How an email marketing company got it wrong (but still got it right)

A surprising lesson from our very first send-off using Emma's new A/B/C split test feature

Editor's note: Scroll to the bottom of this post to request early access to Emma's new split testing feature.

A couple weeks ago, we tested out Emma's new subject line split test feature with a mailing to customers about fall holiday designs, and we learned just how hard it is to predict the behavior of our email audience.

There. I said it. We're an email marketing company – subject line experts, some may say – and we were surprised by the results of our first official subject line split test.

Here's how it went down. We started by coming up with a list of subject line candidates. After jotting down whatever came to mind, we reviewed our list and pointed out the distinguishing characteristics of each. Some were short and punchy, some long and explanatory, some teased a video, some used the word "free" or mentioned Emma by name.

The coolest part? We didn't have to pick just two subject lines to test -- Emma's split test feature lets you test a third variation, so it's a-ok to stick a wildcard in there and see what happens. Here were our picks:

Variation A: Emma's fall templates are free and ready in your account

Variation B: Free fall templates + a video of how we made the email you're about to open

Variation C: Fall design is here: free templates + affordable custom design

If we didn't have the option to split test, we would've sent Variation B to everyone. It's a little longer than your average subject line, and previous subject line tests (we've done a few over the years, but without the use of this handy feature) have indicated that our audience responds well to subjects packed with details. Plus, it mentions video, which we figured would entice recipients to open and click.

As soon as the split test was launched, Emma staffers postulated which subject line would perform best. We huddled around the whiteboard casting our votes on Post-Its and analyzing words and punctuation with unbridled scrutiny.

I'm not the only one who thought Variation B was our ticket to a high open rate.

The result? Well, we segmented our audience into two groups, customer contacts and non-customer subscribers to our emails, and we ended up with two winning subject lines:

Customers preferred Variation A: Emma's fall templates are free and ready in your account.

A screenshot of our account, mid-test. Customers favored a subject line that mentioned that we added free fall templates to all accounts.

Our broader email community preferred Variation C: Fall design is here: free templates + affordable custom design.

Our audience of email subscribers preferred the subject line that made a broader mention of fall design at Emma.

So there you have it. Sometimes you don't know your audience as well as you think you do. But isn't it great to have a tool that helps you gain that insight? I sure think so. 

And you sure can have a ton of fun being wrong. 

Can't wait to try split testing your own email marketing campaigns?

We're looking for a few eager folks to try out split testing alongside us. Just email us your username and we'll add this feature to your account right away.