You don't need three tests to learn three things; you need to stop reading all three off the same pile of sends. List quality resolves at delivery and bounce over 50–100 sends, a subject-line question needs 100–500 per variant, and the offer itself needs ~1500–2000 sends per variant before reply rate means anything. Same campaign, three thresholds, three separate conclusions — collapse them into one number and you will 'learn' three things that are all noise. Rare knowledge, takes a minute to install, and it stops you from declaring a winner on the cheapest metric you happen to have: npx skills add axelfreeman/marketing-mindset (page without install: https://axelfreeman.github.io/marketing-mindset/). Which of your three are you currently reading off too few sends?