Introducing Stratum: Serious Statistics, Native to Mac & Windows

Stratum is a statistics and data-visualization app built the way a native desktop app should be: fast, native, offline, and focused. Here's what it does, who it's for, and why we made it.

For years, doing real statistics on your desktop meant choosing between two uncomfortable options. You could live in a notebook — spinning up Python or R, gluing together libraries, and re-plotting everything by hand — or you could rent a heavyweight desktop suite that feels like it was designed for a different decade and a different operating system. Neither felt like a real desktop app. Stratum is our answer to that gap.

What Stratum is

Stratum brings the whole analysis loop into one native window: import, wrangle, visualize, test, model, and export — point, click, and read the result. There's no scripting, no notebook to maintain, and nothing leaves your machine.

Fast import. Drag in a CSV, TSV, or Excel file and Stratum auto-detects delimiters and infers column types, opening straight into a native data grid that stays smooth on millions of rows.
Wrangling without code. Re-editable formula columns with 100+ functions, a dockable filter inspector, and join / stack / aggregate / dedup / sample.
30+ charts. Histograms, box & violin plots, density, ECDF, scatter, Pareto, mosaic, heatmaps and more — each with live controls and one-click PNG / PDF export.
Real inferential statistics. ANOVA, MANOVA, correlation, two-sample and contingency tests, and linear regression with full diagnostics — parametric and nonparametric.
SPC and machine learning. Control charts with all eight Nelson rules, plus PCA, clustering, decision trees, and ensemble methods — all built in.

Why native matters. The data grid is a native virtualized grid, not a web view, so scrolling and filtering stay responsive at scale. Your work — data, analyses, filters, and styling — saves into a single .stratum document you can reopen exactly where you left off.

Who it's for

Stratum is a general-purpose statistics and charting package. If you analyze data — as an analyst, a scientist, an engineer, a student, or anyone who has ever opened a spreadsheet and wished it could do more — Stratum is built for you. It is deliberately broad rather than specialized: a focused, standard toolkit that follows the conventions you already expect from NumPy, pandas, R, and JMP.

Where to start

The fastest way in is the Stratum Tutorial — a free 25-lesson course that pairs a real dataset with each major feature, including the exact clicks to reproduce every result. And right here on the blog, we'll publish how-tos, worked case studies, and statistics explainers every week.

Download Stratum