How it works
From PDF to Excel, in 3 steps
What happens between upload and export, and why each stage is designed the way it is.
- 1Upload
- 2Identify
- 3Extract & validate
- 4Export & trace
- 1
Upload
Drop one or many PDFs. Files are encrypted in transit (TLS 1.3) and hashed (SHA-256) on receipt. The hash and timestamp anchor the audit chain.
- Drag-and-drop or batch upload (multi-file)
- Up to 100 MB per file, 200 pages per PDF (cloud)
- Both text-based and scanned PDFs accepted
- 2
Identify
ChromaParse fingerprints the PDF against its template library (Waters Empower, Agilent ChemStation/OpenLab, Thermo Chromeleon, Shimadzu LabSolutions...) and selects the matching parser.
- Automatic vendor + workstation + version detection
- Falls through to OCR layout analysis for non-text PDFs
- Logs the chosen template into the audit trail
- 3
Extract & validate
Per-cell extraction with confidence scoring. Numerical fidelity is exact — no rounding, no scientific-notation surprises.
- Retention time, peak area, height, %area, peak names
- System suitability tables, sample / method headers
- Low-confidence cells flagged for review with severity
- 4
Export & trace
One-click Excel/CSV export. Every cell carries source-trace metadata — click it in Excel and ChromaParse highlights the exact bbox in the source PDF.
- Excel / CSV / JSON export
- Source-trace metadata embedded per cell
- Direct LIMS field mapping (config-driven)