Upload a CSV, pick your numeric columns, and get every anomaly ranked and explained — multivariate Mahalanobis scoring, the driving column named per row, and a 2D anomaly map. Free.
or click to browse · max 3 MB
Free analyses run on up to 10,000 rows. Larger files are randomly sampled to that size — sign up to analyze your full dataset.
Scanning for anomalies...
Sent to — ranked anomalies with driving features, score distribution, 2D anomaly map, column deviation profile, R code, and AI insights.
Analyze another fileEvery row is scored by its Mahalanobis distance from the multivariate center of your selected columns — a scale-free distance that accounts for how the columns co-vary, so a row can be flagged for an unusual COMBINATION even when no single value looks extreme. Rows beyond the 99.9% chi-square threshold are flagged, and each anomaly is explained by its dominant feature: the column with the largest robust z-score (median/MAD), stated as above or below typical.
Use it to sweep any table of numeric measurements for errors, fraud candidates, sensor glitches, or genuinely rare events — before those rows silently distort averages, models, or dashboards.
Not for categorical columns, and not a fraud verdict — it finds statistical outliers; deciding what they mean is domain work. Multi-modal data (distinct clusters) may need a cluster-aware method instead.
Built for: Analysts and operators auditing data quality or hunting unusual records
Typical data source: Any spreadsheet or CSV with several numeric columns
Any table with several numeric columns. For example, transaction records:
Minimum 20 rows · Best with 100-50,000 rows and 3-12 numeric columns
Standard-library analysis: find the unusual rows in your dataset across several numeric columns. Robust per-column z-scores (median/MAD) combined with multivariate Mahalanobis distance flag the rows that don't fit the overall pattern — with the driving column named for every anomaly, a score histogram, a 2D anomaly map, and the columns that cause the most trouble. Works on any dataset: map 2 or more numeric columns.
The bulk of rows pile up at low scores; the long right tail is where anomalies live — the further right, the stranger the row.
The most unusual rows ranked by score, each explained: which column drives it, its value versus typical, and the direction.
The whole dataset on its two main axes with anomalies highlighted — see whether outliers scatter randomly or cluster in one direction.
Which columns carry the extreme values across the flagged rows — where to focus data-quality attention.
Plain-English interpretation — what the numbers mean, what's significant, and what to do next.
Which of my rows don't belong?
Map your numeric columns. Every row gets a multivariate anomaly score; the flagged rows come back ranked, each with the column that makes it unusual and how far above or below typical it sits.
See our FAQ for details on pricing, data privacy, and how the analysis works. Every report includes a Methodology section showing the statistical test, assumptions checked, and diagnostics run.
Run any analysis on your own data — validated R analyses, interactive reports, AI insights, and PDF export.
Try Free — No Credit CardTell us what went wrong, in your own words. We capture the page you're on automatically, so no need to describe where you are.