How to A/B test your emails without overthinking it

A/B testing, sometimes called split testing, means sending two versions of an email to small parts of your list, seeing which performs better, then sending the winner to everyone else. It replaces guessing with evidence.

Most email marketing tools have A/B testing built in on paid plans. Even if yours does not, you can test over time by comparing similar emails.

What you can test

• Subject lines. The most common test and often the easiest win.
• Preview text, the short line shown next to the subject.
• Sender name, such as your name versus your brand name.
• Send time, such as morning versus evening.
• Content, such as a short email versus a longer one.
• Call to action, such as different button text or placement.

Test one thing at a time

This is the golden rule. If you change the subject line and the send time together, you will not know which change made the difference. Change one thing, keep everything else the same.

Pick the right number to judge by

Decide in advance what "better" means:

• For subject lines, opens are the usual measure, though they are less reliable than they used to be. If you can, look at clicks too.
• For content and buttons, use clicks.
• For anything that leads to a sale, use conversions, meaning people who actually bought or booked.

Make sure the test is big enough

Small tests can mislead. If 50 people see version A and 50 see version B, a few clicks either way can be pure chance. As a rough guide:

• If your list is small, test with your whole list split in half, then use what you learn next time.
• If your list is larger, many tools let you send each version to a smaller sample first, such as 10 to 20 percent each, then send the winner to the rest.

Let a test run long enough for people to open their email. A few hours is the minimum. Many tools let you choose how long to wait.

Simple tests to start with

Subject line: question vs statement

Sender name: person vs brand

Button text: generic vs specific

Length: short vs long

Keep a test log

Write every test in a simple sheet: what you tested, the two versions, the result and what you learned. After a few months, you will have your own rule book built from your real readers, which beats any general advice.

What not to test

• Tiny changes that would not matter even if they won, like one extra comma.
• Tricks that mislead readers, like fake "Re:" in subject lines. They may win once, then hurt trust.
• Too many versions at once on a small list.

The short version

Test one thing at a time, judge by the right number, make sure enough people see each version and write down what you learn. A/B testing is not about one big win. It is about getting a little better with every send.