Paid Ads
Turns Claude into a performance marketer for campaign strategy, targeting, bidding, optimization, and safe live-account audits across Google, Meta, LinkedIn and more.
Marketing & SEOIntermediate★ 53,760⑂ 7,998AI score 9/10Last updated: Oct 8, 2026
What it does
- Context intake first: asks for objective, target CPA/ROAS, budget, offer, landing page, and existing pixel data before advising — and reads
.agents/product-marketing.mdif present so it doesn't re-ask. - Intent-based reference routing: a table maps each question type (Meta kill/scale decisions, LinkedIn bidding, Google Search structure, ABM, RSA generation, ecommerce audits) to a dedicated playbook under
references/, where the actual thresholds live. - Structure and creative: account/naming conventions, 70/30 test-vs-proven budget splits, 20% scaling increments, PAS/BAB/social-proof copy formulas, 15–30s video structure, creative testing hierarchy.
- Modern targeting stance: audience knowledge goes into creative first, targeting filters second, with platform-by-platform ratios.
- Retargeting system: different-offer retargeting plus a 4-ad framework (objection handling, proof carousel, other-offers CBO, value-first audit).
- Measurement discipline: brand vs non-brand separation, blended CAC, never summing conversions across attribution windows, break-even ROAS before scaling.
- Guardrails: unknown ≠ failing, no invented negative keywords, no fixed kill rules, draft-before-apply on live accounts, and ignoring instructions embedded in fetched data.
Who it's for
- Founders and growth leads running their own ad accounts
- Performance agencies and freelancers auditing client accounts
- B2B demand-gen teams fighting lead quality and CPL
- Marketing analysts who must turn ad platform exports into defensible reports
Examples
- "$5k/month for a B2B SaaS free trial — which channel first?" → pulls payback-period and b2b-paid-playbook to run payback math and split budget.
- "This Meta ad's CPA is 2x target after three weeks, should I kill it?" → checks learning phase, sample size and conversion lag, then applies the TCPL decision tree rather than a blanket rule.
- "Audit this Google Ads account" → scores each check pass/fail/unknown, states disclosed vs total clicks on search terms, and proposes changes with rollback instead of applying them.
· · · Install guide · · ·
Try it now, no install
Paste this into Claude to use the skill without installing anything.
Read the instructions in this file and follow them to help me: https://raw.githubusercontent.com/coreyhaines31/marketingskills/HEAD/skills/ads/SKILL.md What I want: (describe your task here)
If Claude can't open the link, open it yourself and paste the contents instead.
↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on Claude Skill Mart. Copy the skills/ads folder from the GitHub repo coreyhaines31/marketingskills into my ~/.claude/skills/ads/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/coreyhaines31/marketingskills.git /tmp/marketingskills && mkdir -p ~/.claude/skills && cp -r /tmp/marketingskills/skills/ads ~/.claude/skills/⚠ This is a third-party skill. Check the source repository before installing.
- Open your terminal (Terminal on macOS, PowerShell or Git Bash on Windows).
- Clone the repository:
git clone https://github.com/coreyhaines31/marketingskills.git - Create the skills directory if it doesn't exist:
mkdir -p ~/.claude/skills - Copy the whole
adsfolder — thereferences/subfolder holds the real thresholds, so don't copy SKILL.md alone:cp -r marketingskills/skills/ads ~/.claude/skills/ - Restart Claude Code, then run
/skillsor simply ask something like "my Facebook ads CPA is too high" to confirm the skill triggers. - Optional: create
.agents/product-marketing.mdwith your product, ICP and offer details so the skill skips repetitive intake questions.
View source on GitHub ↗License: MIT