Upload a CSV, pick an outcome and your candidate predictors, and get a permutation-importance ranking measured out of sample — with error bars, a verdict on which ranks are genuinely separated, and a warning when correlated columns are splitting the credit. Free.
Free analyses run on up to 10,000 rows. Larger files are randomly sampled to that size — sign up to analyze your full dataset.
Shuffling predictors and measuring what the model loses...
Sent to — a ranked importance chart with error bars, drop-column and standalone comparisons, a rank-reliability verdict, a collinearity diagnostic, R code, and AI insights.
Analyze another fileA linear model is fitted for a numeric outcome and a logistic model for a two-level one, and scored by R-squared or AUC on cross-fitted predictions so that no row is ever scored by a model that trained on it. Each predictor is then shuffled inside the held-out folds and the already-fitted model is re-scored: the fall in performance is its permutation importance. The entire procedure — resample the rows, redraw the folds, refit the models, reshuffle — is repeated many times, and the spread across those repeats becomes the interval on each importance and the basis for judging whether one predictor genuinely outranks the next. Alongside it the analysis refits without each predictor (drop-column importance), refits with each predictor alone (standalone performance), reports the direction of each effect, and computes variance inflation plus correlated groups so that overlapping predictors are flagged rather than ranked against one another.
Use it when you have an outcome and several candidate explanations and need to know which columns a model actually depends on, how confident that ranking is, and which columns are duplicating each other.
Not for establishing that changing a predictor would change the outcome — that needs an experiment or a causal design. Not for per-observation explanations, which are a different quantity from the global importance reported here. Not when the real relationship is a strong interaction or a sharp non-linearity, which a linear or logistic model will understate.
Built for: Analysts and data scientists deciding which variables matter, and which are worth continuing to collect
Typical data source: Any table with one outcome column and several candidate predictor columns
One row per observation, with the outcome and every candidate predictor as columns. For example, weekly marketing performance:
Minimum 60 rows · Best with 300-4,000 rows with 3-12 candidate predictors
Standard-library analysis: of the columns you think explain an outcome, which ones does a model actually depend on? Each candidate predictor is shuffled in turn and the model is re-scored on rows it never trained on, so the fall in performance measures how much the fit really leans on that column. The whole procedure — split, refit, shuffle — is repeated on resampled data, so every importance arrives with an uncertainty interval instead of a single number, and the report says which steps in the ranking are genuinely separated and which are a coin flip. A drop-column importance, the direction of each effect, a collinearity diagnostic that names correlated groups, and the distinction between redundant and irrelevant come with it. Works on any dataset: map a numeric or yes/no outcome and two or more candidate predictors.
Each predictor ranked by what the model loses when that column is shuffled, with error bars — overlapping bars are not distinguishable from each other.
Permutation importance beside drop-column importance and standalone performance, so you can see which predictors are unique and which are duplicating each other.
For every step in the ranking, the share of resampling reruns in which it held — the card that says which parts of the order are real.
Correlated groups with the joint contribution that is actually identified, and the difference between a redundant predictor and an irrelevant one.
The exact procedure, what the interval covers, and the explicit limits — including that no SHAP value is computed and that nothing here is causal.
Plain-English interpretation — what the numbers mean, what's significant, and what to do next.
Which of these columns is the model really using?
Map the outcome and every candidate predictor. Each one is shuffled in turn and the model is re-scored on rows it never trained on, so you see the fall in performance that each column is worth — with an interval, not a bare number.
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.