Parser Information
For file-based documents (PDFs, uploaded files, storage-based sources), the Document Statistics page displays information about how the document was parsed.
Parser Information Card
| Field | Description |
|---|---|
| Parser Used | The parsing method used to extract text from the document |
| Quality Score | A confidence score (0-100%) indicating extraction quality |
| Quality Issues | Any detected issues with the extraction (if applicable) |
Parser Types
| Parser | Description | When Used |
|---|---|---|
| Standard | Fast, free text extraction for digital PDFs | Digital PDFs with selectable text |
| Advanced | High-accuracy OCR for complex documents (Growth+ plans) | Scanned PDFs, images, complex layouts |
Quality Score Interpretation
| Score Range | Meaning | Recommendation |
|---|---|---|
| 80-100% | Excellent extraction quality | No action needed |
| 60-79% | Good extraction with minor issues | Review if accuracy is critical |
| Below 60% | Poor extraction quality | Consider re-syncing with Advanced parser (Growth+ plans) |
When Parser Info is Not Available
Parser information may not be available for:
- Documents synced before this feature was added
- API-based sources (Confluence, Notion) that don't require file parsing
- Website sources that use HTML extraction
If parser information is missing for a file-based document, you can click "Re-sync Document" to reprocess it and capture parser metadata.