Product Categorisation API
Structured taxonomy
for every product
in your catalogue.
Clusteva takes raw product titles and catalogue text and returns precise Google taxonomy categories — via a single API call. Built for e-commerce developers, feed management tools, and catalogue enrichment pipelines.
Request
POST /v1/categorize HTTP/1.1
Host: api.clusteva.com
Authorization: Bearer <token>
Content-Type: application/json
{ "id": "p1", "product_name": "Sony WH-1000XM5 Wireless Headphones" }
---
{ "id": "p2", "product_name": "12-Cup Programmable Coffee Maker, Black" }
---
{ "id": "p3", "product_name": "Men's Slim-Fit Chino Pants 32x30" }Response
HTTP/1.1 200 OK
{ "id": "p1",
"category": "Electronics > Audio > Audio Components > Headphones & Headsets > Headphones",
"taxonomy_id": "google_543626",
"confidence": 0.97 }
---
{ "id": "p2",
"category": "Home & Garden > Kitchen & Dining > Kitchen Appliances > Coffee Makers & Espresso Machines > Drip Coffee Makers",
"taxonomy_id": "google_1388",
"confidence": 0.94 }
---
{ "id": "p3",
"category": "Clothing & Accessories > Clothing > Trousers & Jeans",
"taxonomy_id": "google_204",
"confidence": 0.99 }How it works
Send product text
Pass a product title. Plain text, no preprocessing required.
Clusteva categorises
The API resolves each item against the full Google product taxonomy — 5,500+ categories, continuously maintained.
Use structured output
Get back the category path, taxonomy ID, and a confidence score. Wire it into your pipeline in minutes.
Use cases
Catalogue cleanup
Retailers with legacy catalogues often have thousands of products sitting in wrong or missing categories. Clusteva re-classifies them in bulk without manual review.
Marketplace onboarding
When sellers upload products, enforce consistent taxonomy immediately. Reduce listing rejections and improve search relevance from day one.
Feed management
Google Shopping and comparison engines require accurate category IDs. Clusteva ensures your feed always satisfies taxonomy requirements.
Enrichment pipelines
Add categorisation as a step in your data pipeline alongside title normalisation, attribute extraction, and deduplication.
The taxonomy problem is harder than it looks.
Google's product taxonomy has over 5,500 leaf categories. Mapping products to the right node — not just a plausible one — requires training data, ongoing maintenance, and careful handling of edge cases.
Most teams that build this in-house underestimate the maintenance burden. A rule-based classifier drifts as taxonomy updates ship. A model trained once degrades on new product types.
Clusteva owns the taxonomy layer so you don't have to. Call the API, get accurate categories, move on.
Taxonomy coverage
5,500+ categories
Full Google product taxonomy, including all leaf nodes.
Integration time
< 30 minutes
One endpoint, standard REST. No SDK required.
Maintenance burden
Zero
Taxonomy updates are absorbed on our side.
API reference
One endpoint. Predictable output.
curl -X POST https://api.clusteva.com/v1/categorize \
-H "Authorization: Bearer $CLUSTEVA_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"id": "sku-847",
"product_name": "Ergonomic Lumbar Support Mesh Office Chair"
}'{
"id": "sku-847",
"category": "Furniture > Office Furniture > Office & Desk Chairs",
"taxonomy_id": "google_2045",
"confidence": 0.96
}Also available as POST /v1/categorise (same behaviour).
Responses follow a stable schema. Confidence scores are calibrated against Google taxonomy acceptance rates.
Get started