What is a Taxonomy?

What is a Taxonomy?

A Taxonomy is a hierarchical classification system that organizes data, content, or metadata into parent-child categories. While an Ontology is a web of relationships that maps how different concepts interact (e.g., “A Customer buys a Product”), a Taxonomy is a top-down tree designed to classify what a thing is (e.g., “A Laptop is a type of Computer, which is a type of Electronics”).

In Data Lakehouses and enterprise data catalogs, a shared Taxonomy makes data much easier for people to find. If a company loads millions of PDF reports, JSON logs, and Parquet files into an Amazon S3 bucket without classifying them, the data is hard to search. The lake becomes a swamp because employees have no consistent way to drill down and filter the data.

The Architecture of the Tree

A Taxonomy is built on “Is-A” (Parent-Child) relationships. Each node in the tree is a more specific sub-category of the node directly above it.

Consider an e-commerce company structuring its product data. The Taxonomy defines the hierarchy:

1. Root: All Products
2. Level 1: Electronics
3. Level 2: Computing
4. Level 3: Laptops
5. Level 4: Gaming Laptops

Many taxonomies aim for Mutually Exclusive and Collectively Exhaustive (MECE) categories. A specific product (e.g., a gaming laptop) belongs under Gaming Laptops, not Kitchen Appliances. Some taxonomies allow an item to appear under more than one parent (a polyhierarchy), but each category still has a clear place in the tree.

Taxonomies in Data Governance

Taxonomies are one of the main tools Data Stewards use to organize an enterprise data catalog (like Alation or Collibra).

When a new Apache Iceberg table is created in the Data Lakehouse, the Data Steward does not just type free-form keywords into the description. They tag the table using terms from the shared taxonomy.

Security and Access Control

Because the Taxonomy is hierarchical, platforms that support tag-based policies can use it to simplify access control. The security architect writes a single rule: “The Marketing team cannot access any data tagged under the Human Resources -> Payroll branch.” When tables and columns are tagged against the Taxonomy, the policy applies to everything under that branch. A new table tagged as Payroll is blocked for the marketing team as soon as the tag is applied, without a new rule.

Summary of Technical Value

A Taxonomy is a core organizing framework for enterprise data. By setting a standard, top-down hierarchy to classify assets, it reduces inconsistent categorization. It gives people a clear structure for discovering data and gives tag-based security and governance policies a consistent set of categories to act on across the Open Data Lakehouse.

Learn More

To learn more about the Data Lakehouse, read The Lakehouse Built for Everyone by Alex Merced. You can find this and other books by Alex Merced at books.alexmerced.com.