A/B tests is a research method, which consists in effectiveness verification of e-mailing, WWW page, layout clarity, newsletter template, contact form or another component which effectiveness, reception level and conversion we would like to measure. A/B tests are often called an experiment, since testing persons, often trial and error like, experiment with the form and content of elements the efficiency of which they want to verify.
There’s one objective: it needs to be established which of the tested versions is for some reason superior, that is which generates more clicks, which boosts sales conversion, from which there’s more redirections. Let’s address the most tested element – e-mail messages.
Step 1: What is your objective?
Basically, one needs to answer the question: What’s the goal of carrying out A/B tests? What do we want to assess? Do we hope for open rate increase? Or maybe we care about the other variable – click through rate? These are most common ratings of e-mailing and newsletters the senders are not particularly satisfied with. Various elements of the message have influence on other ratings.
It’s very likely that the changes in graphic layout or design will turn out useless if nobody click your emails. Then, it’s worth to investigate created e-mail messages topics or anti-spam filters and their ‘requirements’. If the content of the e-mail messages doesn’t go through the filters, then the reason for low open ratings is clear.
Step 2: Determine the subject of your examination.
Right at the beginning you must know what you want to test. Below, you’ll see most tested elements of e-mail message:
- e-mail subject
- call to action button types – for instance: encouraging to download some materials (e-books, case study, etc.)
- layout, elements arrangement (visual part)
- headers
- images / graphics / videos
- content (of an e-mail)
- buttons and graphic elements colors
There is some sort of unwritten agreement that the most important elements are to be tested in the first place. These are basically e-mail subjects, which addressee sees at first and initially draw his attention. Next, these are call to action buttons and only later the headers and the arrangement of particular elements. Without the initial attention, it’s hard to get one later on.
Step 3: Start testing!
Once you determine which particular element of your e-mail you want to test, it’s time to act. In order to accurately carry out A/B tests, you’ll need a suitable tool. Here, Marketing Automation system comes handy. It will allow to create such type of test, carry it out and thanks to advanced analytics panels it will enable to collect appropriate feedback afterwards.
Summary
It’s truly worth to remember about one detail, which when missed may influence test quality and pervert the result. It’s the time. Time is the key – both versions of tested messages must be sent at the same moment. Without it, the result won’t be authoritative. The added value of such tests is the possibility to make quick decisions related to information architecture on WWW page based on their results. To sum up, the tests are easy to implement and collected data is not too difficult to interpret.
Follow
