Ad spend looks healthy until downstream signals reveal it's not. This episode reframes rescue work as a hypothesis engine: test creative+landing pairs (not creatives alone) to expose mismatch-driven waste quickly. Hosts Gustavo (measurement) and Michelle (brand) guide a compact 72‑hour runbook: a fast-spend audit to pick a lone line item, two low‑dev paired variants (creative headline + landing hero match), immediate tagging and simple minimum-sample rules, and a conservative scaling checklist that prevents overreacting to novelty. We include an anonymized 30‑second guest soundbite from a team that doubled qualified demos, plus a clear low‑traffic contingency (pooled sequencing and quick user calls). Jargon is kept minimal and defined—e.g., “micro‑conversions” (early signals like demo starts or signups)—and listeners get a short URL in episode notes to download the budget rescue checklist and CSV template for immediate use.