The most important things you should know, summarized for you

How to do A/B Testing and Improve Your Conversion Rate: Stop Guessing, Start Testing

These days, everything revolves around data. Executives don't want to make decisions unless they have evidence. This is of course a good thing, and fortunately there are many ways to get information without having to rely on one's instincts. One of the most common methods, especially online, is the A/B test. Many of the web’s most powerful companies, including Amazon and Google, use this technique.

In the following we will explain, what an A/B test is, how to conduct it, and how you can optimize your conversion rate (that is, its ability to turn visitors into customers).

What is A/B testing?

A/B testing is a method of comparing two versions of a webpage or app against each other to determine which one performs better by the target audience. Two or more variants of a page are shown to users at random at an experiment. A statistical analysis is then used to test the efficiency of Version A and Version B with regard to various indicators such as the conversion rate. It can be tested for example which version leads to the most clicks, registrations or purchases. A/B testing is possible on websites, in native mobile apps for iPhone or Android and via server-based APIs.

There are different A/B testing solutions:

Classical A/B Test:
Visitors are shown two or more variants of a page under the same URL. Thus the success of different variants of a certain element can be measured.

Split Test:
It redirects your traffic to one or more URLs. Especially this can be interesting if you host new pages on your server.

Multivariate Test:
A multivariate test is used to measure how several changed elements affect one page. For example, you can change a banner, the color of a text, or even your design. You can use the test to check which combination has achieved the best performance.

How to run an A/B test

For an AB test, you first need additional information that will help you identify your conversion problems and understand user behavior.

  1. To a successful A/B testing, a strong hypothesis is the first step to a successful A/B test. If the identified problem is for example a high drop-out rate on a registration form, the hypothesis could be as follows:
    “The color change of the registration button, from green to white, will increase the number of registered contacts."
    But which elements should be considered when testing? These highly dependent on the company, industry and product. However, the elements can be roughly divided into titles and headings, images, buttons, forms and page structure.

  2. After defining hypothesis, a project team needs to be build for the A/B Test. Then, tests hypotheses are sorted by priority. This is followed by the creation of a roadmap.

  3. The execution of a test varies depending on the selected AB Testing solution and functionality.

    Some AB Test tools are complex and require the assistance of technical experts to modify the source code of the pages under test. Which AB Testing solution and which mode of operation is chosen depends on the level of development and resources of the respective company. Every company is different, so a solution must be chosen that adapts to the needs and limitations of the company.

  4. After testing, the evaluation of the results follows

  5. In addition, the tests should be documented and archived to inform the responsible staff for the conversion rate.

  6. If a version achieves by far better performance than the original version, it should be put into operation. And it must be checked, whether the advantage identified during the test is maintained in the long term.

A/B Testing is an optimization process that is carried out continuously. Conclusions are drawn after each test, these then lead to new test hypotheses. It is an instrument for pursuing a conversion optimization strategy, but it should not be detached from other activities like web analytics, usability tests or customer feedback.

A/B testing tools:

Two examples of A/B testing tools are Google Optimize and Adobe Target. Google Optimize is part of the Google Marketing Platform and has the advantage in contrast to other testing tools that it is linked to other numerous Google products, such as Google Analytics or Google Ads. Google Optimize is the free version, which has already a broad range of features. For more complex testing, the paid variant, Google Optimize 360, offers more features and the ability to run more tests simultaneously and more combinations in multivariate testing.

Adobe Target works especially well for those users who already use Adobe Analytics. In addition to A/B testing, which is easy to set up with existing step-by-step instructions, Adobe Target's core features include multivariate testing, geotargeting, and automated personalization.


Using A/B testing software is a powerful way to increase your website’s conversion rate gauging your audience’s response to a design or content, without disturbing your users’ experience or sending out disruptive feedback surveys.

It is a valuable resource for anyone making decisions in an online environment. With a little knowledge and hard work, you can mitigate many of the risks faced by most beginning optimizers.

Amazon CEO, Jeff Besoz, is also convinced of A/B testing. He states:

“If you double the number of experiments you do per year you're going to double your inventiveness.”

Now if you believe in the power of A/B testing to continuously optimize conversion rates, we recommend: Stop Guessing and Start Testing!

Contact us
We are happy to support you with improving your conversion rate
Get in contact with us
Request a Case Study