A/B Testing in a Nutshell

By Testing

A/B Testing is a question of choice.

Do the majority of your visitors prefer this or that? Does a red button work better than a green button? Are there too many form fields and is it putting people off?

How do you know?

Welcome to A/B testing, the tool that let’s you try different scenarios on real life customers to find out what works best.

When you first design a website or landing page you are making a series of assumptions about your site visitors, such as:

  • Age
  • Demographics
  • Needs / Desires / Wants

All of this is a good way to start thinking about the design of a website or landing page but are all (to some extent) assumptions.

In the end, the truth is in the data. How well is a particular page performing? Are there things that are causing problems or confusion? Are there things we could add or remove that would help with trust and conversions?

In a nutshell

A/B testing, in it’s simplest form, allows you to create variations on a design that can then be “split” tested against a set of visitors.

Visitor A might see the version with a red call to action button and Visitor B might see the version with a green one.

Once you run this “test” for long enough (or for enough page views) you can start to gather data and take an opinion on which one works best.

It may sound crazy but the difference in the colour of a button could have quite a difference on your bottom line if it comes to making sales online.

Small changes make the best tests

A/B testing is best suited to optimising well designed pages. It is not really intended to test large differences between pages (although it can be used for that earlier in the design phase). If you have too many changes between test types then you have no real idea what was the element that made the difference. So to judge on a broad scale if Design A is preferred over Design B it can help but as to what it was that made that design work better you would not really know.

By taking a page that is already performing well, then focusing on running tests where you have changed one or two elements you can rationally evaluate the best result and incrementally improve the performance of that page to maximise conversions.

Ensuring the ROI

You could invest a lot of time split testing pages to try and find out whether you can improve your conversion rate but it is important to understand the potential “value” of that work.

If a 5% increase in conversions is worth a lot of extra profit then it would be worth going to that level of detail. If it only means a few dollars then it may just be a waste of time. It pays to carefully consider where to spend your time on trying to improve your bottom line and A/B testing is just one of the many tools we use to improve your profits.

If you would like to find out more about A/B testing or our other services contact us on 02 8097 7957, email us at hello@conversionworx.com.au or complete our online enquiry form.

Tagged under: