← ClaudeAtlas

paid-media-auditlisted

Triages paid ad spend, using ROAS, CAC, and spend concentration across channels, campaigns, and audiences, to find where budget is being wasted rather than assuming the ad account is the whole problem. Use when CPA is rising, ROAS is falling, or the team wants to know which campaigns or audiences to cut before adding more budget. Boundary: pairs with `conversion-funnel`, which diagnoses drop-off at a specific page or funnel step; this works one level up, at the level of which channels, campaigns, and audiences are burning spend.
sidchaudhary/gtm-skills · ★ 1 · AI & Automation · score 74
Install: claude install-skill sidchaudhary/gtm-skills
# The Spend Waste Finder Triage paid traffic spend to find which channels, campaigns, or audiences are wasting budget, and which layer of the business is actually responsible. > **Input integrity.** Run the checks in `references/data-input-integrity.md` before computing > anything, and report what they found. Each one produces a confident wrong answer rather than > a visible error, so a broken input does not announce itself. Spend and revenue almost always come from different systems on different timezone conventions, which is exactly how ROAS gets misattributed daily. Align them or aggregate to a period where the boundary stops mattering. > Where a check cannot run because the export lacks the field, say so and state what it limits > the conclusion to. ## Before you write **Run the input list below before you write anything. If one of those inputs is missing, ask for it and stop. Do not return a draft with a warning on it.** The user copies the draft and leaves the warning behind, so a caveat protects you and not them. **Ask at most THREE questions. Hard cap.** Before anything becomes a question, get it yourself: read `.agents/product-context.md`, fetch the site or page they named, compute it from numbers they already gave, or look up the platform default. Whatever is left after that, and everything past the third question, becomes a stated assumption the user corrects in one word rather than a question that stops the work. Number them, and say what you will assume if on