⚙️

HS Code Classification

ActiveCoreMUSTFREE

What it does

Classify any product into its correct HS (Harmonized System) code using POTAL's v3.3 GRI pipeline. The system uses 595 codified rules across all 21 Sections of the HS nomenclature — zero AI calls, zero per-request cost. Supports text-based classification with 9-field input, image-based classification, and batch processing for hundreds of products at once.

Required Fields for 100% Accuracy

1 required / 9 total

Field	Type	Required	Description	Example
`productName`	string	!	Product name in English, as specific as possibleInclude material + form. Exclude brand names.	`Men's cotton t-shirt, knitted`
`category`	string	opt	Product category (e.g., apparel, electronics, food)Use broad category, not material. "apparel" not "cotton".	`apparel`
`material`	string	opt	Primary material (WCO material group)Use WCO material terms: cotton, polyester, stainless steel, wood, etc.	`cotton`
`processing`	string	opt	How the product is manufacturedknitted, woven, molded, cast, forged, printed, etc.	`knitted`
`composition`	string	opt	Material composition breakdownInclude percentages when mixed materials.	`95% cotton, 5% elastane`
`weight_spec`	string	opt	Weight specificationPer-unit weight helps distinguish tariff subheadings.	`200 g/m2`
`price`	number	opt	Unit price in USDPrice-break rules apply to some HS headings (e.g., "valued over $14/kg").	`14.99`
`origin_country`	string	opt	Country of origin (ISO 2-letter code)Use ISO codes: CN, US, DE, JP — not full country names.	`CN`
`destination_country`	string	opt	Destination country (ISO 2-letter code, default: US)Affects HS10 subheading selection for 7 countries with 10-digit codes.	`US`

Tips for Best Accuracy

✓All 9 fields provided = highest confidence classification
✓productName alone works but may return lower confidence
✓Adding material + processing dramatically improves accuracy for textiles and metals
✓price field is critical for HS headings with "valued over/under $X" rules
✓destination_country affects 10-digit HS code selection (US, EU, GB, CA, AU, JP, KR)

Common Mistakes

✗Putting material in the category field — "cotton" is a material, "apparel" is the category
✗Using full country names instead of ISO codes — use "CN" not "China"
✗Too-generic product names — "shirt" is vague, "men's cotton knitted t-shirt" is precise
✗Including brand names — "Nike Air Max" should be "running shoes, leather upper, rubber sole"
✗Mixing up processing and composition — processing = how it's made, composition = what it's made of

How to use it

Get your API key

Prepare your product data

At minimum, provide a productName. For best accuracy, include category, material, and processing details.

Send a POST request

Call POST /api/v1/classify with your product data. The response includes the HS code, confidence score, and classification path.

Review the result

Check the confidence score and alternatives array. If confidence is below 0.8, consider providing more product details.

API Reference

POST/api/v1/classify

Request

{
  "productName": "Men's cotton t-shirt",
  "category": "Apparel"
}

Response

{
  "hsCode": "6109.10",
  "description": "T-shirts, singlets and other vests, of cotton, knitted or crocheted",
  "confidence": 0.95,
  "alternatives": [
    {
      "hsCode": "6109.90",
      "confidence": 0.82
    }
  ]
}

cURL

curl -X POST https://potal.app/api/v1/classify \
  -H "X-API-Key: pk_live_your_key" \
  -H "Content-Type: application/json" \
  -d '{"productName": "Men\'s cotton t-shirt", "category": "Apparel"}'

Related Features

⚙️Duty Rate Calculation

Look up MFN, MIN, and AGR duty rates for 240 countries

⚙️Confidence Score

Multi-dimensional confidence scoring for every classification result

⚙️Batch Classification

Classify hundreds of products at once via CSV or JSON batch API

⚙️HS Code Validation

9-field input validation ensuring accurate classification inputs

⚙️Image Classification

Classify products from photos using AI vision analysis

Having issues?

🐛 Report a Bug ❓ Ask a Question 💡 Suggest Improvement