← Technical glossary
Definition

Clustering

Unsupervised methods that group observations by similarity so analysts can explore structure and design different actions.

Why it matters

The practical consequence.

Clustering can reveal useful customer, store, or product groups when the groups are stable and lead to meaningfully different decisions.

Example

How it appears in a system.

A retailer might identify high-frequency low-margin shoppers separately from occasional high-value purchasers.

Continue learning
Browse engineering insights →