A/B Testing
View a walk-through of the A/B Testing process in the emfluence ClickLearn section here.
You can use the A/B Testing available within the Compose step to create up to five separate versions of your email, schedule out your test sends, and either schedule or manually send a winner based on a specific metric. NOTE: All of the test sends must be sent at the same time. If you are testing day of week or time of day you will want to use the Segment Groups tool in the Contacts section.
In the Compose step of your email you can turn on A/B Testing at any point by clicking the Create A/B Split button at the top of the page. When you click this button the system will create a current copy of your original email (which will then become Version A). Further copies will always be created on the current state of Version A. If your segments are going to be largely identical, you will probably want to create your email prior to duplicating and creating your segments (rather than have to recreate the same content on each one).
You can click into the different test segments by clicking on the letter of the segment you wish to edit. You can continue to create copies of your A version up to a maximum of 5 different test segments.
By default, your individual fields are locked down. You can unlock them simply by clicking into the input field. NOTE: Best practice would be to only test one variable at a time.You can set different Google Analytics variables for your different test segments, however you cannot add the segments separately to different emfluence Campaigns.
On the Recipients tab, you will choose your groups, just as you would for a normal manual email. Next, you will set your A/B Split Contact Distribution using the sliders provided (with the remaining segment going to your winner). NOTE: If you choose segements too small to be statistically relevant you will get a warning on the Send step. However, as this is a warning rather than an error, you can still proceed and send your segments.
On Delivery Options, you can choose to either send both the tests and winner manually, or schedule your test sends and either send your winner manually or schedule your winner. You can only schedule your winner if you schedule your test sends. To set a schedule on your test emails, toggle to Set a delivery date/time, fill in the day and time in the pop-up lightbox, and click Apply. Once applied, you will then have the option to also set a delivery/time for your winner email. If choosing to set your winner delivery, you will need to chose which metric to judge the winner (if you are sending manually, you can choose any segment you wish at the point you are ready to send).
Once your test sends have gone out you will see metrics provided for the test emails on the Summary page.
Prior to it being sent, you will find a Send Now button under the Winner Email, along with a scheduled date and winning metric (if it was scheduled). If you chose, you can cancel the winner email and not send to the remainder of your list. This action cannot be undone.
The Results below will allow you to compare your test segments. You can drill into each segment by clicking the letter for more data about each send.
