A/B Testing Email Subject Lines: A Simple Framework

Subject lines have an outsized effect on whether your email gets opened, which makes them a natural place to test. Done well, A/B testing turns guesswork into a steady stream of small improvements. Here is a framework you can reuse for every campaign.

Change one thing at a time

Test a single variable so you can attribute any difference to it. If you change the length, the tone and the offer all at once, a winner tells you nothing you can repeat. Pick one lever: length, personalisation, a question versus a statement, urgency, or an emoji.

Split the audience fairly

Send each version to a random, comparable slice of your list at the same time. Sending at different times introduces timing effects that muddy the result.

Use a big enough sample

Small tests produce noisy results. If your list is tiny, a few extra opens can flip the outcome by chance. Give each variant enough recipients that the difference is meaningful, and be honest when a result is too close to call.

Measure the outcome that matters

Opens are the obvious metric for subject lines, but many clients now pre-load images, which inflates open counts. Where you can, look through to clicks and conversions, because a subject that wins opens but attracts the wrong readers is not a real win.

Ideas worth testing

• Short and punchy versus descriptive.
• With and without the recipient’s name or company.
• A question versus a direct statement.
• Curiosity versus a clear, specific benefit.

Keep a simple log of what you tested and what won. Over months, those small, compounding lessons become a subject-line playbook tuned to your own audience.