AB Test Calls to Action

 
Line 13: Line 13:
Design variations for your call to action, e.g. "Sign Up" could have variations of "Sign Up Now", "Find out more" and "Read the details".  
Design variations for your call to action, e.g. "Sign Up" could have variations of "Sign Up Now", "Find out more" and "Read the details".  
-
You then need to configure your website so that different users are shown these variations. If your content management platform doesn't have an integrated AB Test module, there are a number of third-party services available, such as [http://www.google.com/websiteoptimizer Google Website Optimizer] (Free) and [http://visualwebsiteoptimizer.com/ Visual Website Optimizer] (Free trial available, at time of writing). These services will also accurately measure the results of each variation and give you the statistics necessary to determine when the results are significant (i.e. not just by chance).  
+
Next, configure your website so that different users are shown these variations. If your content management platform doesn't have an integrated AB Test module, there are a number of third-party services available, such as [http://www.google.com/websiteoptimizer Google Website Optimizer] (Free) and [http://visualwebsiteoptimizer.com/ Visual Website Optimizer] (Free trial available, at time of writing). These services also accurately measure the results of each variation and provide the statistics necessary to determine when the results are significant (i.e. not just by chance).  
-
Once the results are in, modify your website accordingly, and repeat the process if necessary, based on your findings (i.e. try subtle variations of the new winning phrase).
+
Once the results are in, modify your website accordingly, and repeat/iterate the process as necessary, based on your findings (i.e. try subtle variations of the new winning phrase).
=== Why Use This Pattern  ===
=== Why Use This Pattern  ===
-
It's almost impossible to know which text will work best unless you run tests (what works for other people may not necessarily work for you), and an AB test may be one of the quickest, easiest and cheapest methods for improving your sign-up/purchase rate.
+
It's almost impossible to know which text will work best unless you run tests (what works for other people may not necessarily work for you), and an AB test may be one of the quickest, easiest and cheapest methods for improving your sign-up/purchase rate.  
 +
 
 +
=== References  ===
 +
 
 +
#[http://dustincurtis.com/you_should_follow_me_on_twitter.html Dustin Curtis: You Should Follow Me on Twitter]
 +
#[http://www.niemanlab.org/2009/10/how-the-huffington-post-uses-real-time-testing-to-write-better-headlines/ How The Huffington Post uses real-time testing to write better headlines]<br>
 +
#[http://blog.appsumo.com/be-careful-what-you-ask-for-how-2-words-dropped-ctr-36/ Be careful what you ask for. How 2 words dropped CTR 36%!]
 +
#[http://meta.wikimedia.org/wiki/Fundraising_2010/Banner_testing Wikipedia 2010 Fundraising Banner: AB Test Results]
=== Examples Gallery  ===
=== Examples Gallery  ===
-
[[Image:Dustin-curtis-ab-test.png|thumb|left|Dustin Curtis increases CTR by 173% using an iterative AB Test]]<br>
+
[[Image:Dustin-curtis-ab-test.png|thumb|left|Dustin Curtis increases CTR by 173% using an iterative AB Test]]<br>  
=== Categories<br>  ===
=== Categories<br>  ===
[[Category:Editorial]] [[Category:Microcopy]]
[[Category:Editorial]] [[Category:Microcopy]]

Latest revision as of 04:17, 17 January 2011