Getting Started with Data Analysis

Blast Office provides a chat-first data analysis interface that combines natural language conversation with the power of SQL and DuckDB.

The Chat-First Interface

Data analysis in Blast Office works through conversation. Describe what you want to analyze, and the AI will help you load data, write queries, and visualize results.

Loading Data

To start analyzing data:

1. Open a new Data Analysis session from the sidebar
2. Load your data by dragging files into the interface or using the file picker
3. Supported formats include CSV, TSV, Parquet, JSON, and Excel files

Writing and Executing SQL

You can write SQL queries directly or let the AI help:

Natural Language Queries

Ask questions in plain English:

• “What are the top 10 customers by revenue?”
• “Show me the monthly sales trend for 2024”
• “Which products have the highest return rate?”

The AI translates your questions into SQL and executes them.

Direct SQL

Write SQL queries directly in the input field. Blast Office uses DuckDB as the query engine, which supports:

• Standard SQL syntax
• Window functions
• CTEs (Common Table Expressions)
• JSON processing
• Parquet file support

Visualization

Results are displayed in tables by default. The AI can also create visualizations:

• Bar charts and line graphs
• Histograms and scatter plots
• Pie charts
• Time series plots

Ask the AI to “visualize this data” or “create a chart showing the trend” and it will generate an appropriate visualization.

Tips

• Start with broad questions and drill down into specifics
• Use the AI to explore unfamiliar datasets — ask “describe this data” or “what columns are available?”
• Save interesting queries for later reference