Upload a CSV, pick your outcome and drivers, and get the statistically correct regression — logistic for yes/no, Poisson for counts with an overdispersion check, least squares for continuous — with every effect as an odds ratio, rate ratio, or coefficient. 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.
Inspecting the outcome and fitting the right model family...
Sent to — family decision trail, effects with confidence intervals, partial-effect curve, predicted-vs-actual calibration, diagnostics, R code, and AI insights.
Analyze another fileThe analysis inspects the outcome column and chooses the generalized linear model family from its values: exactly two distinct values gives binomial (logistic) regression; non-negative whole numbers give Poisson regression, with a Pearson-dispersion overdispersion test (threshold 1.5 plus a chi-squared check) that upgrades the fit to quasi-Poisson when the counts vary more than Poisson allows; any other numeric outcome gives gaussian ordinary least squares. Effects are reported with 95% Wald confidence intervals on the family's native scale — exponentiated to odds ratios or rate ratios for the logit/log links — alongside deviance explained, AIC where defined, a partial-effect curve for the strongest numeric predictor, and a predicted-vs-actual check.
Use it when you want a driver analysis but the outcome's type should decide the model — or when you're not sure whether linear, logistic, or Poisson regression is the right call for your column.
Not for multi-category outcomes (use classification drivers), not for time-to-event data (use survival analysis), and not for counts with wildly different exposure windows per row.
Built for: Analysts and operators who want the right regression without pre-deciding the family
Typical data source: A spreadsheet or CSV with one outcome column (revenue, churn flag, order count) and several candidate driver columns
One outcome plus candidate drivers — the outcome's type picks the family. For example, weekly revenue per store:
Minimum 30 rows · Best with 200-20,000 rows with 2-6 predictors
Standard-library analysis: regression that inspects your outcome column and chooses the correct model family instead of assuming one — what it adds over plain linear regression. A yes/no outcome gets logistic regression with odds ratios; a non-negative whole-number count gets Poisson regression with rate ratios (upgraded to quasi-Poisson when an overdispersion test demands it); a continuous outcome gets ordinary least squares with plain coefficients. The family decision and its evidence are reported openly, every effect carries a 95% confidence interval and plain-language significance, and you get the strongest predictor's partial-effect curve plus a predicted-vs-actual fit check (binned calibration for logistic). Works on any dataset: map one outcome and 1-8 predictors.
Each predictor's effect on the family's native scale — odds ratio, rate ratio, or plain coefficient — with confidence intervals and plain-language significance.
The strongest predictor's partial effect: predicted outcome across its range with everything else held typical.
Predictions against reality on the diagonal of perfect prediction — a binned calibration plot when the outcome is yes/no.
The family decision trail with its evidence, the overdispersion verdict for count models, deviance explained, AIC, and rows used.
Plain-English interpretation — what the numbers mean, what's significant, and what to do next.
Which regression does my data actually need?
Map your outcome and candidate drivers. The analysis inspects the outcome — two values, whole-number counts, or continuous — picks logistic, Poisson, or ordinary least squares accordingly, and shows you the evidence behind the choice.
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.