# MCP Analytics

> MCP Analytics is a statistical analysis and reporting platform. Upload CSV data or connect live sources (GA4, Google Search Console), get interactive HTML reports built on validated R scripts with AI insights and PDF export. Available as a web app and via the Model Context Protocol (MCP) for AI assistants.

Every analysis runs on peer-reviewed R statistical packages — not AI-generated code. Same data + same module = same result every time. Reports include interactive Plotly charts, AI interpretation, R code, methodology descriptions, and citation formats.

## Key Capabilities

- **Statistical Testing**: t-test, ANOVA, chi-square, Mann-Whitney, Kruskal-Wallis, ICC, ANCOVA, Holm-Bonferroni
- **Regression**: Linear, Ridge, Lasso, Elastic Net, Logistic
- **Machine Learning**: Random Forest, XGBoost, K-Means, DBSCAN, PCA, Isolation Forest, Naive Bayes
- **Time Series**: ARIMA, Simple Trend, Revenue Forecasting
- **Business Analytics**: ROAS / Ad Spend Efficiency, Marketing Attribution, Price Elasticity, MRR Analysis, Churn Prediction, RFM Segmentation, Product Profitability
- **Research Methods**: ICC, ANCOVA, Cox Proportional Hazards, Survival Analysis, Difference-in-Differences
- **Engagement**: GA4 Page/Source Analysis, GSC Rankings/Quick Wins/Content Performance

## How It Works

1. Upload a CSV or connect a data source (GA4, Google Search Console)
2. 5-signal semantic matching finds the right analysis module for your data
3. R-based analysis runs in Docker, producing an interactive HTML report
4. Every result gets embedded in a searchable knowledge layer — query across all past analyses

## Pricing

- **Free**: 2,000 credits on signup (~15 reports)
- **PAYG**: Credit packs from $4.99/500 credits. Haiku AI.
- **Academic Bundles**: 10 reports/$19, 20/$29, 30/$39. Sonnet AI. Reports never expire.
- **Business Bundles**: 10/$29, 20/$49, 50/$99, 100/$149. Opus AI. Reports never expire.
- **Academic Pro**: $199/month for 200 reports
- **Business Pro**: $499/month for 500 reports

## MCP Integration

Install in any MCP-compatible client (Claude Desktop, Cursor, Windsurf, Claude Code):

```json
{
  "mcpServers": {
    "mcpanalytics": {
      "command": "npx",
      "args": ["-y", "@mcp-analytics/mcp-analytics"],
      "env": { "MCP_ANALYTICS_API_KEY": "YOUR_API_KEY" }
    }
  }
}
```

Alternative: Direct HTTP with API key (`https://api.mcpanalytics.ai/mcp/api-key`) or OAuth2 (`https://api.mcpanalytics.ai/auth0`).

## Public Discovery (No Auth)

Browse the full tool catalog without an account — the `/mcp/api-key` endpoint works without a key for discovery:
- MCP protocol discovery: `POST https://api.mcpanalytics.ai/mcp/api-key` with `{"jsonrpc":"2.0","method":"tools/list","id":1,"params":{}}` (no API key needed for initialize + tools/list)
- Tool catalog JSON: `GET https://api.mcpanalytics.ai/.well-known/mcp.json`
- Smithery: `smithery mcp add mcpanalytics/mcp-analytics`

## Main Pages

- [Home](https://mcpanalytics.ai/)
- [How It Works](https://mcpanalytics.ai/how-it-works.html)
- [The Engine (MCP)](https://mcpanalytics.ai/the-engine.html)
- [Knowledge Layer](https://mcpanalytics.ai/knowledge-layer.html)
- [Compare vs Julius/Tableau/ChatGPT](https://mcpanalytics.ai/compare.html)
- [For Business](https://mcpanalytics.ai/for-business)
- [For Students](https://mcpanalytics.ai/for-students)
- [For Analysts](https://mcpanalytics.ai/for-analysts)
- [For Researchers](https://mcpanalytics.ai/for-researchers)
- [Free Tools](https://mcpanalytics.ai/free/) — try any analysis without signup
- [Sample Reports](https://mcpanalytics.ai/analysis/)
- [Pricing](https://mcpanalytics.ai/pricing.html)
- [Web Application](https://app.mcpanalytics.ai/)

## Learning Resources

- [Articles](https://mcpanalytics.ai/articles/): 290+ guides to statistical methods
- [Tutorials](https://mcpanalytics.ai/tutorials/): 107+ step-by-step walkthroughs
- [Blog](https://mcpanalytics.ai/blogs/): 70+ case studies
- [Whitepapers](https://mcpanalytics.ai/whitepapers/): 100+ technical deep-dives

## Optional

- [Full reference (llms-full.txt)](https://mcpanalytics.ai/llms-full.txt)
- [Sitemap](https://mcpanalytics.ai/sitemap.xml)
- [Robots.txt](https://mcpanalytics.ai/robots.txt)