Ali-expandings

Performance Marketing Skill — Data skill for Claude Code

Data community

Performance marketing data analysis skill.

How to install Performance Marketing Skill

This entry records only its repository, not the path inside it, so there is no exact command to give. Open Ali-expandings/performance-marketing-skill and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Performance Marketing Skill does

Performance marketing data analysis skill.

Alternatives in Data

Marketingskills — Marketing skills for Claude Code and AI agents 16.9k ★
MCP Server Chart — 🤖 A visualization mcp & skills contains 25+ visual charts using @antvis 4k ★
Posthog Analysis — Analyze PostHog usage data with correct filters, timelines, and caveats 1.6k ★

README

performance-marketing

A skill for analyzing large marketing and sales datasets. It works in two modes:

Quick: ad exports (Meta, Google, TikTok, GA4, Shopify) in, a ranked list of what to scale, cut, fix and test out.
Audit: CRM deals (HubSpot or any CSV export) joined with ad spend across periods. It detects growth patterns and funnel leaks, then builds a 12-tab Excel workbook and a 10-slide executive deck.

All math runs in scripts (DuckDB/pandas), so large files never need to be read whole, and every number in the final report is checked against the data.

It improves itself

Each run is graded by `scripts/selfcheck.py` (numbers traced to data, findings addressed, completeness, caveats, valid outputs). Lessons are logged with `scripts/learn.py`; lessons that recur are promoted into rules the next run reads, or into detector thresholds. `scripts/selftest.sh` reruns the pipeline on synthetic data with planted patterns so a change can't silently break detection.

Install

Copy this folder into your editor's skills directory as `performance-marketing`, then:

bash scripts/setup.sh
bash scripts/selftest.sh

Try it

python3 examples/make_audit_sample.py demo
python3 scripts/audit.py --deals demo/deals.csv --owners demo/owners.csv --meta demo/meta_ads.csv --profile demo/profile.json --out demo/out
python3 scripts/build_workbook.py --metrics demo/out/metrics.json --narrative examples/sample_narrative.json --profile demo/profile.json
python3 scripts/build_deck.py --metrics demo/out/metrics.json --narrative examples/sample_narrative.json --profile demo/profile.json

Per-client settings (currency rates, stage names, phone formats, sales owners and start dates, creative keywords) go in a profile: see `profiles/example.json`.