Email A/B Testing For Conversions

Lindsey Sherman - Tue, July 15, 2014

Your emails are about to become a lot more effective.

Last week, SharpSpring launched our newest feature, A/B Testing for Email. To celebrate, we’ve compiled some best practices and testing ideas for you to get started making your emails work harder for you.

A/B Testing, also known as split testing, shows two versions of marketing variables and lets your consumers decide which is the “winner,” based on which version they react better to (more opens, more clicks, etc). These variables can be part of your website design, landing page, display or text ad, lead form, or in this instance, the emails that you are sending.

Providing different versions of your email to statistically significant groups of subscribers allows you to measure the effect each of these versions has on your goals. As a result, when you optimize your email marketing strategy you are working off of actual data, not just a hunch.

Goals

The first, and probably most important step, is determining your goal for the test. There are many variables that you can change in your email marketing campaign that will impact how effective your emails are. Do you want to get a higher open rate? You can try testing your subject line to improve it.

Here is a list of some common goals to begin testing around:

  • What is the open rate?
  • What is the click-through rate?
  • What is the on-site conversion rate from the email traffic?
  • What is the bounce rate?

Segmentation

Once you have determined your goal, it’s time to segment. With SharpSpring, this step is accomplished by creating Dynamic Lists.

There are a few methods for segmenting your lists for the test. No matter which method you choose, the segments need to be randomized in order to ensure the most accurate results. SharpSpring does the randomization within your Dynamic List for you, which allows you to focus on building relevant, well-segmented lists.

The first, which is the most simple, is a straightforward 50/50 split, with one half receiving one version of the email and the other receiving a slightly modified version. 

Another option is to do a test on a smaller percentage of your list, and then send out the winning variable to the rest of your list. This is a good method to use if you’re testing something dramatically different from previous emails, or if you are running a limited time offer. Make sure that your initial test is going out to enough subscribers to gain statistical significance, or your results could be skewed or inconclusive.

Here are a few more tips to get the best segmentation:

  • Remove inactive users that can drag your results down
  • Highly active users should be split evenly between the two variations, or not included in the test. These users are more likely to engage anyway, so separating them out will ensure that too many in one variable won’t skew results. 

Testing…1, 2, 3…

Now is the time to pick the variable you’d like to test! Below are a few ideas to get you started, but the possibilities are endless when it comes to A/B testing ideas.

  • Personalization in headline or body copy? Or a more generic email?
  • What time of day performs best?
  • What day of week performs best?
  • Punctuation – does a question in the headline work better? Exclamation points?
  • Length- short and sweet, or long and detailed?
  • Tone– Straightforward? Funny? Mysterious?
  • Offer in the headline? In the copy? (Free v. percentage off, demo v. trial)
  • Call to Action – Buy now? Purchase Now? Add to Cart? Play around with various calls to action depending on your end goal.
  • Change the location of the call to action button throughout the email (top v. bottom, left v. right)
  • Change the color or shape of the call to action button
  • Image heavy email, purely text or a mix?
  • What kind of images you show- Stock images? Products? Employees? Cute animals?
  • Who the email is from- is it more impactful from the CEO? Marketing Ddirector? Sales person?
  • Utilize your Dynamic Lists that you’ve created in the SharpSpring interface to take your testing a step further, and test specific variables with different segments.

These are just a few of the many ideas that you can start testing with, and a great first step in making your email marketing work more efficiently for you. Just remember to only test one of these at a time, or you won’t be able to differentiate which change is affecting your metrics.

Now that you’ve got some great ideas on what A/B testing is and how to make it work for you, get started! We’re excited to hear about your successes with our new tool, or ideas for improvements.

  • Author: Lindsey Sherman

  • Previously, Lindsey played an integral role in the digital marketing teams at reputable marketing agencies. In her current role, she is responsible for marketing SharpSpring to an online audience. Overseeing and implementing PPC, display and social media advertising, she is adept at identifying unique selling points and industry specific challenges (as well as solutions) for a variety of markets.