To make use of the Website Optimizer tool, you are simply required to choose the important aspects that you want to thoroughly test. These can include areas like headline, select copy text, image, or promo text. Then an experiment is done across cross-section of the site traffic. The idea is to help you discover exactly what format and type of content appeals to your site visitors and the one they respond to best.

After sufficient representative data is collected, you are provided with reports plus a course of action suggested in order to well optimize your business site.

Test different content combinations

Tracking the effectiveness of your content combinations won’t alone tell you exactly how to make it more effective. The only way to check what kind of content will work best on your business site is to randomly test different content combinations. In case you’re running a multivariate test, you will get two sub-tabs of data: reports either by combination, or those by page section.

It is important to make a structured and elaborate testing plan for a business site for improving end results through AB & multivariate (changing of multiple elements in multiple versions) testing. Some of the factors that need to be kept in mind are in context of the end objectives are as follows:

• Strategies to best to engage especially first time as well as repeat visitors in order to reduce bounce rate.
• Elements to look out for while doming clickstream analysis like forward & reverse path analysis.
• Vital recommendations on the best advanced & custom reports to use for classifying your visitors by traffic source to fathom their behavior.
• Integrating personalization, social bookmarking, reviews, ratings, and video as part of Web 2.0 approaches into your business site.
• Maximizing your landing page, basket as well as checkout conversion rates.
• Planning and managing AB and multivariate testing of pages with tools like Google Website Optmizer.

An A/B experiment, as you must be aware, lets you test compare the performance of two or more different versions of a webpage. Begin with your original page, one whose content you wish to test. Then you may create its alternate versions. You can modify the content of a given page, alter its look and feel, or even move around the layout of the alternate pages. The traffic flow to both your alternate versions and your original page is observed, so as to understand what visitors respond to best.

In essence, an A/B experiments is a simpler version of testing done with Website Optimizer. If you straightaway plan multivariate testing, you may not find it easy to analyze which aspects are impacting the traffic flow and influencing viewers the most. To begin with, just change a few elements between versions for conducting an A/B test. We shall know more about the tests in our upcoming post.