Split testing is such an important and overlooked part of email marketing, maybe people feel overwhelmed… You can split test subject lines, fine, but what do you test, how often should you test, what’s the best practice and how should you start?

One piece of advice. Don’t overthink it, just start small and learn as you go. Split testing is such an amazing resource within your email program, so just start today and test ‘something‘. Small tests will tell you so much more than big ones, here’s why… If you test a single word in a subject line 3 times and there’s one clear winner, you know exactly what the result was. Whereas if you test an entirely different subject line, sure you know one worked and drove 0.1% more opens (yippee), but how do you know what the success factor was?

This collaborative post from some of our friends and fellow Mailchimp partners is here to help you with all of this.

Sit back and learn.

Emily Ryan
Emily Ryan Co-Founder and Email Strategist at Westfield Creative Contributed to 2 posts

Start with testing the “from name”

You may want to dive in testing different send times and subject lines, but I always like to nail down the best “from name” first. It’s one split test that you can tell the results from pretty quickly.

Your “from name” for your emails is really important for getting people to open your email. Should your email be from “Nike” or “Emily at Nike” or “Emily, Nike & Co.” — just as an example. For small businesses, I think it’s important to test an actual name — people connect more with a person than maybe a company name. So test both options and see which has the higher open rate.

Make sure you test this over several emails to see if you see a trend– you might be surprised at the results of the A/B test.

Once you’ve nailed down the from name, THEN dive into determining great send times and subject lines.

Test the subject to find what works best for your audience

A/B testing the subject of an email campaign is always useful in order to understand your subscribers’ preferences and improve the open rates and overall results of your campaigns. Here some test you can run:

  • Personalization in the subject with the first name of your subscribers (provided you always have one or you can use a default value) vs an impersonal subject
  • Use / don’t use emoji 
  • A subject that anticipates the content vs a “repetitive but recognizable” subject (eg. My Topic Bullettin #N)

A smart tip I got from a colleague: write down several versions of the subject, A/B test the first two or three, then use the winner and the remaining versions to create an A/B test campaign for those who didn’t open the first one.