A/B Testing for Improved Email Marketing

By: Jake

Many small business owners who manage email lists of their clients have asked us how they can improve the success rate of their mailings. Indeed, sometimes it can feel like you’re sending messages out into the Internet ether and crossing your fingers that you’ll be lucky enough to get some kind of response. One of the first questions we ask in return is whether these businesses have tried A/B testing.

 

Some version of A/B testing has been around since the dawn of marketing. While it can come off as one of those things that “professional” marketers do, the truth is that if you send any volume of email, A/B testing can help you hone in on your core audience and drive engagement.

 

Before you get a chance to say, “Jack, I run a small business, not a statistics laboratory,” let us assure you. It’s actually quite simple.

 

What is A/B testing?

 

The idea of A/B testing comes from one of the fundamental problems of content marketing: it’s nearly impossible to judge what content will resonate with your followers until they actually start responding to it. But, if that’s the case, how do you know if you’re sending out the right type of emails?

 

You can track general response levels over time, but as you might imagine that takes a long time and too many people forget to do the necessary work. Or, you could try A/B testing.

 

In a nutshell, A/B testing involves splitting your email list in half, and sending two different versions of your email message (two different subject lines, two different calls to action, etc.) to see which one has a higher response rate. From there, you adapt your email marketing strategy to reflect these findings for any future messages you send.

 

What you’ll find out you might surprise you: It’s entirely possible that your followers are twice as likely to respond to your email if your subject line is phrased as a question. Or, you might find that they engage much more if you only cursorily mention your product, rather than going with the hard-sell approach. Sometimes, silly things like different formatting (two columns vs. one) will give you dramatically better results. The point is, much of this stuff is arbitrary, and you need A/B testing in order to figure out what works.

 

How to Do It

 

In order to do email A/B testing properly, you’ll have to send your emails from an email-marketing platform that allows you to track open rates and click through rates for your emails (these are the numbers you’ll compare to each other). These platforms tend to be very affordable for small businesses with small mailing lists. If you don’t already use one, two worthwhile options to look into are MailChimp and Emma.

 

Most of these platforms also have a neat feature that allows you to get the benefits of A/B testing in your very first mailing: When you schedule an email blast, it will first go out to a “test group” of 10%-40% of the recipients, split evenly between the A and B version of the message. The results are then tallied, and whichever message got a higher response rate in the test group gets sent out to the remaining 60%-90% of your list, which should improve your response rate.

A/B testing is one of the easiest ways to hone your email marketing strategy, so we recommend that all small businesses try it at least for a short period. As you can see from the above, modern technology has made the A/B testing process quite simple, keeping the time requirement low and the potential upside high.