eduardocornelsen/full-funnel-ai-analytics
View on GitHub ↗Full-Funnel AI Marketing Analytics. A modern data stack powered by dbt MetricFlow and MCP. Natural language insights across Google/Meta Ads, CRM, and 5 data warehouses. Includes XGBoost lead scoring and a $0/mo architecture.
20 ★4 forksPythonUpdated 25d ago
What you need to know
Natural language marketing analytics platform powered by MCP, dbt Semantic Layer, and ML lead scoring, working with Claude Desktop, OpenCode, Gemini CLI, and Antigravity IDE.
Install
pip install -e . fullfunnel demo docker compose up
Usage
- •Type /marketing, /attribution, /pipeline, /campaign, or /traffic, or ask open-ended questions about marketing funnel data
Key features
- ✓Natural language querying of marketing data
- ✓dbt Semantic Layer (MetricFlow) governed metrics
- ✓6 mock MCP servers (BigQuery, Google Ads, Meta Ads, GA4, HubSpot, Salesforce)
- ✓Multi-touch attribution models
- ✓XGBoost lead scoring with FastAPI and MLflow
- ✓n8n automation
- ✓Zero-drift golden metrics validation
- ✓Works across 5 warehouses and 4 AI clients
- ✓$0/month base cost
Best for
Marketing analytics and revenue operations education and demos
Caveats
- ⚠Uses mock/synthetic data by default; swap to production connectors for real data
Platforms: Python · DockerClients: Claude Desktop · OpenCode · Gemini CLI · Antigravity IDE
Documentation ↗Reviewed 2026-08-11
Topics
agentsai-analyticsantigravitybigqueryclaudedata-warehousedatabricksdbt-coreduckdbgeminigenaillmmcpmetricflowmodel-context-protocolopencodepostgressemantic-layersnowflakesupabase
- Stars
- 20★
- Forks
- 4
- Language
- Python
- License
- MIT
- Created
- 2026-03-15
- Last push
- 2026-08-10