User Guide

Managing Documents

Learn how to upload and manage your legal documents

Uploading Documents

Thea supports uploading various document formats to extract chronological information and build timelines.

Supported Formats

  • PDF (.pdf) - Including scanned documents with OCR text
  • Microsoft Word (.docx) - Modern Word documents
  • Plain Text (.txt) - Simple text files

How to Upload

  1. Navigate to your project
  2. Click the Documents tab
  3. Click Upload Documents or drag and drop files into the upload area
  4. Select one or more documents from your computer
  5. Wait for the upload to complete

You can upload multiple documents at once. Drag-and-drop works for batch uploads.

Document Processing

After uploading, Thea automatically processes each document through several stages:

1. Text Extraction

  • Extracts all readable text from the document
  • Preserves structure (paragraphs, headings) where possible
  • Handles multi-column layouts in PDFs

2. Intelligent Analysis

  • Identifies chronological content (dates, events, sequences)
  • Detects mentioned parties and organizations
  • Recognizes legal terminology and document types
  • Creates searchable representations (embeddings) for smart retrieval

3. Availability

Once processing completes, the document content becomes available for:

  • Timeline suggestions: Thea can propose timeline structures based on document patterns
  • Event extraction: Pull chronological events directly into timelines
  • Conversational creation: Reference document content when building timelines through chat
  • Source citations: Link events back to source documents for verification

Processing time: Typically 30 seconds to 2 minutes depending on document length and complexity.

Viewing Your Documents

The document list shows all files in your project:

  • Filename: Original document name
  • Upload Date: When added to the project
  • Status Indicator:
    • Processing: Document is being analyzed
    • Completed: Ready for use
    • Failed: Processing encountered an error
  • Actions: Download, view, or delete

Click on a document name to preview its content (if supported).

Document Organization

Projects as Containers

Documents are organized within projects (folders):

  • Each project contains its own set of documents
  • Timeline suggestions are specific to that project's documents
  • Keeps different cases separate

Best Practices for Organization

  • One project per case: Keep documents for each case in separate projects
  • Descriptive filenames: Name files clearly before uploading (e.g., "Complaint_2024-03-15.pdf")
  • Upload in batches: Add all related documents at once so Thea can see the full context
  • Check for duplicates: Avoid uploading the same document multiple times

Document Features

Viewing and Downloading

  • Click a document name to open a preview (PDF viewer or text preview)
  • Use the Download button to save a copy locally
  • Preview shows the processed text for verification

Deleting Documents

To remove a document from your project:

  1. Click the delete icon next to the document
  2. Confirm the deletion

Warning: Deleting a document:

  • Removes it from the project permanently
  • Removes any timeline suggestions based on that document
  • Does NOT delete events already created from that document (events remain but lose source citations)

Source Citations

Documents remain linked to extracted events:

  • Events show which document they came from
  • Click Show Source in an event to see the original text
  • Useful for verification and court citations

Troubleshooting

Document Won't Upload

Possible causes:

  • File format not supported (only PDF, DOCX, TXT)
  • File is corrupted or damaged
  • File size is too large (typically 50MB limit)
  • Network connection interrupted

Solutions:

  • Verify file format and try converting if needed
  • Try uploading a smaller test file first
  • Check your internet connection

Processing Failed

Common reasons:

  • Password-protected or encrypted PDF: Remove password protection before uploading
  • Scanned image-only PDF without OCR text: Requires OCR preprocessing
  • Corrupted file: Try opening the file locally to verify it's not damaged
  • Unsupported character encoding: Some legacy text files may have encoding issues

What to do:

  • Check the file can be opened normally on your computer
  • For image PDFs, use OCR software to create a searchable PDF first
  • Try re-saving the document in a standard format
  • Contact support if issues persist

Processing is Taking Too Long

  • Documents over 100 pages may take 3-5 minutes
  • Complex PDFs with many images process slower
  • Check the status—if it's stuck for more than 10 minutes, try deleting and re-uploading
  • Page refresh does not interrupt processing

Privacy and Security

  • Documents are securely stored and encrypted
  • Only you can access documents in your projects
  • Text extraction happens securely in the cloud
  • Documents can be permanently deleted at any time from project settings

Next Steps