Use case
BOL, CMR and customs forms across four languages
Split multi-doc scans, classify each form, extract line items — RO/TR/DE/BG included.
The problem
A shipment's paperwork is a stapled multilingual mess, and the dock doesn't wait for data entry.
- 01Four languages, one pack. Romanian, Turkish, German and Bulgarian forms mixed together defeat template parsers.
- 02Stapled scans. Several shipments in one PDF. Someone has to separate them before anything books.
- 03Unsorted forms. BOL, CMR, invoice, packing list — each needs its own schema, and nobody sorts upfront.
- 04Dock-side delays. Trucks wait while paperwork gets retyped. Every hour is detention cost.
The solution
Fieldry splits, classifies and extracts the pack — tables stitched, totals checked, evidence attached.
- 01Split first. Multi-shipment scans separate into individual documents automatically.
- 02Classify each page. BOL, CMR, invoice or customs form — routed to the right schema without sorting.
- 03Multilingual by default. Interpreter model reads all four languages plus Cyrillic — no dictionaries.
- 04Tables and totals. Packages, weights and HS codes as rows; totals reconciled with evidence.
Extraction checks
Search the freight checks.
See what each check returns.
Every check Fieldry runs on a shipment pack.
Split first
Multi-shipment scans separate into individual documents automatically.
Classify each page
BOL, CMR, invoice or customs form — routed to the right schema without sorting.
Multilingual by default
Interpreter model reads all four languages plus Cyrillic — no dictionaries.
Command preview
Auto-separation
Multi-shipment PDFs separate into individual documents before extraction.
Anatomy of a shipment pack
A driver hands over a 25-page scan: BOLs, CMRs and invoices in three languages, two shipments stapled together, one page sideways.
Fieldry splits the shipments, classifies every page, extracts against each schema — and the booking proceeds from structured data.
- Input25-page mixed-language scan
- OutputPer-shipment structured records
- ReviewExceptions only
From dock scan to booked shipment
Ops books from structured records; disputes resolve at the source line.
AnalystLine items, weights and totals extracted with tables stitched and reconciled.
FieldryShipments split; pages classified to BOL, CMR, invoice and customs schemas.
FieldryMixed pack arrives — stapled shipments, mixed languages, sideways pages.
Document
4 milestones, newest first.
Key benefits
From dock scan to booked shipment without retyping.
No templates
New carrier, new layout — parses anyway against your schemas.
Detention killer
Sync extraction clears paperwork while the truck is still at the dock.
Dispute evidence
Every charge traces to page and bbox; rate fights end fast.
TMS integration
REST, webhooks and n8n push structured records into your systems.
Parse your ugliest pack
Upload a real multi-doc scan on free credits and compare against your booking.
Get startedRelated use cases
Related industries
Bills of lading, CMR notes, commercial invoices, packing lists and customs forms — across Latin and Cyrillic scripts.
Split separates them first; each document is classified and extracted on its own with per-result metering.
OCR handles rotation, skew and warehouse photo quality; low-confidence fields route to review with the snippet attached.
Yes — structured JSON via API or webhook, plus an n8n node for no-code routing into your stack.
Parse your first document today
One API for extraction, evidence and automation.