Split Testing Your Way To A Winner

Published by

Split Testing

With our recent update to Maxemail including improvements to our split testing feature, we thought we’d put together a simple strategy when testing this way to ensure your email marketing is onto a winner.

Split testing is a fantastic way to compare different versions of an email to ascertain which one performs better to create better results. These results can then be analysed to determine which version of the email to send to the rest of your subscribers.

First things first you need to decide what you are going to conduct your split test on. It’s important to try and only test one component at a time as cramming your emails with multiple variations of features will make analysing results difficult and less accurate.

What are results like with your current emails? Are you lacking in clicks or opens? If your open rate is particularly low you may want to look into running a test on subject lines to gain an insight into which subject line style encourages recipients to open your email.

If your open rate is quite high but click through rate is low you would be better off conducting a split test towards your main email content.

Some of the most popular testing variations include:

  • Call to Actions
  • Images
  • Headers
  • Content
  • Offers
  • Subject Lines
  • Preview Text
  • Layouts
  • Personalisation

  • Next is to work out how many subscribers you are going to send the test to. Most Marketers tend to send to their entire list as this provides the most accurate results however, depending on your email service provider, this could incur charges so may not be possible. The main thing to think about is having a decent size list for your needs.

    For example, if you are testing ‘Subject Line A’ against ‘Subject Line B’ sending to around 2,500 subscribers per subject line should provide you with decent and accurate results. If you are testing ‘Call to Action A’ against ‘Call to Action B’ in a targeted email then you could get away with under 1,000.

    If you are testing the waters with something fairly extreme you may not want to send to your whole list in-case of issues and problems. In this instance it would be wise to use a slightly smaller list of a few hundred, but the larger the better.

    Once your test has been sent it’s extremely important to wait for at least 24 hours before analysing the results, however waiting longer will provide a more accurate outcome. Some tests may take days or even a week to fully come through but it will include late openers.

    Now onto the analysing!

    Even if you were basing your test on one component, it’s a good idea to look at all of the results. Most email builders, including Maxemail, should have built-in reporting tools to help you analyse the results at a glance. If possible, take a look at the conversion rates – after all this is the main goal! Once you have worked out which email performed best, you can then send it to the remainder of your subscribers. However, if you are unsure about the results or if they were unexpected, there is no harm in running a second similar test to ensure the results are accurate.

    Key Take-Away Tips

  • Test one component at a time
  • Send to the largest list possible
  • Repeat tests for maximum confidence of your results
  • Don’t rush into analysing

  • Our new update to the Split Test interface within Maxemail includes improvements that will make creating split tests easier and bring them inline with the other recent Maxemail interface changes. If you would like to find our more about Maxemail get in contact with us today!

    About Emma Weir

    Marketing Executive at Emailcenter UK Ltd.

    Emailcenter_HEX