A/B Split Testing – Path to a Better Product

5 April 2017

A/B Testing (Split Testing) is a process whereby users are shown different versions of your website or app and then records how well each perform. The goal by monitoring the manner in which users interact with the page, is to see which version yields a higher conversion rate.

twitter

Your website and app has a purpose whether that be for users to buy products, convert visitors into paid users or to generate revenue from paid advertising. As a web-master you need to know how users interact with your ‘channel’ and how you can increase wanted interactions and overall conversions.

A/B testing allows you to make more use of the existing traffic your channel already has. Acquiring new traffic can be very expensive and time consuming, however converting your already existing traffic is far less costly in all areas. A small change to your landing page or website can increase conversions dramatically and this is where A/B testing comes in.

By designing two versions of specific landing or content pages, a targeted A/B campaign can be used to ‘learn’ which of the two generated higher results.

You can test almost any aspect of your channel such as:

  • Headlines
  • Sub headlines
  • Paragraph Text
  • Testimonials
  • Call to Action text
  • Call to Action Buttons
  • Links
  • Images
  • Content near the fold
  • Social proof
  • Media mentions
  • Awards and badges

A/B has a simple scientific process similar to any experiment.

  1. Ask a question  – “Why are only 20% of my user base finishing a transaction”
  2. Do background research – This helps you to understand the behaviour of your user base and what they look for within your channel (use Google Analytics or other analytics tools to assist in this process).
  3. Construct a hypothesis (Prediction of what you think will happen) – “Including a progress bar will encourage more people to finish their transactions”
  4. Calculate the number of visitors you need for the test – Use an A/B test duration calculator.
  5. Test you hypothesis – Did version B (that contained a progress bar) yield more conversions that version A (original/control)?
  6. Analyse and conclude – If your hypothesis was correct then continue to step 7. If not then return to step 3 and create a new hypothesis.
  7. Share you results – Let your clients/those who need to know about the results know the outcome.

A/B Testing allows you to produce an end product that the user can interact with happily and without any confusion. The simple process can optimise your traffic conversion on the users you already have whilst resources elsewhere can be spent on growth. Furthermore A/B testing shows you a user’s behaviour and why they behave the way they do, allowing you to continue to meet a user’s needs even after the initial conversion.

A/B testing a product also saves you time and money in the long term. By testing how your users interact with your channel and what they don’t like, need or want when producing your channel can result in a more efficient channel and profitable channel.

Conclusion:

Overall A/B testing is a crucial part of creating a user channel and without it you can waste resources. By following this simple process you can optimise and produce a user channel that will yield a high conversion rate of free to paid users by utilising user experience and interface aspects.

If you would like to know more, give us a call and one of our specialists will discuss how A/B testing could be used on your site.