When it comes to optimizing your adult website’s user experience, traffic and revenue streams, nothing can take the place of formalized testing: comparing one specific approach to another by splitting visitors between two variations of a particular element; such as images, design features or call to action, in order to determine which is the most profitable. It is not something that a webmaster can effectively do manually, however, due in part to issues such as fluctuations in traffic sources, which will prevent a “try this one day, try that another day” approach from obtaining reliable results.
For true split A/B testing, half your site’s visitors should see one element; half should see the other.
It is an intriguing proposition for adult webmasters struggling to incorporate an easy and effective split testing regimen into their PHP-powered websites, including sites powered by WordPress and other popular content management systems offering customizable code.
Boasting endless possibilities with universal results, phpA/B (www.phpabtest.com) offers A/B testing with a simple setup, using a PHP script for easy integration and Google Analytics for detailed statistics.
“Setting up an A/B test couldn’t be easier. Almost everything is done for you,” the script’s author, Brian Cray, states. “You can be running multiple tests in only a matter of minutes.”
It is an intriguing proposition for adult webmasters struggling to incorporate an easy and effective split testing regimen into their PHP-powered websites, including sites powered by WordPress and other popular content management systems offering customizable code.
“Whether you want to test changes to your copy, HTML, CSS, JavaScript, or test completely different designs, phpA/B offers a way to do it,” Cray notes, adding, “phpA/B integrates automatically into your Google Analytics, allowing deeper insight into your tests’ impact on users.”
The high performance PHP-based tests run before users even begin loading your web pages, while avoiding unexpected problems arising from JavaScript-based A/B testing solutions. Furthermore, phpA/B automatically filters out search engine robots, preventing them from skewing your data and test results.
According to the publisher, phpA/B includes a “trial mode” that displays a random variation upon browser refresh, without recording any data to Google Analytics.
Up to five tests can run simultaneously (a limit imposed by Google Analytics), by giving each test a different name and ensuring that they are placed in different Google Analytics reporting slots, using the set_ga_slot() parameter.
The custom {phpab} tag as well as PHP logic is able to specify testing variations, which can be limited to changes in CSS or HTML only, with examples provided in the script’s documentation. Of course, these variations can utilize a mix of CSS, HTML and JavaScript, without creating any problems for the script set.
The active variations are detectable via JavaScript, allowing for more flexibility and detailed analysis.
Advanced Google Analytics users that have made major customizations to their tracking code may face problems with test results not showing up, since phpA/B may not automatically be able to add the additional Google Analytics code it needs to work. The documentation offers a workaround for this.
The phpA/B script uses Google Analytics’ custom variables API for reporting, with _setCustomVar() calling a session-level custom variable that includes the test name and the user’s variation. The script intelligently detects whether the site is using traditional or asynchronous tracking and then inserts the appropriate calls.
Using phpA/B requires PHP version 5 or above to be installed on your server, plus Google Analytics and a thirst for knowledge. The phpA/B script is free for commercial and personal use, licensed under version 2 of the GNU General Public License.
In testing, phpA/B effectively rotates between control content and alternative content, on browser refresh and via unique visitor — providing the click stream data users need to optimize their websites for better performance and sales. Try out phpA/B and see if it works for you.
