Extract data from customs declarations
A customs declaration is the document that clears goods across a border, and it is where a single misread number turns into a held shipment or a penalty. A customs broker filing entry for an importer, or a trade-compliance team auditing what was filed, works from declarations such as the US CBP Form 7501 entry summary or an equivalent single administrative document, each tying a shipment to a tariff classification, a declared value, and the duties owed. A broker handling entries all day processes declarations covering one HS code or forty, for shipments arriving from Shenzhen, Rotterdam, or Monterrey, each with an importer of record, a country of origin, and an Incoterms basis that determines whether freight and insurance are inside the customs value.
The difficulty is the line-item grid and the codes. Each line carries a Harmonized System tariff code, the ten-digit HTS classification such as 8479.89.9499, a description, a country of origin, a quantity, a customs value, and a duty rate that produces the duty amount. The classification drives everything, and an importer that splits one product across two HS codes doubles the lines. Customs value is built from the goods value plus or minus freight and insurance depending on the Incoterms, so a CIF basis and an FOB basis compute differently. Entry numbers, the importer of record number, and the manifest reference connect the declaration back to the Bill of Lading and the commercial invoice. Duty and fee lines, including the merchandise processing fee, sum to the total payable.
Talonic reads the declaration and returns the entry header plus a line array, each line with its HS code, origin, quantity, customs value, duty rate, and duty amount, so a broker or compliance team can reconcile the filing against the invoice and the Bill of Lading from structured data.
What gets extracted from customs declarations
How extraction works for customs declarations
Customs declarations are produced by broker entry software and national customs systems, so a US CBP Form 7501 and a European single administrative document carry the same data in different layouts. Talonic classifies the declaration and maps it to the customs-entry schema in the Field Registry, which separates the entry header from the per-line tariff detail. Each line keeps its Harmonized System code, the ten-digit HTS classification, country of origin, quantity, customs value, and duty rate, and the duty amount is recomputed from value and rate to catch a transcription error. The customs value basis is read from the Incoterms so a CIF and an FOB filing are handled correctly. Entry, importer-of-record, and manifest references are captured so the declaration reconciles against the Bill of Lading and the commercial invoice. Every value returns with a confidence score and pixel-region provenance under DIN SPEC 91491 conformity, so a broker can verify an HS code or value against the source filing.
Sample extraction
A US CBP Form 7501 entry summary with two tariff lines
{
"entry_number": "ENT-2026-774512",
"importer_of_record": "Acme Imports LLC",
"importer_number": "12-3456789",
"country_of_origin": "China",
"incoterms": "CIF",
"currency": "USD",
"lines": [
{
"hts_code": "8479.89.9499",
"description": "Industrial machine parts",
"origin": "CN",
"quantity": 288,
"customs_value": 48200,
"duty_rate": "2.5%",
"duty_amount": 1205
},
{
"hts_code": "8544.42.9090",
"description": "USB-C cable assembly",
"origin": "CN",
"quantity": 600,
"customs_value": 9000,
"duty_rate": "2.6%",
"duty_amount": 234
}
],
"merchandise_processing_fee": 99.4,
"total_duties_and_fees": 1538.4
}Frequently asked
What is an importer of record?
The importer of record is the party in whose name goods are entered, responsible for the duties owed, the tariff classification declared, and the records behind the filing. On a US CBP Form 7501 entry summary it sits in the header with its importer number, such as Acme Imports LLC under 12-3456789. Both are returned as header fields alongside the entry number and the manifest reference, so the declaration reconciles against the Bill of Lading and the commercial invoice.
Does it capture each tariff line separately?
Yes. Every line keeps its own Harmonized System code, country of origin, quantity, customs value, and duty rate, so a shipment split across two HS classifications is filed and audited as two lines rather than a merged total.
How is the duty amount checked?
The duty amount is recomputed from the customs value and the stated duty rate per line, so a transcription error between the rate and the amount is flagged before the entry is filed.
Does it account for the customs value basis?
Yes. The Incoterms on the declaration determine whether freight and insurance sit inside the customs value, so a CIF basis and an FOB basis are captured distinctly rather than treated as the same number.
What fields are extracted from customs declarations?
Talonic returns customs declarations as schema-validated, typed fields. Common fields include Entry Number, Importer of Record, Country of Origin, HS / HTS Code, and more, each normalized (dates to ISO 8601, amounts as numbers) and mapped to a stable key so the output shape stays the same across layouts.
How accurate is extraction from customs declarations, and how is confidence reported?
Every extracted cell carries a confidence score from 0.0 to 1.0 and a provenance pointer back to the source page and region, so low-confidence values can be reviewed against the original before the data is trusted downstream. There is no single accuracy number: confidence is reported per field so you can gate on it.
Can I use customs declarations extraction in production?
Yes. The same engine behind this guide is available as a production REST API and Node SDK with sync, async, and streaming modes, schema versioning, signed webhooks, and EU-resident processing. Start free with an API key, then scale on usage-based pricing.
What does it cost to extract data from customs declarations?
There is a free tier for prototyping and agent evaluation with no credit card. Paid usage is credit-based at 1,000 credits per euro: page ingestion is 100 credits per page and registry-resolved queries are free. See talonic.com/pricing for current rates.
Ready to extract from your own customs declarations?
Author note
Reviewed by Talonic engineering, schema review · last reviewed 2026-06-13