriptions
- Header images
- Landing page headlines
- Email subject lines
- Email incentives and offers
- Image content and placement
- Landing page copy
- Form fields
- Form length
- Email send time
- Amazon product listings
There are obviously endless possibilities in A/B testing, but don’t drive yourself crazy by running so many tests that you lose the original purpose of your variable. Its quality could suffer if you over test. A/B testing can be the answer to bad conversion rates, but it can’t answer anything if the source material is becomes poor quality. A/B testing is a precision tool that can greatly increase your conversions if used delicately.
Click below to see the case study on how Kevin Vegors boosted his own email campaigns.