A Guide to WordPress A/B Testing through Google Analytics



When it comes to webpages and websites it’s often the little thingsthat make a difference. You may be amazed but the tiniest incrementaldifferences can make the most significant changes to conversions andsales.

The answer to this is A/B split testing as you may already know. A/B testing allows you to try out twodifferent versions of a page to find which one performs the best. It’s a tried and tested manner to increase conversions and we’re going to show you how to use Google Analytics to do so on your WordPress website.



Before you Test

It’s pivotal to take into account a number of things before you testyour website to ensure the most accurate and apt optimisation.

  • You need to decide on the areas to test and these should be selected in accordance with the result you want.  So, if you wish for moresignups then you will need to test field boxes, sign up forms etc. Ifyou want to increase sales conversions – the checkout can be a goodplace to start.
  • Be focused and consistent when testing. You need to schedule your testing to prevent a pileup of data and to ensure accuracy.
  • Decide on a set period for testing and ensure you have enoughtraffic to test; otherwise the statistics may not be accurate. Ideallyyou will need 100 clicks or more prior to testing to ensure that theresults are accurate. The higher the number of clicks the more likelythe A/B split testing will be accurate.



First things First

Before you begin, you’ll need to do a few things if you’ve not already done them.

  • Install and set up a Google Analytics Account on your WordPress site
  • You also need two versions of the page that you are testing. Onecopy can be the original and the other should have a different layout,typography or something else than the first. The original and variationwill need to be published

How to Set Up your Test



  • Google Analytics offers a quite easy manner to get things up andrunning. Within the program, A/B testing is known as experiments. Thiscan be found by going to the ‘content’ tab on the left of the page,clicking and then scrolling down to experiments. Clicking on your URLwill get you started.
  • Now, choose a name for the test and an objective. This allows you to choose and track the number of visitors to each variation page achievethe objective. As an example, this goal could be the number of peoplethat flow through a certain channel to your goal page. So, how manypeople complete a registration form and land on a thank you page forexample. This would then be counted as one completed objective.
  • The next thing to do is to choose different variations of the pagefor people to land on. As we said make alternative versions of the pageand make them publically accessible. These can have different calls toaction, designs and layouts.
  • When you add these you will be taken to the next page which has acode to add to the WordPress site’s header. Copy and paste this to thetheme’s header. php file.
  • By adding this code visitors will be directed to different versionsof the page. Simply, replace $page_id with the id of your original page. It’s also possible to do this by using a tool for WordPress calledGoogle Content Experiments – this means you don’t have to modify thecode.
  • When this is completed and you’ve added the code to the originalpage, you can then click through to Google Analytics and press the Nexttab. This verifies the experiment with Google and that it’s properlyinstalled. Now, simply press the start experiment to begin A/B testingon the pages.
  • Now that this is complete you should have no problems and you’re A/B testing on WordPress should be underway. It often takes a period oftime before Google begins to showcase the results of the test data. As a digital marketing agency in Leeds and London we’ve seen some of our client’s conversions benefit numerous folds after this testing.
  • During the process some visitors will be sent to the alternativepages and some the original pages. Return visitors will continue to beshown the same variation on their subsequent visits when the experimentis on-going.
  • As time goes on Google Analytics will then inform you which pagesare working better than others and helping you better meet yourobjective. It also decides on a winner if it can see that one page issignificantly doing better than the other page.

Split page testing can play a significant part in the success of awebsite and it’s not unknown for conversions to increase significantlyafter the best page option is chosen. So, why not go ahead and see ifyou can improve your website with a little help from A/B testing.

This entry was posted on by .

Leave a Reply

Your email address will not be published. Required fields are marked *

*

code