Here's a quick overview in case you forgot:
Now's the time to get creative. What do you want to test? Here's a few ideas to get you started:
Use your imagination, the sky is the limit. If you're hard up for ideas, Optimizely has a nice article with 71 Things to A/B Test.
Once you decide on what you are going to test, Clone an exact copy of the email that you created in step 1. I like to name the 2 local emails Email A and Email B. In your Email B, implement the "test" whether it be Subject Line, From Line, or Content within the email itself. Your A/B Test should only be focused on 1 distinct variable (if you want to test more than 2 variables look into multivariate testing). Once you've finalized your changes to Email A and Email B, make sure you Approve the Emails.
Now, locate the "Send Emails" Smart Campaign you built in Step 1. You will need to reference your Audience in the Smart List tab by dragging a "Member of Smart List". Even though we are doing an A/B test, we still want to send all members of the Audience over to the Flow tab.
In the Flow tab, start by dragging a "Send Email" flow action over - and the click the "Add Choice" button. The purpose of the choice is to split our Audience in half, to accomplish this simply set the choice to Random Sample, is, 50. Then set Choice 1 to receive Email A, and the Default Choice to Email B. In layman's terms, take 50% of our target audience and send them Email A, send Email B to everyone else:
The whole point of doing an A/B test, is to be able to take away actionable insight and choose a winning variation of your test. Once you've run a number of A/B tests, you can join the winning variations together to make the perfect concoction of an Email. There are 2 ways to quickly see the results of your A/B test, you can either click on the Email tab of the Smart Campaign you used to send the emails, or use a Marketo Report. Both options will display the results of your in this format:
I would have chosen Email B as the winning variation in this email Test considering that Email B had a higher Open Rate and a lower Unsubscribe Rate, the difference in click rates were negligible, so Email B is the winner.
If you have further questions about A/B Tests, feel free to drop a line in the comments section.
Picture Credit: http://upload.wikimedia.org/wikipedia/commons/b/bb/Super-Famicom-Controller.jpg