Transform marketing data into actionable insights through automated reporting, attribution analysis, and ROI tracking.

What This Skill Does

The Challenge: Marketing teams drown in data from GA4, ad platforms, email tools, and CRMs. Manual reporting is time-consuming and cross-channel attribution is complex.

The Solution: Analytics skill provides GA4 API integration, pre-built KPI frameworks, attribution models, and automated report generation. Includes Mermaid.js visualization and actionable recommendation prioritization.

Activation

Implicit: Activates automatically for Analytics Analyst, Campaign Manager, and Growth Specialist agents.

Explicit: Activate via prompt:

Activate analytics skill to [describe task]

Capabilities

1. Google Analytics 4 API Integration

Query GA4 data programmatically for custom reports and dashboards.

Setup:

npm install @google-analytics/admin @google-analytics/data

List accounts and properties:

node scripts/ga-list-accounts.cjs --summaries

Run custom report:

node scripts/ga-run-report.cjs \
  --property=PROPERTY_ID \
  --dimensions=country,city \
  --metrics=activeUsers,sessions \
  --start-date=30daysAgo \
  --end-date=today

API documentation:

• Admin API: references/ga-admin-api.md
• Data API: references/ga-data-api.md

2. KPI Framework and Tracking

Pre-defined metrics for acquisition, engagement, conversion, and retention.

Core marketing KPIs:

Framework loading: references/marketing-kpis.md

3. Attribution Models

Compare attribution models to understand channel contribution.

Available models:

• Last-click: Full credit to last touchpoint
• First-click: Full credit to first touchpoint
• Linear: Equal credit across all touchpoints
• Time-decay: More credit to recent touchpoints
• Position-based: 40% first, 40% last, 20% middle

Model guide: references/attribution-models.md

Prerequisites

For GA4 API access:

• Google Cloud project with GA4 APIs enabled
• Service account OR OAuth 2.0 credentials
• Credentials file in .claude/secrets/:
  • ga_service_account.json (service account)
  • OR google_client_secret.json (OAuth)

For report generation:

• Configured data sources (GA4, ad platforms, email, CRM)

Configuration

Credentials location: .claude/secrets/

• ga_service_account.json - Service account (recommended for automation)
• google_client_secret.json - OAuth (interactive setup)
• google_tokens.json - OAuth tokens (auto-generated)

Report templates: references/report-templates.md

Best Practices

1. Track leading metrics, not just lagging Monitor engagement metrics (leading) to predict conversions (lagging).

2. Compare apples to apples Use consistent time frames and segments. Compare year-over-year accounting for seasonality.

3. Statistical significance before conclusions Ensure sample size supports insights. Use confidence intervals for A/B tests.

Common Use Cases

Use Case 1: Weekly Campaign Performance Report

Scenario: Generate weekly report comparing channel performance.

Workflow:

1. Fetch GA4 data for past 7 days
2. Calculate key ratios (CAC, ROAS, CVR)
3. Compare with previous week and benchmarks
4. Generate Mermaid.js charts
5. Prioritize recommendations

Output: Markdown report with:

• Channel performance table
• Trend visualizations
• Top 3 actions with owners

Template: references/report-templates.md → Weekly Performance

Use Case 2: Attribution Analysis for Multi-Touch Campaign

Scenario: Understand how channels work together in 30-day customer journey.

Workflow:

1. Export GA4 path data
2. Apply attribution models (last-click, linear, position-based)
3. Compare channel contribution across models
4. Identify undervalued channels
5. Recommend budget reallocation

Output: Attribution comparison table with budget optimization recommendations.

Troubleshooting

Issue: GA4 API returns 403 Forbidden Solution:

1. Verify API is enabled in Google Cloud Console
2. Check service account has access to GA4 property (minimum Viewer role)
3. Confirm property ID format: properties/123456789

Issue: Reports show unexpected data spikes Solution:

1. Check for bot traffic (filter internal IPs)
2. Verify tracking implementation (GTM tag firing)
3. Compare raw events vs processed metrics

Issue: Attribution model results don’t match GA4 interface Solution: GA4 uses data-driven attribution by default. Scripts use rule-based models. Document methodology differences.

Related Skills

• Campaign Management - Campaign tracking and optimization
• SEO Optimization - Search traffic analysis
• Email Marketing - Email performance metrics
• Ads Management - Paid campaign ROAS tracking

Related Commands

• /analytics/dashboard - Generate dashboard snapshot
• /analytics/keywords - Analyze search performance
• /analyze - General data analysis
• /analyze:report - Create custom report