A/B testing in email is easy: split list, send variant A to half, variant B to other half, measure.
Most email tools build in sample-size calculators. Rule of thumb: test needs 500+ recipients per variant for meaningful data.
One test, one variable. Don't change multiple things at once.