How many of you have tried variation of Google Website Optimizer tests to your website to check what’s working for your site and what’s not? Well not many guess many, so here I am going to you take you through the basics of Website Optimizer tests and the factors that should be kept in mind before starting a optimizer test:
What is Google Website Optimizer?
Google Website optimizer is a completely free tool that is offered by Google (Obviously) which can be used to test and create variety of experiments on your webpages/landing pages. All the visitors have different approaches when go through your website and with the GWO (Google Website Optimizer) we can show different versions of the webpages to the visitors and check which versions triggers the most conversions. Visit Google Website Optimizer to get your first experiment started, but before checking it out, here are few thighs that you will need to consider.
Target the Right Test Page
Always ensure that you choose a page that receives high amount of traffic, such as your main category pages or the landing page so that the test will become more meaningful. You can also consider choosing a page that drives traffic through Adwords or a webpage that has been consistently having few conversions. The targeted page should be having a call to action button, link or any such factors that can cause an action like download, sign-up etc.
Choose Conversion Page
A conversion page is the page that represents the result for an action, for example it can be a thank you page after a form submission, after a download etc. Choose the right conversion page while having experiment as this can defines the success of the experiment.
What Test is Right for you?
Google Website optimizer offers two kinds of experiments – A/B tests and Multivariate tests. Let me explain it a bit more:
A/B experiment lets you test two entirely different versions of the page, this is widely opted test if a webmaster will need to experiment with the page design, layout or change the entire look and feel of the webpage. Example you can show visitors two versions of the homepage and check which versions can help get more conversions.
Multivariate experiments lets you test different sections of the page and check which converts more. Example if you want to test different 3 different headlines, three CTA buttons and 3 images, you can opt for multivariate tests.
Get Set with the Contents
Once if have chosen the kind of experiment that you want to run then it’s time for you to select the elements that you want to test and the content that you want to apply (For Multivariate), and also the variation page that you want to test if you are running an A/B test.
For a multivariate test make sure to come up with elements that are very different from the original, that is for example.
Original Page Headline: “Welcome to my Blog”
Variation Page Headline: “Enter … the Conversion Land”
Original Image:“A Simile”
Variation Image: “A Conversion Flow Chart”
Now notice here that the original page elements are simple whereas the variation page elements are more intriguing. This can help you test different reactions of the users.
Length of the Experiment
This is the last and the most important factor to consider before starting a GWO test, while starting a test you can always limit the amount of visitors that see the variations. But always keep in mind to not stop and decide the winning variation until the results reach a successful conclusion (100 % visitors). Now since the length of the test is also finalized you can start a GWO test for your website.