Gut instinct is a poor guide for email optimization. A/B testing — running controlled experiments on specific variables — is the only way to know what actually works for your audience, your product, and your market.
Test in order of impact: (1) Subject line, (2) Opening sentence, (3) Value proposition, (4) Call to action, (5) Email length. Most teams never get to #5 because the first four are already enough to improve performance dramatically.
Send each variant to a minimum of 100 recipients from the same audience segment. Change only one variable per test. Run the test to completion (all 5 touches) before measuring final reply rate — measuring after Touch 1 misses the replies that come later. Document every test in a shared spreadsheet with: hypothesis, variant A, variant B, sends per variant, open rate, reply rate, winner, and what you'll test next.
Testing too many things at once: If you change the subject line, opening, and CTA simultaneously and get a different result, you don't know which change caused it. One variable per test, always.
Declaring a winner too early: With fewer than 50 sends per variant, your results are statistically meaningless. Discipline yourself to wait for proper sample sizes.
Not documenting results: The team that doesn't document test results repeats the same tests indefinitely. A shared swipe file of winning variants is a compounding organizational asset.