ROASTERDB Specialty Coffee Dataset

§ 00 · Cross-Source Pipeline & Flavor Mapping

Direct Roaster
Storefronts,
normalized.

The structured specialty coffee dataset — 8,000+ products from 280+ artisan roasters, with tasting notes mapped to the SCA Flavor Wheel.

Abstract

A structured catalog of 8,000+ specialty-coffee products from 280+ curated artisan roasters across 20+ countries. RoasterDB reads each roaster's live storefront and maps its tasting notes deterministically to standard hierarchical SCA Flavor Wheel nodes — with a source URL on every record so any fact can be re-verified.

Get the Dataset ↗ ↓ Sample Explorer 8,000+ rows

One-time purchase · Instant CSV / SQLite download · Zero recurring fees

8,000+
Specialty Coffees · 280+ Roasters · 20+ Countries
3,400+
Verified-Tier Records (origin + price + SCA flavor, QA-passed)
Coverage: Price 88% · Origin 58% · SCA flavor mapping 55% (11,000+ links) · Source URL on every record. Catalog-depth dataset — see field coverage before you buy.

§ 01 · Headline Metrics · Relational Integrity

4 Tables
Shopify API
Storefront Ingestion Engine

Direct automated extraction across `/products.json` and `/sitemap.xml` feeds without browser rendering latency.

SCA Wheel
3-Tier Hierarchical Taxonomy

Poetic terms (`"Grandma's peach cobbler"`) mapped directly to hierarchical paths (`Fruity > Other Fruit > Peach`).

Bean-Only
Non-Bean Filtering

Multi-layer keyword and product-type boundaries filter out mugs, grinders, apparel, gift cards, subscriptions, and merchandise.

§ 02 · Live Catalog Explorer Preview

Open Source Snapshot
ID Roaster Coffee Title Origin Elevation (masl) Process SCA Flavor Wheel Nodes Score
7704499585122 Onyx Coffee Lab Monarch - Cafes Only Colombia / Ethiopia 1,650 - 2,100m Natural Nutty/Cocoa > Cocoa > Dark Chocolate 92.5
bean_2078 Verve Coffee Roasters Streetlevel Guatemala / Colombia 1,500 - 1,850m Washed Sweet > Brown Sugar > Caramel, Fruity > Citrus > Orange 91.0
9142729113913 Black & White Edwin Noreña - Fruity Flowers Colombia (Quindio) 1,600 - 1,750m Anaerobic Fruity > Berry > Strawberry, Floral > Floral > Jasmine 94.0
882194011 Passenger Coffee Divino Niño 2024 Colombia (Huila) 1,800m Washed Fruity > Stone Fruit > Peach, Sweet > Honey > Honey 93.5
664192018 The Barn Volcan Azul Geisha Costa Rica 1,700m Natural Anaerobic Floral > Floral > Rose, Fruity > Tropical > Passion Fruit 95.0

Want to dig deeper? Try the interactive Sample Explorer on 🤗 Hugging Face → or download the free 100-row sample dataset.

§ 03 · Two Ways to Get the Data

Snapshot + Custom
Tier 1 · Self-Serve Snapshot

Full Snapshot Dataset

One-time download · 8,000+ records · Snapshot 2026.07

$49
  • 8,000+ specialty coffees from 280+ curated artisan roasters across 20+ countries.
  • Relational SQLite + CSV + JSON: normalized tables (`roasters`, `coffee_beans`, `sca_flavor_nodes`, `bean_flavors`), ready for ML, recommendation engines, and B2B apps.
  • SCA Flavor Wheel mapping: 11,000+ bean→flavor links as 3-tier paths (`Category > Subcategory > Descriptor`).
  • Full provenance: every record carries its source roaster, source URL, and crawl timestamp — traceable and re-verifiable.
  • Verified tier included: 3,400+ QA-passed records (origin + sanitized price + flavor) alongside the full catalog.
  • Snapshot 2026.07 · one-time purchase, no recurring fees. Distributed as derived factual attributes with attribution.
Buy Now · Instant Download · $49 →
Secure card checkout via Stripe · Apple Pay / Google Pay · dataset delivered instantly after payment
Need a company invoice or crypto payment? email our desk
Tier 2 · Custom & Fresh Data

Custom & Enterprise

Built to your spec · per engagement

$99
  • Custom roaster lists: we crawl the specific storefronts, origins, or regions you care about.
  • Recurring fresh snapshots: monthly or quarterly refreshes so the data doesn't go stale — a static snapshot ages fast.
  • Delivery your way: flat files, a hosted API endpoint, or a schema mapped to your product.
  • Commercial terms & support: invoicing, licensing scoped to your use, and integration help.

This is where most of the value is — a fresh, curated feed built to your requirements, not a one-off snapshot of public data.

Request a Custom Quote →

Prefer live, self-serve data? Run our scraper on Apify →

§ 04 · Frequently Asked Questions

Structured Knowledge Base

Dataset & Taxonomy FAQ

What is the RoasterDB Specialty Coffee Dataset?

RoasterDB is a structured, factual database containing over 8,000 specialty coffee releases from 280+ artisan roasters across 20+ countries. Every entry includes standardized attributes like origin country, altitude (masl), process method, varietals, and tasting notes mapped directly to the SCA Flavor Wheel.

How are tasting notes mapped to the SCA Flavor Wheel?

RoasterDB converts poetic or unstructured roaster descriptions (such as "Grandma's peach cobbler" or "bright clementine acidity") into strict 3-tier hierarchical SCA Flavor Wheel paths (for example: Fruity > Other Fruit > Peach). This enables precise semantic filtering and machine learning feature extraction.

What database formats are included in the download?

The download package includes a normalized relational SQLite database file (.db) with separate tables for roasters, coffee beans, and flavor node linkages, alongside clean CSV and JSON exports for direct import into pandas, R, Python scripts, or web applications.

Can I request custom scraping or real-time API access?

Yes. Our Tier 2 Custom & Enterprise offerings include tailored daily/weekly crawling pipelines, custom roaster target lists, and hosted GraphQL/REST endpoints. Contact RoasterDB@proton.me for scoped quotes.

§ 05 · Direct Engineering & Dataset Support

Have Questions or Need Custom Data?

Whether you need custom schema formats, API access, or enterprise invoicing, reach out directly to our desk at RoasterDB@proton.me.