Skip to content
Documentation

Working with Data Sources

Querri helps you analyze data from various sources. This guide covers how to upload files, connect to data sources, and work with the Library to manage your data.

Uploading Files

Section titled “Uploading Files”

The simplest way to get data into Querri is by uploading files.

Supported File Formats

Section titled “Supported File Formats”

Querri supports common data formats:

  • CSV (.csv) - Comma-separated values
  • Excel (.xlsx, .xls) - Microsoft Excel spreadsheets
  • JSON (.json) - JavaScript Object Notation
  • TSV (.tsv) - Tab-separated values
  • Other delimited text formats

How to Upload

Section titled “How to Upload”

You can upload files in several ways:

From the Library

Section titled “From the Library”
  1. Navigate to your Library
  2. Click “Upload File” or similar button
  3. Select your file from your computer
  4. Wait for the upload to complete

The file is now stored in your Library and can be used in any project.

During a Chat Session

Section titled “During a Chat Session”

When you start a new project or ask to analyze data:

  1. The AI may prompt you to add a data source
  2. Click the upload option
  3. Select your file
  4. Continue your conversation

The uploaded data becomes available immediately for analysis.

Drag and Drop

Section titled “Drag and Drop”

Many interfaces support drag-and-drop:

  1. Drag your file from your computer
  2. Drop it into the designated area
  3. The file uploads automatically

File Size Considerations

Section titled “File Size Considerations”
  • Most files under 100MB upload quickly
  • Larger files may take longer but are supported
  • Very large datasets (1GB+) may require special handling

If you’re working with extremely large data, consider filtering or sampling before upload.

Referencing Data in Chat

Section titled “Referencing Data in Chat”

Once data is uploaded, you reference it in your questions:

By File Name

Section titled “By File Name”

“Analyze sales_data.csv” “Show me trends in customer_orders.xlsx”

Implicitly

Section titled “Implicitly”

If you’ve already mentioned a dataset: “Filter to 2024” (AI knows which dataset you mean) “Now show me monthly totals” (continues with same data)

Multiple Sources

Section titled “Multiple Sources”

“Join sales_data.csv with customer_info.xlsx on customer_id” “Compare revenue trends between region_north.csv and region_south.csv”

The AI understands context and can work with multiple data sources in a single analysis.

Using the Library

Section titled “Using the Library”

The Library is your central data repository in Querri.

What’s in the Library?

Section titled “What’s in the Library?”

The Library stores:

  • Uploaded files: CSV, Excel, JSON, and other data files
  • Connected sources: Links to databases and external systems
  • Metadata: Information about each data source

Benefits of the Library

Section titled “Benefits of the Library”

Reusability: Upload once, use in many projects Organization: Keep all your data in one place Efficiency: No need to re-upload the same file for different analyses

Accessing the Library

Section titled “Accessing the Library”

Navigate to the Library section (typically in the main menu). Here you can:

  • View all available data sources
  • Search for specific files
  • Upload new data
  • Delete data you no longer need
  • View details about each source

Adding Data to Projects

Section titled “Adding Data to Projects”

When starting a new project:

  1. Begin your chat
  2. The AI prompts for data or you mention what you want to analyze
  3. Select from Library or upload new
  4. Start analyzing

You can add more data sources as your analysis progresses.

Adding Sources During Conversation

Section titled “Adding Sources During Conversation”

You don’t need to upload everything upfront:

You: “Analyze customer purchase patterns” AI: Loads customer data from Library

You: “Compare that with our marketing campaign data” AI: “I’ll need the marketing data. Would you like to upload it or select from the Library?”

You: Uploads marketing_campaigns.csv

AI: Continues analysis with both datasets

This flexible approach lets you build your analysis iteratively.

Working with Multiple Data Sources

Section titled “Working with Multiple Data Sources”

Querri excels at combining data from multiple sources:

Joining Data

Section titled “Joining Data”

“Join orders.csv with customers.csv on customer_id” “Merge product data with sales data”

Comparing Datasets

Section titled “Comparing Datasets”

“Compare Q1 sales to Q2 sales” “Show differences between actual_revenue.csv and forecasted_revenue.csv”

Appending Data

Section titled “Appending Data”

“Combine all regional sales files into one dataset” “Append this month’s data to the year-to-date file”

The AI handles the technical details of combining data sources.

Connected Data Sources

Section titled “Connected Data Sources”

Beyond file uploads, Querri can connect to live data sources:

Database Connections

Section titled “Database Connections”
  • PostgreSQL
  • MySQL
  • Other SQL databases

Cloud Services

Section titled “Cloud Services”
  • Data warehouses
  • Cloud storage
  • API-based services

Setting Up Connections

Section titled “Setting Up Connections”

Connecting to external sources typically involves:

  1. Navigate to Library
  2. Select “Add Connection” or similar
  3. Provide connection details (host, credentials, etc.)
  4. Test the connection
  5. Save to Library

Once connected, query the source just like an uploaded file: “Analyze data from production_db” “Show me the latest records from customer_warehouse”

Data Privacy and Security

Section titled “Data Privacy and Security”

Where is Data Stored?

Section titled “Where is Data Stored?”
  • Uploaded files are stored securely within Querri
  • Connected sources remain in their original location
  • Querri queries data but doesn’t necessarily copy entire databases

Access Control

Section titled “Access Control”
  • Your Library data is private to you
  • Share access when you share projects
  • Remove access by unsharing or deleting data

Deleting Data

Section titled “Deleting Data”

Remove data from the Library when you no longer need it:

  1. Navigate to Library
  2. Select the data source
  3. Click Delete or similar action
  4. Confirm deletion

Projects using deleted data may not function properly, so delete carefully.

Best Practices

Section titled “Best Practices”

Organize Your Library

Section titled “Organize Your Library”
  • Use clear, descriptive file names: “sales_2024_q1.csv” not “data.csv”
  • Delete old or unused data periodically
  • Keep related datasets together (naming conventions help)

Prepare Clean Data

Section titled “Prepare Clean Data”

While Querri can clean data, uploading clean data saves time:

  • Remove extraneous rows (headers, footers, notes)
  • Use consistent column names
  • Ensure dates are in recognizable formats
  • Avoid special characters in column names when possible

Start Small, Scale Up

Section titled “Start Small, Scale Up”

When working with large datasets:

  1. Test with a sample first
  2. Verify your analysis approach
  3. Then run on the full dataset

This prevents wasting time if your approach needs adjustment.

Document Your Sources

Section titled “Document Your Sources”

Consider keeping notes about where data came from:

  • Original source
  • Date obtained
  • Any preprocessing done
  • Refresh frequency for connected sources

This context helps when revisiting projects later.

Troubleshooting

Section titled “Troubleshooting”

Upload Failed

Section titled “Upload Failed”

If an upload fails:

  • Check file format is supported
  • Verify file isn’t corrupted
  • Try a smaller sample of the data
  • Check internet connection

Can’t Find Data in Library

Section titled “Can’t Find Data in Library”
  • Use the search function
  • Check if it was deleted
  • Verify you’re logged in to the correct account

Data Looks Wrong

Section titled “Data Looks Wrong”

If data appears incorrect after upload:

  • Check the original file
  • Verify encoding (UTF-8 is standard)
  • Look for parsing issues (wrong delimiter, etc.)
  • Try re-uploading

Example Workflow

Section titled “Example Workflow”

Here’s a typical workflow for working with data sources:

  1. Upload file to Library: sales_2024.csv
  2. Start new project: “Analyze sales trends”
  3. AI loads data: From Library
  4. You continue analysis: “Break down by region”
  5. Add more data: Upload regions.csv for regional details
  6. Combine sources: “Join sales with region data”
  7. Complete analysis: Create visualizations and insights
  8. Clean up: Remove temporary files from Library if needed

Next Steps

Section titled “Next Steps”