Landing pages | 4 Min Read

Examples of A/B Testing: How To Increase Conversions

Joshua Todd

Joshua Todd

October 24, 2020

Examples of A/B Testing: How To Increase Conversions

Table of Contents

    What is A/B Testing?

    A/B testing is an experiment that examines two variables, A and B, of a website. In a more precise way, it refers to the act of comparing two varying elements against each other by showing it to different visitors at the same time to see the one that drives more conversion.

    Besides, this process eliminates the doubts associated with website creation and optimization. Instead of judging based on your interests or what you feel is right, you can make data-backed decisions based on the test result.

    How to Do A/B Testing On Website

    In the previous section, we defined what A/B testing is and its importance. In this section, we will discuss the ways to carry it out on a web page. 

    Here is a walkthrough guide of the A/B test:

    1. Determine What You Want To Test

    The first step is to determine what you want to test. When you know what you want to test, the work becomes easy. However, avoid the common mistake of some marketers who try several different components at once. 

    When you test several parameters at once, you will have challenges pinpointing the one that yielded results. Test a single component at first and ensure it is appropriate to the metric you choose.

    2. Set Your Conversion Goals. 

    Knowing your conversion goal will give you a good head start. If you understand the conversion goals you want to achieve, you will select your page’s components you need to test.

    Do you like your website visitors to sign up for your event, subscribe to your email list, or purchase your product? It is advisable to test on a single metric at first. You can then assess other metrics later on.

    3. Formulate Your Hypothesis.

    After you have set your conversion goals, the next step is to state your hypothesis. This stage lets you know the changes you might have to make (if need be) before you begin your testing.

    Here, you set your test options. You can choose between using an existing landing page and a new one. 

    4. Create Your Variations.

    The next step is to create a variable option that makes the desired changes to your page. Creating variations for your current page can be quite a herculean task.

    Kissmetrics, Crazyegg, and Omniconvert are testing software that will help you create variations. To run your test conveniently, you can use the drag and drop features available in your website builder.

    4. Run Your Test.

    This stage is where the main action takes place. When you run your test, you do need half of your visitors on a variable, while the other half sees the other.

    You can run a Split URL Testing by testing different versions of your web page hosted on different URLs or a Multivariate Testing (MVT) by changing your web page’s sections for all the possible combinations.

    5. Keep Your Test Running Long Enough.

    It is essential to keep your test running for an extended period. You can calculate your test duration by noting your average daily and monthly visitors, current conversion rate, minimum expected conversion rate, and visitors' percentage.

    The sample size required to reach the appropriate test results also determines the test’s duration. You can repeat the experiments in these tests until you reach the desired conclusion.

    No amount of time is too long to run the test. What matters most is to run the test until you can calculate a statistical significance. This value varies from one website to another, depending on the amount of regular traffic. Testing software can help you predict the time range. 

    6. Analyze Your Results.

    Once the time frame you set is up, analyze the results by checking metrics like percentage increase in the number of subscriptions, time spent on your website, and orders completed.

    Check where there is statistical significance between your control and variant options A and B by analyzing your results.

    If the test is statistically significant, you can use the winning variation, but you can create another one using the insights gained if it is inconclusive.

    Try our free A/B test calculator.

    Typical Example of A/B Testing

    A/B testing is used in different industries and has helped in making better marketing decisions. Several businesses have increased the number of their subscribers and average order value. They have improved their user experience, which resulted in more time spent on their website.

    Taking the eCommerce industry as an example, Amazon is at the leading edge in conversion optimization because of its dedication to delivering a premium customer experience.

    Booking, an example in the travel industry, created three versions of its landing page in 2017 and ran A/B tests for two weeks. This test resulted in a 25% increase in owner registration and a significant decrease in each registration cost.

    Conclusion

    This article has explained what A/B testing is, the benefits, and how to carry it out on your website. I also used Booking.com as an example of a website that used A/B testing in its optimization plan. It has been paying off for them.

    Without a doubt, it will pay off on your website when you do it right.