Email A/B testing: what to test, how long and how to read results
A practical A/B testing guide for email: test one variable, pick a metric that matters, ensure enough sample and avoid false winners.

Test one variable at a time, choose a primary metric such as click rate or revenue per recipient, and run the test until you have enough data rather than stopping at the first lead. Opens are unreliable after Apple's privacy changes.
What to test
- Subject lines and preheader; see preheader guide.
- Hero image and headline.
- CTA wording and button style.
- Layout: one vs two columns; see columns guide.
- Send time.
Choosing the metric
Because Apple Mail Privacy Protection inflates opens, judge tests on clicks, conversions or revenue per recipient. Open rate is only useful for subject-line tests with caution.
Sample and duration
Small lists produce noisy results. Use a statistical significance calculator or your platform's guidance, and test for long enough to capture normal weekday variation.
Keep a log
Record hypotheses, results and learnings so you do not repeat tests. Roll winners into your templates.
How EmailDesign helps
We build measurable programmes: clean tagging, sensible segments and designs you can test and improve over time. See Ecommerce email strategy, or pricing for fixed quotes from £95 per email.




















How big does my list need to be?
Larger lists give more reliable results; small lists should test bigger changes and be cautious.
Can I test more than one thing?
Multivariate tests need much larger audiences. Test one variable at a time.
Should I test every send?
Test regularly on key sends and flows.
What if the test is inconclusive?
Treat it as no difference and move on to a bolder test.
Can you build this for us?
Yes. EmailDesign designs and hand-codes emails for UK brands from £95 per email, tested across 60+ clients and set live in your platform.
Ready when you are.
Tell us what you're sending and where — fixed quote within one working day.