Ask ChatGPT or Claude to analyze your data and the answer changes every time you run it. We build the real analysis behind the chart — validated statistics you own, can re-run, and can put in front of anyone — right inside your AI chat.
Custom analysis, built for your CSV.
Upload your CSV, describe your question, and we build a custom statistical analysis module specific to your business — revenue trends, churn signals, ad performance, forecasting. Yours to own. Rerun next month on fresh data without rebuilding.
Real R, not vibes-based AI.
Validated statistical methods — t-tests, ANOVA, regression, survival analysis, PCA — packaged into a custom module for your data. Every result includes diagnostics, confidence intervals, and a citable PDF. Rerun the exact same analysis next semester on next year’s data.
Your AI already knows us.
Connect from Claude, ChatGPT, Cursor, Windsurf, or your own agent via MCP protocol. Full toolset for module discovery, execution, and customization. Bring your own API key or use platform credits. The analysis engine works regardless of which model you use.
Every analysis runs through the same validated pipeline — you choose how far it goes. From a fast verified read of your data, to a single computed answer you can re-run, to a full commissioned study you own and refresh forever. Every depth comes back independently verified.
More rigor outranks more charts: going deeper buys real statistical methods — hypothesis tests, regression, diagnostics — not just more cards. You pay for depth, not chart count — and only if the build succeeds.
The same pipeline runs at every depth — a Snapshot runs a lean version in minutes, a Deck runs the full study. Depth changes how much of it runs and how robust the result, never the rigor. Here’s what happens between your question and an analysis you own.
Every analysis produces a multi-card interactive report with AI insights and a downloadable PDF. Not a chat response — a complete deliverable.
Not in a chat thread that disappears. Not in a notebook you can't find. Every analysis is permanent, shareable, citable, and reproducible.
Every report stored, indexed, searchable. Compare results across months. Nothing disappears.
One-click APA, MLA, Chicago, or BibTeX. Methodology and assumptions documented per card. Use in papers, decks, filings.
Token-based access, no login to view. Share with your team, your client, your professor. The methodology travels with the report.
Fixed seeds, Docker isolation, R source code in every report. Same input, same answer, every run. Run it independently — get the same result.
Same reproducible product whether you’re a solo analyst or an enterprise team.
Validated R analyses across 8 categories — each produces a full interactive report.
t-test · ANOVA · Chi-square · Linear · Ridge · Lasso · Logistic · Random Forest · XGBoost · K-Means · PCA · ARIMA · RFM · Churn · ROAS · ICC · ANCOVA · Cox PH · and more
Browse all modules →Upload your CSV. Get a real statistical report with interactive charts and AI insights.
Describe what you want to know. The agent picks the right analysis and delivers a full report.
Each card in your report gets its own AI interpretation. Not generic summaries — specific observations about your data, statistical significance calls, and actionable recommendations.
AI reads each chart and table, explains what matters and what to act on.
One-page TLDR for stakeholders who won't read the full report.
Citable R code, assumption checks, what the test actually proves.
TV spend shows the strongest ROI at $4.20 per dollar, significantly outperforming Radio ($2.15) and Newspaper ($0.87). The model explains 89.7% of variance (R² = 0.897), suggesting marketing budget reallocation from Newspaper to TV could increase revenue by approximately 12-18%.
Every result gets embedded into a high-dimensional vector space. Related insights cluster together automatically.
Search by meaning, not keywords. The more you analyze, the more connections you find.
Ask a question, pick a tool, or let the agent decide. The result is a full report with charts, tables, and AI insights.
The result is converted to a 768-dimension vector and placed in a shared semantic space alongside every other analysis.
Ask “What do we know about churn?” and the system retrieves the most semantically relevant results — across all datasets, tools, and time.
Each dot is a past analysis. Proximity = similarity in meaning.
A query finds the nearest neighbors — regardless of when they ran or what tool was used.
The more you analyze, the more connections you find.
Learn more about the knowledge layer →Whether you're running homework stats or forecasting next quarter's revenue.
MCP Analytics is an MCP server. Install it in any compatible AI client and your agent gets access to the full module library — run statistical analyses, ML models, and business analytics directly from conversation.
One developer installs the MCP server. The whole team views reports in the web app. Two interfaces, one platform.
# Add to your MCP client config
{
"mcpServers": {
"mcpanalytics": {
"command": "npx",
"args": [
"-y",
"@mcp-analytics/mcp-analytics",
"--api-key",
"YOUR_API_KEY"
]
}
}
}
Also supports direct HTTP and OAuth2
Every report has feedback buttons on each slide. Flag an issue and we'll send you an updated version. Still not satisfied? Credits back, no questions asked.
Every analysis you run becomes searchable knowledge. The more you use it, the smarter your organization gets.