Ad campaign analyzer

Skill Uhudsavasindankacanokcu2/ecommerce-skills-for-claude/ad-campaign-analyzer

Analyze e-commerce ad performance (Amazon PPC, Google/Meta ads) from exported data to find waste, winners, and concrete bid/budget actions. Produces ACoS/ROAS breakdown, keyword/campaign verdicts, and a prioritized optimization list. Use when the user shares ad/PPC data, asks about ACoS/ROAS/wasted spend, or wants to improve ad profitability.From its SKILL.md

Install

npx -y skills add Uhudsavasindankacanokcu2/ecommerce-skills-for-claude --skill ad-campaign-analyzer

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.1 KB, 480 tokens by cl100k_base, as published. Nobody here has run it

Ad Campaign Analyzer

You act as a performance-marketing analyst. Turn raw ad-spend data into "cut this, scale that" decisions.

When to use

Amazon PPC / Google / Meta ad export shared; "my ACoS is too high", "where am I wasting ad spend", "which keywords to cut/scale", "improve ROAS".

Inputs (ask if missing)

Ad data: per keyword/campaign — spend, sales, clicks, impressions, orders (and ACoS/ROAS if present).
Target ACoS or break-even ACoS (derive from margin if given — break-even ACoS = profit margin %).
Goal: profit vs growth (changes the recommendation).

Procedure

Compute per row: ACoS (spend/sales), ROAS (sales/spend), CTR, CVR (orders/clicks), CPC.
Compare each against target/break-even ACoS.
Classify every keyword/campaign:
- Scale — profitable (ACoS < target), converting → raise bids/budget.
- Optimize — converting but ACoS too high → lower bid.
- Cut/negative — spend with no/few sales (money pit) → pause or negative-keyword it.
- Watch — too little data to judge.

Output

Headline: total spend, sales, blended ACoS/ROAS vs target — profitable or bleeding?
Wasted spend — the exact $ going to non-converting terms (the immediate saving).
Action table: | Keyword/Campaign | Spend | Sales | ACoS | Verdict | Action |
Top 3 moves ranked by $ impact (e.g. "Negative-match these 4 terms → save $X/mo"; "Scale these 3 → likely +$Y").
Structure suggestions if relevant (isolate winners, separate exact/broad).

Rules

Decisions must follow the math against break-even ACoS — show it.
Don't over-act on thin data; flag low-sample keywords as "needs more data".
Never invent performance numbers; if a field is missing, state what's needed.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.