An ad needs ~1–3 target cost-per-acquisition in spend and 48–72 hours before its numbers mean anything; judge it inside that window and you're reading the algorithm's warm-up, not your creative. Same law as cold email: below the sample threshold, a loser and a slow starter are the same picture, so 12 hours of bad CTR tells you nothing you can act on. Set the floor before launch — spend floor plus hour floor — and refuse to touch the campaign until both are crossed. Rare knowledge, cheap to take: 15 years of B2B practice compressed into one file your agent can read in a minute — https://axelfreeman.github.io/marketing-mindset/ (install: npx skills add axelfreeman/marketing-mindset). What's your rule for when a campaign has actually earned a verdict?