Test every single subject line

“Your subject line is important because testing and optimising them drives response rates like nothing else. But, there’s a common problem with subject lines, over 70% of email marketers don’t split test their subject lines very often.” (Source: Smart Insights)

You might want to test different elements of your campaigns from now and then, like the call-to-action or the use of personalization, but the subject is the one element that should be tested every single time you send one out, regardless of the type of campaign you’re sending.

Luckily, this is the easiest thing to test in email marketing, and ESPs have made it very easy.

The Goal

The goal, as we stated at the beginning of this module, is one: Get more people to open your emails. Simple.

What to test?

There are different things you can test in subject lines:

  • The length
  • Should you include numbers?
  • Using personalization
  • Should it be a statement or a question?
  • Should it have a positive or negative tone?

Set some rules

In order to get the most out of your test, you should establish some rules:

  • What percentage of the list will receive each variant. It should be a small fraction of your list and both variants should be sent to the same amount of subscribers. Most people do 10% – 10%
  • Establish how long you will wait until you have a winner. You don’t want to declare a winner prematurely. Mailchimp suggests 4 hours but I know some people wait 24 hours
  • Establish what will determine success.

How to A/B Test your subject line


This homework should NOT wait, you need to start testing your email subject lines immediately, on your next campaign.

Here is the process:

  1. Decide what you are going to test (length, personalization, statement or question, etc.)
  2. Determine what percentage of your list will receive each variant
  3. How is the winner decided? In this case will be by open rate
  4. How long are you going to allow to get the results

When you look at the results, it’s very interesting to see what the difference was, some tests will be close (55% to 45%), but there will be cases when you’ll get a clear winner (75% to 25%). It’s important to learn from this experience and try to figure out what was obviously wrong with the losing subject line.

Once you’ve ran your first test, come back to this lesson and share your results.

  • If you’ve tested your timing and determined that you should send your email at a very specific time like 2pm on Thursday for optimal open rates, wouldn’t this make testing a subject line and waiting for 4 – 24 hours a bit tricky? (b/c i would have to test it on another day entirely, or several hours before my ideal open rate) – Couldn’t this skew my test results? Have you ever tested a week in advance to avoid this conundrum or am i being too picky since we are in essence only testing subject line? (which in theory should yield a clear result – even if the time sent is less than optimal?) – Thx!