How to A/B test your email approaches
Stop guessing what works. Test two approaches and let the data decide.
When to A/B test
- You have two ideas for opening lines
- You're debating casual vs professional tone
- You want to try different CTAs
- You're not sure if pain-focused or benefit-focused works better
Creating a test
Go to A/B Tests and click Create Test.
Fill in:
- Product — which product this test is for
- Name — e.g., "Pain vs Question Opening"
- Description — optional notes
Then define your variants:
Variant A:
- Subject line
- Email body
Variant B:
- Subject line
- Email body
Click Create.
How it works
Once active:
- New drafts for this product randomly get Variant A or B
- Each variant tracks: sent, opened, replied
- Results update in real-time
Reading results
The test shows:
- Sends per variant
- Open rates
- Reply rates
- Statistical significance
When there's enough data, the system identifies a winner.
Declaring a winner
Click Complete Test when you have enough data (20+ sends per variant recommended).
The system calculates:
- Which variant performed better
- Confidence level of the result
- Recommendation for future emails
Applying learnings
After a test concludes:
- Update your product description or approach based on the winner
- Create a new test for the next variable
- Keep iterating — each improvement compounds
Test one thing at a time. Subject line in one test, tone in another. Otherwise you won't know what made the difference.