Drop a CSV and see how every numeric column moves with every other one — Pearson or Spearman, with p-values, confidence intervals, a multiple-testing correction, and a scatter plot behind every cell.
Free, no sign-up. Runs entirely in your browser — the file is never uploaded.
Paste CSV text instead
How to read it
r runs from −1 (move in opposite directions) to +1 (move together). Around ±0.1 is negligible, ±0.3 moderate, ±0.5 and beyond strong — but what counts as “strong” depends on the field.
Pearson measures straight-line relationships and is sensitive to outliers. Spearman works on ranks, so it catches any consistently rising or falling relationship and shrugs off extreme values. When the two disagree a lot, look at the scatter plot.
p-values test “is r different from 0?”. With many columns you run many tests, so some will look significant by chance — the Holm correction (on by default) adjusts for that. A • marks pairs that stay significant at 5%.
Missing values are handled pairwise: each pair uses the rows where both columns have a number, and n is shown for every pair. ID-like columns (row numbers, customer_id, postcodes) are left out, because correlating labels is meaningless.
Correlation is not causation: two columns can move together because a third thing drives both.
one file vs. live data
This matrix is a snapshot. QuantumLayers watches it.
Connect a database, API or spreadsheet and QuantumLayers recomputes correlations as the data changes, alongside PCA, ANOVA, regression and AI-written insights — and can email you when a relationship shifts.