Google’s free Analytics service has long been one of the most widely used analytics tools for website owners. Google Analytics provides powerful tools for webmasters and website managers to discover how their visitors are coming to their site, what pages they’re viewing while they’re on the site, and how long they’re spending on each page. Google Analytics also integrates with Google’s AdWords program, so that a web manager can integrate their advertising program goals into their website, and so that they can track and measure the effectiveness of their various advertising campaigns.
For some time many webmasters have also used Google’s Webmaster Tools to conduct “A/B testing” on various pages of their websites. An A/B test is one in which a website has two different versions of a particular web page (an “A” version and a “B” version), randomly serves those different versions to site visitors, then measures how the visitors behave. The purpose of this test is to see which version of the page is more effective in accomplishing a particular goal.
For example, a website might have two different versions of a newsletter sign up page, and A/B testing will enable the webmaster to learn which version leads more visitors to sign up.
Google has now added the ability to do A/B testing directly to Google Analytics. This means that you can now connect your various Analytics and AdWords goals directly with A/B tests to learn not only what types of advertisements bring the most visitors to your site, but which versions of various action pages those visitors best respond to. Within Google Analytics, A/B testing is called “Content Experiments.”
To set up your first content experiment, first log in to your Google Analytics account. Look under the “Content” section of the main sidebar menu, and click “Experiments.” Next click the “Start experimenting” link.
To set up your first experiment, decide what you want to test. Content Experiments can be created not only for pages, but also for particular elements within a page. You can also test two or more different headlines or headers, images or icons, blocks of text, or calls to action.
Once you’ve decided what you want to test, you add URL for original page, plus URLs for the variation pages you want to test. Within a particular Content Experiment, you can test up to six variations of a page. Note that the variations you want to test should only have one particular element that changes. If the pages you’re testing have multiple variations, then you might not know which changed element is responsible for the different performance among your visitors.
Build up your understanding of Google’s new Content Experiments by starting small, and learning everything you can about each test you run.
Just a normal software developer who love to write code. Developing professional software since 1999. After working in the software industry for many years, I've started my own website to share knowledges and experiences.