Latest News & Blog

A-B Testing - Version A vs Version B

A-B Split Testing for Websites

A-B testing is a data-driven approach that helps businesses and website owners improve their digital marketing tactics based on real user data.

It is widely used to optimise web pages, emails, digital advertisements, and other online content.

At eCBD, we use A-B testing as part of our professional SEO Services to help our clients get more revenue from their websites.

Here’s the thing about A-B testing: It is an absolute game-changer when it comes to enhancing your online marketing and achieving outstanding results. 


What is A-B Split Testing on a Website?

A-B testing, also known as split testing, is a method used in website optimisation to compare two different versions of a webpage and determine which one performs better. 

It involves dividing a sample of real website visitors into two groups, often referred to as Group A (the control group) and Group B (the variation group).

Real traffic is then sent to each version of the web page and a metric (such as conversions) is used to measure which version performs better.

Why conduct A-B Testing on your website?

The purpose of A-B testing is to assess the impact of changes or variations made to one version of the content (Group B) compared to the original or existing version (Group A).

These changes can include alterations to elements such as headlines, images, text, layout, colours, or calls to action.

By presenting these variations to users and measuring their interactions and responses, you can gather data and insights on which version is more effective in achieving your desired outcome.

Examples of the outcomes you are measuring could include things like: 

  • increased conversion rates
  • increased conversions such as phone calls, enquiry form submissions or email list sign-ups;
  • longer average time on site;
  • more pages viewed during the user’s session;
  • increase in average revenue per session;
  • lower bounce rate or higher engagement rate for a page;


Getting Started with A-B Testing

Set Clear Goals

Before diving into A-B testing, you must define your goals. What do you want to achieve with your website? Whether it’s increasing sales, sign-ups, or user engagement, having clear objectives is essential.

Create A Hypotheses

Hypotheses are educated guesses about what changes on your website will lead to better results.

You should formulate hypotheses based on your goals and existing data.

Splitting Your Traffic

To conduct A-B tests, you’ll need to split your website traffic evenly between the control group (A) and the variation group (B).

This ensures that your results are statistically significant.

The good news is, that most of the A/B testing tools (mentioned below) currently available on the market can do this automatically for you.


Implementing A-B Tests

Testing Elements

You can test various elements, such as headlines, images, videos, call-to-action buttons, and even layout or colour changes.

Each test should focus on one specific element to ensure you can isolate its impact on your goal.

A/B Testing Tools

There are numerous A-B testing tools available, such as:

  • Google Optimize (now sunset);
  • Optimizely;
  • VWO;
  • Crazy Egg;
  • Freshmarketer

These online tools simplify the testing process and provide valuable insights.

Running the Split Tests

Once your tests are set up, it’s time to run them. Ensure that you collect sufficient data to make reliable conclusions.

We recommend running split tests for at least 60-90 days and that you are driving at least 2 visitors per day to the pages you are testing.

Analyzing and Interpreting Results

Determining Statistical Significance

Analyzing the results requires an understanding of statistical significance. Tools will help you determine whether the changes you made had a meaningful impact.

Drawing Conclusions

Based on the data, you can draw conclusions about which version performed better.

Implement the winning variation to improve your website.


Advanced A-B Testing Strategies

Multivariate Testing

Multivariate testing allows you to test multiple changes simultaneously, offering insights into how different elements interact with each other.


Tailoring your website to individual user preferences through personalisation can significantly enhance user engagement and conversions.

Continuous Testing

A-B testing isn’t a one-time task. To stay competitive, you should continuously test and refine your website’s pages.

By stacking A-B tests, you can dramatically improve the most important pages of your site and increase your return on investment (ROI).


A simple example of website A-B split testing in action

For example, let’s say you want to try and get more phone calls from your website so you decide to conduct an A-B split test.

Version A is your current website design, completely unchanged.

Version B is your variant.

Since you aim to increase phone calls on the B version, you make the following changes:

  • Add your phone number to the header of your website.
  • Add a prominent click-to-call button on the mobile version of your site.

Now, you run your A-B split test for 60 days which sends 50% of your traffic to Version A and the other 50% to Version B.

When you check in on the results at the end of the 60 days, it turns out that your variant (version B) attracted double the amount of phone calls compared to the original (version A).

Based on this evidence, you finish the A-B test and swap your website design to permanently match version B of your test. 


Frequently Asked Questions

How long should I run an A-B test?

The duration of a test depends on your website traffic. It’s crucial to run tests long enough to gather statistically significant data, but not so long that you miss out on potential improvements.

Is A-B testing suitable for small websites?

Yes, A-B testing can benefit websites of all sizes. Even small changes can lead to significant improvements in user engagement and conversions.

What if my A-B test results are inconclusive?

Sometimes, test results may not show a clear winner. In such cases, consider refining your hypotheses and running the test again.

Are there ethical considerations in A-B testing?

Yes, ethical concerns can arise in A-B testing, especially regarding user privacy and consent. Always prioritize user trust and transparency.

Can A-B testing be applied to mobile apps too?

Absolutely. A-B testing can be used to optimize the user experience in mobile apps, just like on websites.

Is A-B testing a one-size-fits-all solution?

While A-B testing is powerful, it’s essential to tailor your approach to your specific goals and audience.


Experimenting with  A-B Testing is a good way to stay ahead of your competition.

By harnessing the power of data-driven decision-making, you can create a website that not only engages users but also drives increasing levels of conversion.


Need A-B Split Testing or Website SEO? Contact e-CBD

Optimise your website, increase traffic, pump up your conversion rates and increase sales for your business.

Work with our in-house team of digital marketing and SEO experts.

For further insights on conversion rate optmisation techniques, reach out to a Gold Coast SEO Agency like e-CBD.

Reach us at (07) 5531 3810 or submit an online inquiry today.

Connect with us

Keep in the loop or engage with us via

Acknowledgement of Country

We respectfully acknowledge the people of the Yugambeh language region, the traditional owners of the land on which we stand, and pay our respect to their elders past and present, and all Aboriginal and Torres Strait Islander Peoples who now live in the local area.

Go to top