Extract data from waybills
A waybill is the transport contract that rides with the goods on the road and the rail. For a truck crossing from Rotterdam to Munich the governing document is the CMR consignment note, standardized by the 1956 Geneva Convention on the Contract for the International Carriage of Goods by Road; for a rail wagon it is the CIM note under COTIF. Unlike an ocean Bill of Lading, and unlike an air waybill issued under IATA rules, a road or rail waybill is not a document of title: the named consignee takes the goods on arrival, and the paper exists to prove the carrier accepted the consignment, to record its condition, and to fix who pays the freight. A haulier such as DB Schenker or Girteka issues one per consignment, the driver signs it at collection, and the consignee signs it at delivery. The CMR note is a multilingual, multi-copy form and its important marks are handwritten. The consignor keeps copy 1, the consignee gets copy 2, and the carrier retains copy 3, so the scan that reaches an office is often the smudged carrier copy with a driver reservation scrawled in box 18. Field labels appear in German, French, and Spanish on the same pre-printed form (Absender, Expediteur, Empfanger), the vehicle registration and any trailer plate sit in their own boxes, and the goods description carries the UN/ECE Recommendation 21 package type, the gross mass, and an HS commodity code. Freight charges fall to the sender or the consignee, and a groupage note lists several cargo lines under one contract. Feed the waybill to Talonic and the consignor, consignee, notify party, carrier, and its vehicle plate come back as structured fields, alongside the places of receipt and delivery, the pickup and delivery dates, and a cargo array with each line package count, gross mass, volume, and commodity code. Box labels are read whether the form is printed in German or French, the Incoterm is normalized, and freight charges are typed as numbers in a single currency. A CMR note issued 2026-04-21 for a groupage load from Rotterdam to Munich, 8,400 kg across two lines carried by DB Schenker and delivered 2026-04-23, loads into a transport management system with its consignment structure intact rather than re-keyed from a photographed page.
What gets extracted from waybills
How extraction works for waybills
Road and rail waybills reach an office as printed carrier copies, forwarder TMS exports, and photographed CMR notes signed at the loading bay, and the pre-printed CMR and CIM forms label their boxes in several languages at once. Talonic classifies the document and maps it to the transport model in the Field Registry, which holds the consignor, consignee, notify party, carrier, and vehicle registration alongside the cargo lines rather than one description block. Multilingual box labels such as Absender, Expediteur, and Empfanger are resolved to canonical fields, the package type follows UN/ECE Recommendation 21, gross mass is normalized to a value plus an explicit unit, and the Incoterm is read as stated. A groupage note with several cargo lines under one contract is kept as an array. Each field returns with a confidence score and a pixel-region pointer under DIN SPEC 91491, so a transport clerk can check a gross mass or a delivery reservation against the source note before the carrier is paid.
Sample extraction
A CMR consignment note for a two-line groupage road load
{
"document_number": "CMR-2026-04-5521",
"document_date": "2026-04-21",
"shipper.name": "Rotterdam Freight Kontor B.V.",
"consignee.name": "Baumann Fertigung GmbH",
"consignee.address": "Werk 2, Dachauer Str. 220, 80992 Munich, Germany",
"carrier.name": "DB Schenker",
"carrier.license_plate": "NL-72-BXD / trailer OP-441",
"place_of_receipt": "Rotterdam",
"place_of_delivery": "Munich",
"pickup_date": "2026-04-21",
"delivery_date": "2026-04-23",
"incoterms": "DAP",
"currency": "EUR",
"total_weight": 8400,
"freight_and_charges": 1840,
"line_items": [
{
"cargo_description": "Aluminium housings",
"number_of_packages": "12 pallets",
"package_type": "PLT",
"gross_mass": 5200,
"commodity_code": "7616.99"
},
{
"cargo_description": "Steel fasteners",
"number_of_packages": "8 pallets",
"package_type": "PLT",
"gross_mass": 3200,
"commodity_code": "7318.15"
}
]
}Frequently asked
How is a road waybill different from a Bill of Lading or an air waybill?
A road or rail waybill, such as a CMR or CIM note, is not a document of title: the named consignee takes the goods on arrival. An ocean Bill of Lading can be negotiable, and an air waybill follows IATA rules. Talonic captures each on its own schema.
Does it read the multilingual CMR form?
Yes. The pre-printed CMR note labels its boxes in German, French, and Spanish at once, and Talonic resolves those labels to canonical fields so the consignor and consignee land in the right place.
Are the vehicle and trailer plates captured?
Yes. The tractor registration and any trailer plate are captured in their own field, because a haulage operation reconciles a delivery against the vehicle that carried it.
What about a groupage note with several consignments?
Each cargo line returns as a row in an array with its own package count, gross mass, and commodity code, so a groupage load under one contract stays itemized.
Ready to extract from your own waybills?
Author note
Reviewed by Talonic engineering · last reviewed 2026-07-07