Ning Blog

How to perform efficient A/B testing for a landing page

How to perform efficient A/B testing for a landing page
4.25 (85%) 4 votes

Every day we should take new steps and learn something new. So we’re moving ahead with a series of guidelines on different aspects of landing pages and offering you a new guide.

In this guide, we’ll consider what A/B testing is, what the elements of an A/B tests are, which testing tools you can use, the necessity for experiments as well as ways to achieve your conversion goals.

What is A/B testing?

Like any other testing, A/B testing (also called split testing) compares among 2 variants to identify the variant that performs best. To conduct A/B testing, launch two or more pages simultaneously and select the page that performs best. You can select any component of a page to be tested: headings, videos, pictures, calls to action, and so on. Based on testing results, you can optimize your landing page in the most suitable way to transform your visitors into customers.

Types of testing pages

Landing pages that participate in testing are named in a special way. Naming conventions are the following:

Like a winner of a sports competition, the champion page is the page that ultimately shows the best conversion performance.

Like the competitors or rivals who challenge the leader in a sports competition, the pages against which the winner is compared are “challenger” pages. They are also called “control” or “treatment” pages.

Any new version of a landing page being tested is called a page variant.

Page elements to test

Any A/B test may include several elements. If you are not familiar with digital marketing, you may ask “What should I test on my landing pages?” – and be supported by a choir of those who want to know in detail how to test a landing page.

You can test most different landing page elements:

Tips on Running A/B Testing

  1. Test one element at a time. Launching several tests simultaneously can distort the overall picture.
  2. Select the time span for your test carefully: the period shouldn’t be too short nor too long for your testing results to be optimal.
  3. Test single elements (button color or a single CTA) or page variants with totally different design and content. Making high-level changes can give you a comprehensive view of the situation. On the other hand, even a change of a single word can have drastic results (in one case, substituting the word “receive” for the word “get” in a CTA increased conversions more than 50%).
  4. Set the goal of statistical significance in advance. Experts recommend aiming at a range between 97% and 99%.

A/B Testing Tools

If your boss requires immediate results in terms of improving landing page performance and you have almost no time, you can resort to different software tools to simplify your life. These online editors facilitate the testing process and present comprehensive data on page performance. There are free tools (Google Analytics Experiments, A/Bingo, Five Second Test) and paid tools (KISSmetrics, Adobe Target, Unbounce) that can be wielded to help you master web marketing.

When we take a closer look at these tools, it’s obvious that the paid ones offer better opportunities for full-fledged testing. They are complete testing platforms that help you gain deeper insights into your landing page performance.

However, an alternative approach to A/B testing, called the multi-armed bandit algorithm, has recently got into the public eye. The algorithm includes two stages: 1) splitting traffic equally 10% of the time, and 2) sending traffic to the current best performer 90% of the time.

Although this may seem a very promising approach to people who aren’t familiar with digital marketing, the experts who dive deeper into the question claim that all that glitters is not gold. A multi-armed bandit algorithm may outperform A/B testing, but the results should be approached with care. Below you can find the reason why.

This algorithm is applicable only if you are solely interested in average conversion rates and do not take into consideration statistical significance. This algorithm is useful for continuous and permanent optimization, while A/B testing is best for strict experiments with an emphasis on statistical significance. These tools have different purposes; it’s up to the testers to decide which one is applicable for their experiments.

Testing Steps

The A/B testing procedure is very simple:

  1. Create a landing page (follow certain steps and employ special tools).
  2. Publish your landing page (again, with the help of special online tools).
  3. Create variations of the page for A/B testing.
  4. Launch A/B testing.
  5. Analyze the results.
  6. Launch another A/B test.

Let’s consider how you assign traffic to your landing pages that are being tested. You can create as many page variants as you want to test (one page for each idea). Then you assign equal traffic to each variant. In this scenario, you treat all pages equally and can get results in the shortest possible time. If you test new ideas against an existing page, then you should direct smaller amounts of traffic to the new variants as these new ideas are connected with some risks.

The matter of determining which page shows the best performance is subjective to an extent, and the results cannot be known 100% in advance. At the beginning of a test, marketers might rely on their great expertise, gut feelings, and substantiated expert opinions. However, it’s not enough. Actual testing can show completely different results. The final word always lies with your customers who will determine the champions and losers of the page variants you test.

Challenges for Marketers

Making experiments (not only with your landing pages but also with other marketing tools) is one of the requirements for a top-rank marketer or a natural-born SEO specialist. An unquenchable thirst for new options and a constant search for new variants and better methods are the telltale signs of a high achiever. You should be ever vigilant and open to unconventional methods of marketing.

Actually, it’s very simple to achieve high conversion goals, to decrease bounce rates, and to get more sales. Just ask your customers, and you’ll get the correct results. Your visitors – the best critics – will evaluate your online marketing efforts with the utmost precision. Upon getting their feedback, you’ll be able to improve your strategy and tactics.

Tips and insights received during A/B testing will help you pave the way to the bright future of your business. No matter whether you run a brick-and-mortar company or a click-and-mortar project, your online presence and customer awareness will increase immensely if your landing pages are properly tested and well tuned.

If you want to get our educational content regularly, stay with us!