Master A/B Split Testing (and Email Marketing ROI) With This Simple Plan

Master A/B Split Testing (and Email Marketing ROI) With This Simple Plan

“Out of sight, out of mind” is sadly a little too true when it comes to email marketing, especially when people are new to it. It’s easy to send a campaign out and not really give it another thought as everyone’s attention quickly switches to the next campaign that needs to get pulled together. Everyone’s busy, so I understand why the focus is usually on getting a campaign out the door and not on evaluating how to make it better, nor analyzing the results of the campaign to see what did or didn’t work.

And that makes doing some simple testing to improve email marketing campaigns (and therefore ROI) that much more important. Simply testing subject lines can give a busy email marketer the opportunity to make one change that can increase a campaign’s performance, without taking up a lot of time.

As someone with over 20 years in the industry, I can assure you this doesn’t have to be hard. In fact, I’ve laid out a plan for you, to turn you into an inveterate tester. Simply start with making subject line testing a habit, and eventually you’ll find ways to test other tweaks as well, until you become a savvy email marketer who is continually optimizing campaigns for ever better results.

How to use A/B split testing to improve your subject lines

Sure, you could test many other elements of your email (because there are plenty of parts to just one email), but let’s start with subject line testing to build your confidence and get you into the habit of using tests to optimize campaigns. You might think, “Oh it’s just the subject line.” But if you only tested subject lines for each campaign, you would incrementally improve your email marketing with every email you send because your subject line plays a critical role in your open rate.

Besides, testing subject lines is relatively simple. All you do is come up with two subject lines you want to compare, labeling one A and the other B. Make sure they are at least a little bit similar so you’re comparing apples to apples and not apples to oranges. Two totally different subject lines won’t give you results you can work with. You want differences that are minor to help you tweak. (Unless you’re just getting started, in which case you might want to test two contrasting subject lines at first, because you’re starting in the dark and won’t learn much if you’re comparing two bad subject lines!)

Next, send emails to a small portion of your list. If your list has 10,000 names on it, choose maybe 100 to 200 names. Split that small portion in half and send one the email with the A subject line and the other the email with the B subject line.

Finally, see which one gets the higher open rate. That is your winner and that is the subject line you’ll use for the campaign that’s sent to the whole list.

Does that sound complicated? I hope not, and if you are intimated by the idea of testing, only try this for a couple of campaigns and you’ll realize it doesn’t add much to your long to-do list. Besides, isn’t a higher open rate worth a little effort?

After you’ve done it a couple of times and you’re comfortable with A/B split testing your subject lines, you’ll want to do this for every campaign, until it becomes second nature. With each test, you should be able to further refine your subject lines. (Note: Some ESPs will automate the testing for you. For example, with iPost, you can set up the campaign and then the platform will automatically send the winning email to the remainder of the list. Now that’s easy!)

Moving beyond only subject line testing

Next? It’s time to expand your testing beyond subject lines, because every tweak you can make has the potential to incrementally improve your marketing. For example, let’s say you test your calls to action (CTAs) and CTA A gets a three percent click-through rate while CTA B gets a 4.1 percent click-through rate. You might be thinking, “That’s only 1.1 percent. That’s hardly a difference.” Wait. Back that up before you make an assumption to see what the 1.1 percent really means. Let’s use the 10,000 names email list we used above and work through the math:

  • Say you send to 10,000 names and your deliverability rate is 91 percent, so 9,100 emails are delivered.
  • Then say you’ve tweaked your subject line so now you’re getting an open rate of 21 percent, meaning 1,911 people opened your email.
  • At the three percent click-through rate, only about 57 people clicked through while at the 4.1 percent open rate, about 78 people clicked through.
  • That means 21 more people will click through with CTA B.
  • Now say 30 percent of the people who clicked through actually convert and make a purchase and your average purchase is worth $100.
  • If 21 more people click through, that means 7 more people will buy and that means you’ll increase your sales from that one email by $700…all because you tested your CTA and used the better one.

These numbers are for illustrative purposes only, but I hope they make the point that the little improvements and tweaks do convert to actual dollars and email marketing ROI!

In addition to your subject lines and calls to action, other email elements you can easily test include:

  • Preheader text
  • Heading
  • Body text
  • Deadlines vs. no deadlines
  • Testimonial vs. no testimonial
  • Types of photographs
  • Bullets vs. numbers
  • The placement of your call to action
  • Button color
  • And countless other details

One caveat: You don’t want to test more than one element at a time, or else you won’t know what’s causing the better performance. But you should test something with every single email campaign you send. Strive to make it a habit and it will become one, one that serves you well.

But simply start with your subject lines. Subject line testing is easy to do, and it will lead to other kinds of testing later as you gain confidence, make it part of your processes, and come to rely on the insight you gain…and the email marketing ROI you enjoy.

Photo Credit: webzin Flickr via Compfight cc

About author

Your email address will not be published. Required fields are marked *