Why you should A/B test your email campaigns

Email campaigns are an important aspect of many businesses marketing strategies, and with so many email campaigns constantly bombarding users, it’s more important than ever to cut through the clutter and achieve great results.

A/B testing can help you determine the best way to present your emails to ensure a high open and click-through rate.

What is an A/B test?

An A/B test allows you to set up and send two different email campaigns to a fraction of your audience and monitor which version receives the most opens or clicks. The test is run for a set time, and after that time is up, the most successful campaign is sent to the remaining subscribers.

Testing subject lines

One of the most common forms of A/B testing is to send the same email with two different subject lines and test which one gets more opens. You could test the subject lines  “Get $100 off email campaigns at Orion Creative” and “Hi Scott, huge sale on email campaigns this month,” and see which one gets the most opens. Did the more personable version with the subscriber’s first name (“Hi Scott…”) get a better result, or did including the exact discount price entice more people to open your email?

In this way, you’ve tested two different subject lines on a small sample of your subscribers, and determined which is more effective. That subject line is then sent out to the rest of the recipients, ensuring you’ve used the best possible subject line for your campaign on the majority of recipients. You’ll also know for your next send what type of subject line gets you the best results.

Testing from names

The From Name is another small change that can yield interesting results – you could set up a test with the very corporate sounding From Name “Orion Creative Solutions Pty Ltd” and the more personable “The Orion Crew.” Testing this in an A/B test can find out which one your audience prefers, and you can then use that going forward.

Testing multiple layouts

The final  A/B test type is to test two different layouts – for example, you could send two entirely different email designs to see which gets more clicks. One might have large, enticing images and the other could have smaller images but larger headings and call-to-actions.

You could also try an email with a two-column layout with lots of call-to-actions and other bits and pieces, while the second email could have a simple one column layout with just the most important information on display. You might find your audience appreciates less clutter and just wants you to get to the point!

Even small changes to the layout can have interesting effects. If you have multiple articles in each email, perhaps changing the way the articles are displayed might yield more positive results – having a larger, more obvious call-to-action may result in a greater click-through rate. Perhaps a larger or different font on your article headings might result in more clicks. Having your social media icons in a more prominent position may result in getting a few more followers – there are hundreds of tiny changes you could make to fine tune your campaign to your specific subscribers.

On top of all that, there is also the actual copy of the email to think about! An easy test you could perform is to try writing two versions of the same email – one in a professional, corporate way, and one in a more relaxed, friendly tone of voice. Even seemingly insignificant aspects of the copy, such as the length of article headings, can have a large impact on what does or doesn’t get clicked. When it comes to variations in your copy, the options are almost limitless.

Why test?

As every audience is different, testing these variables allow you to determine the best way to approach your specific subscribers. What works for a high-level IT corporation’s subscribers won’t work for a small, family-friendly bakery. It’s up to you to determine how to best get your subscribers’ attention and A/B testing is a simple, effective and cheap way to do it.

Running a number of tests on the subject line over a couple of months will soon turn the guessing game of choosing the “best” subject line into a science as you’ll have the hard data to back up your decisions. The same goes for email layouts and From Names. Find what works best for you and fine tune your campaign to achieve the results you want.

Case study

We recently ran an A/B test for one of our clients, and came up with some interesting an unexpected results! We ran a subject line A/B test to invite people to sign up for an event taking place across Australia and New Zealand.

Subject A: Your ticket to the new way of working – don’t get left behind
Subject B: [firstname], business has changed forever – don’t get left behind

The results are below (click to enlarge):
While the non-personalised version (Subject A) of the campaign received slightly more opens (199 as opposed to 187) , the real surprise was the click-through rate. Subject A received 91 overall clicks as opposed to Subject B’s 10.

Originally we expected the personalised version to do much better, as it seemed more approachable. However, the audience for this event is very professional/technical and perhaps preferred the more formal approach. As the open rates were so close, we’d need to do further testing (perhaps on layout or copy) to really get some useful data.

However when we compare this to the original version of the email campaign the week prior, we can see the A/B tested version led to a 2.5% open rate increase – which may not seem like a lot, but due to the size of the list it translates to an additional 200 people viewing the campaign! All for taking 5 minutes to create multiple subject lines and perform an A/B test. Not too shabby! If we continue to refine the campaign with multiple tests, we could be getting our email in front of a few hundred people extra each time.

How much will it cost me?

This depends entirely on what you’d like to test – something as simple as a Subject Line or From Name A/B test won’t cost you a cent, so there’s no reason not to get testing!

However, if you’d like to test multiple layouts, we’ll need to design and build them for you, which will range in price depending on how much needs to change. Get in touch if you want to discuss further or leave a comment on this article.