Turn CSV and Excel files into interactive BI dashboards

Automated Data Analyst (ADA) - Open-source AI for CSV & Excel

Upload a CSV or Excel file and get automated EDA, an interactive business intelligence dashboard, anomaly detection, forecasts, and evidence-backed recommendations-without configuring a BI tool.

Try the live dashboard · Read the engineering story · See the roadmap · Contribute · Report a bug

If ADA saves you time, star the repository-it helps more analysts and open-source contributors discover the project.

ADA is a privacy-conscious, open-source AI data analyst built with Python, Streamlit, pandas, and Plotly. Upload a CSV, XLSX, or XLSM file and it automatically cleans the data, detects the business schema, creates interactive visualizations, flags anomalous periods, projects a guarded baseline forecast, explains material changes, and recommends what to investigate next.

Ask questions in natural language-effectively chat with your CSV or Excel data-and ADA returns a pandas-backed answer with the calculation behind it. The complete deterministic workflow runs without an API key; optional AI features receive computed schema and evidence, never raw rows or cell values.

See ADA in action

Ask a business question. Get the number and its calculation.

Focus on one segment. Watch the entire analysis regroup itself.

_{Anomalies, forecasts, drivers, and evidence stay inspectable instead of disappearing behind generated prose.}

For the architecture, tradeoffs, and failure modes behind the product, read I Built an AI Data Analyst That Tells You When It Hallucinates.

Why ADA is different

Most CSV analyzers stop at charts. ADA keeps four layers explicit:

Every evidence card and chat answer exposes its calculation. The deterministic product remains authoritative whether or not a model is configured.

How ADA compares

From spreadsheet to decision

flowchart TD
    A["Upload CSV or Excel worksheet"] --> B["Clean and infer types"]
    B --> C["Detect metric, date, segment, ID"]
    C --> D["Calculate evidence, anomalies, and forecast"]
    D --> E["Dashboard, executive brief, and drill-down"]
    C --> G["Ask ADA: question → QueryPlan → local execution"]
    G -. "schema + question only, when rules cannot parse" .-> H["Optional AI query planner"]
    D -. "computed evidence only" .-> F["Optional AI strategic read"]

ADA automatically looks for:

• A primary outcome such as revenue, sales, profit, cost, amount, or units
• A time field for period movement, anomaly detection, and the baseline forecast
• A useful segment such as product, category, channel, region, customer, or status
• Identifiers, missingness, outliers, concentration, and numeric relationships
• The strongest evidence-backed next investigation, separated from observed fact

If the source schema is unusual, users can override the detected metric, date, and segment without rebuilding the dashboard - and drill the whole analysis into a single segment slice.

Product capabilities

• Zero-configuration CSV and Excel analytics with an included synthetic demo
• Ask ADA: plain-English questions (totals, rankings, breakdowns, trends, growth, counts, time and segment filters) answered locally with the calculation shown
• Anomaly radar: periods outside a robust trendline band are flagged on the chart, in the evidence ledger, and in the recommended actions
• Guarded baseline forecast with month-of-year seasonality, an uncertainty band, and its backtested error printed next to the chart
• Drill-down focus: analyze one segment value and automatically regroup by the next useful dimension
• Movement waterfall reconciling the latest change by segment, plus a segment-by-period intensity heatmap
• Worksheet picker for multi-sheet Excel workbooks
• Conservative cleanup, type inference, duplicate removal, and a visible cleaning audit
• Executive headline, four business KPIs, and plain-English briefing
• Evidence ledger with the calculation behind every displayed signal
• Prioritized recommendations linked to deterministic evidence
• Optional AI query planner and structured strategy synthesis using the OpenAI Responses API
• Downloadable Markdown executive brief and cleaned CSV
• Responsive Streamlit interface built for non-technical users
• File limit and row cap for predictable hosted performance

Common use cases

• Automated exploratory data analysis (EDA) for CSV and Excel files
• Natural-language data analysis and chat-with-data workflows
• KPI and business intelligence dashboards for operators
• Time-series anomaly detection and guarded forecasting
• Evidence-backed executive summaries and downloadable reports
• A transparent, self-hostable alternative to black-box CSV analysis tools

Architecture

The codebase favors pure analysis functions and dependency injection at the model boundary. That keeps the business engine testable without Streamlit, network access, or API credits.

Run locally

git clone https://github.com/saineshnakra/automated-data-analyst.git
cd automated-data-analyst
python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate
python -m pip install -r requirements.txt
streamlit run app.py

No secret is required. A visitor can enter their own API key in the session-only sidebar field. A trusted private deployment can instead set OPENAI_API_KEY in the environment or in .streamlit/secrets.toml:

OPENAI_API_KEY = "your-key"

Never commit that file; it is already ignored. Avoid putting an owner-funded key on a public deployment unless you also add authentication and spending controls.

Test and develop

python -m pip install -r requirements-dev.txt
ruff check .
python -m unittest discover -s tests -v

GitHub Actions runs linting, the complete test suite, and bytecode compilation on every push and pull request.

FAQ

Does my data leave my machine?
No. Cleaning, schema detection, every chart, every evidence card, and every Ask ADA answer are computed locally with pandas. If you opt into the AI layer, only column schema and computed evidence are sent - never rows or cell values.

Do I need an OpenAI API key?
No. ADA is a complete analyst without one. A key only adds the query-planner fallback for unusual questions and the strategic narrative.

What formats can I analyze?
CSV (comma, semicolon, or tab delimited), XLSX, and XLSM - including picking a specific worksheet from a multi-sheet workbook.

How is this different from pasting a CSV into a chatbot?
A chatbot gives you fluent prose you cannot audit and your rows become part of a prompt. ADA turns questions into explicit query plans, executes them with pandas on your machine, and prints the calculation under every answer.

Is ADA a PandasAI alternative?
ADA serves a related chat-with-data use case but takes a different approach: its core uses deterministic pandas calculations rather than model-generated code, and every answer exposes its calculation. ADA is a complete Streamlit application, not a drop-in replacement for the PandasAI library.

Can I self-host it?
Yes - it is a standard Streamlit app. pip install -r requirements.txt && streamlit run app.py, or deploy it to any host that runs Python.

Contribute

Contributions are welcome, especially around new deterministic metrics, question shapes for Ask ADA, schema-detection fixtures, chart accessibility, file formats, and adversarial test datasets. Start with the good first issues, read CONTRIBUTING.md, choose an item from the roadmap, or open a focused proposal.

Good contributions make an insight more accurate, more explainable, or easier for a non-technical user to act on. Every new recommendation should include a test and the calculation that supports it.

If ADA is useful to you, a star helps other operators find it.

Deploy

Deploy app.py on Streamlit. The repository includes its app theme, dependency manifest, server upload limit, and headless configuration. Add OPENAI_API_KEY through Streamlit's secret manager only if the optional strategy layer should be available.