Batch Processing
For cases with many documents, batch processing lets you analyze and redact multiple files efficiently.
When to Use Batch Processing
Batch processing is ideal when:
- A case has 5+ documents requiring redaction
- You need to process similar document types
- Time is limited and automation helps
- You want consistent redaction across documents
Starting Batch Processing
Attach Multiple Documents
Batch processing isn’t a separate area — it happens automatically whenever you attach more than one file. There is no “Start Batch Analysis” button.
- Open the case and start a case action that attaches documents (Request Info, Submit Info on an Awaiting Information case, or Close Case)
- In the file picker, select several files at once (hold Ctrl on Windows or Cmd on Mac)
- The redactable files are scanned automatically and you are taken to the Review Detected PII screen, with one expandable panel per file
Document Selection Tips
- Attach documents of similar types together for more predictable results
- Review document quality before attaching
- Large batches may take several minutes
Batch Analysis
How It Works
The system processes documents in parallel:
- Multiple documents analyzed simultaneously
- Progress shown for each document
- Individual results available as they complete
- Overall batch progress indicator
Monitoring Progress
The progress view shows:
- Documents queued
- Documents in progress
- Documents completed
- Any errors encountered
Timeout Handling
Documents have a processing timeout:
- Complex documents may take longer
- Very large files may timeout
- Retry individual documents if needed
Reviewing Batch Results
Summary View
After batch completion:
- Total documents processed
- Total PII entities detected
- Documents requiring attention
- Quick statistics
Per-Document Review
On the Review Detected PII screen, each file has its own expandable panel. For each file you can:
- Expand the panel to see the detected entities, their type, and confidence
- Toggle individual entities on or off for redaction
- Note that the data subject’s own PII is shown but not selected for you
Bulk Selection Controls
Across the whole batch you can use:
- Select All - select every detected entity for redaction
- Deselect Subject Data - clear entities identified as the data subject’s own information
- Smart Scan - apply a recommended selection
- Skip Redaction - attach the files unchanged, without redaction
Batch Redaction
Redact Selected
- Make your selection using the per-entity toggles and the bulk controls above
- Click Redact Selected (n) to process all selected files
- Wait for processing to complete (up to 3 files are processed in parallel)
Review and Confirm
After processing, a review screen lets you check each redacted file before it is attached:
- View - open a file to verify what was removed
- Download - save a copy of the redacted output
- Remove - drop a file from the batch
When you’re satisfied, click Confirm & Attach to Case (n) to add the redacted files to the case. You can also choose Pause & Resume Later to continue the review another time.
Batch Processing Sessions
Session Tracking
Each batch operation creates a session:
- Session ID for reference
- Start/end timestamps
- Documents included
- Processing status
- Results summary
Resuming Interrupted Sessions
If batch processing is interrupted:
- Session state is preserved
- Incomplete documents can be retried
- Completed documents are saved
- Resume from where you left off
Performance Considerations
Document Limits
| Aspect | Limit |
|---|---|
| Documents per batch | 50 |
| Total batch size | 1 GB |
| Max document size | 100 MB (10 MB on trial) |
| Documents analysed in parallel | Up to 3 at once |
| Timeout per document | 5 minutes |
Processing Time Estimates
| Document Type | Typical Time |
|---|---|
| Text PDF (1-10 pages) | 10-30 seconds |
| Scanned PDF (requires OCR) | 30-60 seconds |
| Word document (.docx) | 15-30 seconds |
| Excel spreadsheet (.xlsx) | 20-45 seconds |
| Image file (.png, .jpg, etc.) | 20-40 seconds |
| Email file (.eml, .msg) | 10-25 seconds |
| Text file (.txt, .csv, .json, etc.) | 5-15 seconds |
Optimizing Batch Performance
- Group similar documents - Similar types process more predictably
- Smaller batches for urgent work - Start with priority documents
- Check during processing - Review completed docs while others process
- Retry failures separately - Don’t re-run entire batch for one failure
Best Practices
Before Starting
- Review documents to ensure they need processing
- Remove duplicates from the batch
- Ensure documents meet quality requirements
During Processing
- Don’t close the browser during batch operations
- Monitor for errors
- Review completed documents as they finish
After Completion
- Review all detections before applying redactions
- Check edge cases manually
- Verify redacted outputs before sending
Quality Control
- Spot-check a sample of redacted documents
- Verify critical documents individually
- Document any manual adjustments
Error Handling
Common Errors
“Document extraction failed”
- File may be corrupted
- Password protection present
- Unsupported format variant
“Analysis timeout”
- Document too complex
- Try single document processing
- Check document size
“Redaction failed”
- Coordinate mapping issue
- Try re-analyzing the document
- Manual redaction as fallback
Retrying Failed Documents
- View error details
- Address the issue if possible
- Remove from batch or retry
- Process manually if needed
Audit Trail
All batch operations are logged:
- Batch session details
- Documents processed
- Detections made
- Redactions applied
- User who performed the operation
- Timestamps for all actions