Clusters — segmentation | ATX AnalyticsTable Docs

Clusters — segmentationk-means segments found in several numeric columns at once: features, label, automatic k, scaling, the scatter and the profile, quality, reproducibility and cross-filtering. A cluster chart answers "what kinds of customers (products, desks, counterparties) do we have?": it looks at several numeric columns at once — spend and visits, P&L and volume, exposure and turnover — and groups the rows that resemble each other, using k-means. Insert one from Dashboard → Clusters, from Reports → Clusters on a sheet, or pick Clusters as any chart's type; it uses the same sources, title, legend, fonts, number format, palette, PNG, SVG and CSV buttons and PDF reports as every chart, and saves with the project.

Where to find it

Reports → Charts → Clusters
Dashboard → Add Widget → Clusters

What this page covers

Clusters — finding segments in several columns at once
Clusters — quality, reproducibility, large tables, privacy and Ask Greg

Settings and controls documented

Label column (what one point is)
Combine rows per label
Features
k from a parameter
Outlier distance (× IQR)
Cluster names
Random seed
Row limit

Full page

The full page — procedures and the explanation of every setting — is available to signed-in users.

Glossary terms on this page

KPI dashboard: definition

ATX Institute on LinkedIn

This site needs JavaScript for the full experience. Every page above is reachable without it.