ImportYeti Scraper — US customs import records

The importer and the shipper named on a US inward vessel manifest are disclosable to anyone who asks CBP for them, unless that importer has filed a biennial certification claiming confidential treatment. That default is the reason an outsider can see who supplies whom at all.

ImportYeti indexes those records, and this actor reads them: which American companies import, who ships to them, how many shipments, what the goods were, the HS code, the ports at both ends, and the first and last shipment dates.

It's a plain HTTP scraper. No browser, no login, no ImportYeti account. One request per company page.

What you get

resultType picks the row shape:

Mode	One row is	Rows per company
partners (default)	one supplier ↔ buyer relationship	usually 30 to 50, sometimes 80-odd
shipments	one bill of lading	up to 100
profiles	one wide row, partners and lanes nested	1
search	one match from ImportYeti's index, no profile page opened	—

A trading-partner row from a real run:

{
  "profileType": "company",
  "profileName": "Nike",
  "profileUrl": "https://www.importyeti.com/company/nike",
  "profileAddress": "1 Bowerman Dr, Beaverton, Or 97005, Us",
  "profileCountry": "United States of America",
  "profileCountryCode": "US",
  "profileTotalShipments": 560,
  "rowType": "partner",
  "counterpartRole": "supplier",
  "counterpartName": "Apl Logistics Vietnam",
  "counterpartUrl": "https://www.importyeti.com/supplier/apl-logistics-vietnam",
  "counterpartAddress": "C O Apl-Nol Vietnam Limited 17 Ba Huyen Thanh Quan Dist 3 Hcmc Vietnam",
  "counterpartCountry": "Vietnam",
  "counterpartCountryCode": "VN",
  "counterpartLocation": "Xa Ba Thanh",
  "shipments": 189,
  "shipmentsLast12Months": 0,
  "weightKg": 584921,
  "teu": 86,
  "productDescription": "Roll Gka Nike Off Dock Cargo Dest Footwear, Women Polyester Top, Sports",
  "hsCode": "6402.99",
  "hsCodes": ["6402.99", "6404.11", "6101.10", "6403.99", "6109.10"],
  "topProductCategory": "Footwear",
  "originPort": "Vung Tau",
  "originPortCountry": "Vietnam",
  "destinationPort": "Long Beach, Ca",
  "destinationPortCountry": "United States of America",
  "destinationPortRegion": "California",
  "firstShipmentDate": "2015-01-09",
  "lastShipmentDate": "2023-04-01",
  "isRelatedParty": false,
  "counterpartTopBuyers": [
    { "name": "Apl Logistics", "shipments": 52390 },
    { "name": "New Balance Athletics", "shipments": 2602 }
  ],
  "scrapedAt": "2026-08-21T16:20:14.087Z"
}

Shipment rows add shipmentDate, billOfLading, masterBillOfLading, billTypeCode, quantity, quantityUnit, containers, declaredValueUsd and shippingRoute.

Profile rows add alsoKnownAs, website, email, phone, estimatedShippingCostUsd, topCountries, and nested tradingPartners[], tradeLanes[], hsCodes[], recentShipments[].

Across the 1,000-row run priced below, HS code, product description, last shipment date and the counterparty's ImportYeti URL were on 100% of rows, and origin port on 90.7%.

Input

{
  "searchQueries": ["solar panels"],
  "startUrls": ["https://www.importyeti.com/company/nike-usa"],
  "resultType": "partners",
  "maxItems": 200,
  "maxProfilesPerQuery": 3
}

Search terms work on brands (Nike), importer legal names (Nike Usa), supplier names and products (lithium battery). ImportYeti files one brand under many name variants, so maxProfilesPerQuery controls how many of those variants actually get opened.

includeUnnamedPartners is off by default. US manifests are sometimes filed with the counterparty blank and ImportYeti shows those as "Missing in source document"; leaving it off means you're not charged for rows with no name on them.

Price

Event	Price	When
Result	$0.000025 ($0.025 / 1,000)	per row in the dataset
ImportYeti profile opened	$0.0005	per company page that returned data
Actor start	$0.0005	once per run

Two events instead of one because the two costs are genuinely different. Opening a company page means fetching and parsing about a megabyte of HTML; writing a row costs almost nothing. Charging everything per row would mean overcharging you for the 45 partner rows that come out of a single page.

Real numbers from real runs:

• Nike USA's full supplier list — 1 page, 45 rows: $0.002125
• 1,000 partner rows across 37 companies: $0.044
• 40 Home Depot bills of lading: $0.002
• A run that finds nothing: $0.0005, with no result or profile events charged

Blocked pages, 404s, empty tables and bad URLs return a diagnostic row and are never charged.

What it doesn't do

• No air freight, no exports. ImportYeti covers US sea import manifests from Customs and Border Protection. Trucking, rail, air and outbound shipments aren't in there.
• Top partners and recent bills of lading, not the full history. That's what ImportYeti's own page carries. Home Depot USA has 230,803 shipments on record; the page lists its 48 largest supplier relationships and the 100 most recent BOLs.
• Ports are per-lane, not per-shipment. ImportYeti publishes port pairs for a company, not for each relationship. A row's originPort is the busiest lane on that company leaving the counterparty's country. Where ImportYeti lists no lane out of that country the field is null rather than a guess — 9.3% of rows in the 1,000-row run above.
• Names are as filed. US manifests are typed by hand, so you get Wal Mart, Nike Usa, Apl Logistics Vietnam. Match on the ImportYeti URL, not the name.
• Email and phone are mostly withheld. ImportYeti masks contact details behind its own paid plan (XXX-XXX-X319). The fields are returned when the page shows them, which isn't often.
• Throughput has a ceiling. ImportYeti meters company page loads per IP address — roughly 25 before it starts serving a stripped page, and the cut-off lasts several minutes. The actor rotates addresses on every request, which is what lets a run read hundreds of profiles, but a very large run can still hit the wall. It says so in the log and charges you nothing for it.

Questions

What is ImportYeti? A free search engine over US Customs and Border Protection bill-of-lading records. It shows which US companies import by sea, which overseas factories ship to them, and what they ship.

How do I find a company's suppliers? Put the brand in searchQueries and leave resultType on partners. Each row is one supplier, with its country, shipment count, HS code, product description and ImportYeti profile URL.

How do I find a factory's customers? Paste a supplier URL (https://www.importyeti.com/supplier/...) into startUrls. On a supplier page the same table runs the other way, so counterpartRole comes back as buyer.

Do I need an ImportYeti account? No. Everything here comes from public pages.

Is scraping ImportYeti legal? The underlying records are US government customs data, published by CBP. This reads pages ImportYeti serves publicly and takes nothing from behind its login. What you do with the data afterwards is on you.

What is an HS code? The Harmonized System code that classifies the goods — 6402.99 is footwear, 9403.60 is wooden furniture. hsCode is the code that counterparty declares most often; hsCodes lists the top five.

Why is a port missing on some rows? See above. A null means ImportYeti's lane table has no route out of that counterparty's country, and a guessed port would be worse than none.

Can I get more suppliers for one company? Not from one page. The partner table is as long as ImportYeti makes it, which was 45 rows for Nike USA and 48 for Home Depot USA. Raise maxProfilesPerQuery instead: large importers file under several names (Nike, Nike Usa, Nike Canada) and each has its own page and its own partner table.

Output fields

profileName · profileType · profileUrl · profileAddress · profileCountry · profileCountryCode · profileTotalShipments · rowType · counterpartRole · counterpartName · counterpartUrl · counterpartAddress · counterpartCountry · counterpartCountryCode · counterpartLocation · shipments · shipmentsLast12Months · weightKg · teu · productDescription · hsCode · hsCodes · topProductCategory · originPort · originPortCountry · destinationPort · destinationPortCountry · destinationPortRegion · firstShipmentDate · lastShipmentDate · isRelatedParty · counterpartTopBuyers · searchQuery · scrapedAt