Upload a CSV with a date, a metric, and a before/after column. Get an interrupted time-series analysis: the expected path your metric was already on, the gap the event created, and an honest verdict with real uncertainty. 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.
Building the counterfactual and measuring the gap...
Sent to — actual-vs-expected chart, effect sizes with 95% intervals, model diagnostics, R code, and AI insights.
Analyze another fileAn interrupted time-series analysis. The expected path of the metric (the counterfactual) is fitted on the pre-event rows alone — a linear time trend plus day-of-week or month-of-year adjustments when the history supports them — then projected across the post-event window with 95% prediction intervals. The effect is the gap between what actually happened and that expected path: reported per period, as a percentage of the expected level, and cumulatively, with a significance check from the interval, the share of points outside the band, and a t-test of post-event gaps against pre-event residuals.
Use it whenever you know WHEN something changed and want to know whether a metric moved because of it — launches, price changes, campaigns, policy changes, outages.
Not for data without dates, for events with no clean before/after split, or when you ran a proper randomized experiment (use the A/B test tool — randomization beats a counterfactual).
Built for: Marketers, product managers, and operators who shipped a change and need to know if it worked
Typical data source: A dated metric export (daily revenue, signups, sessions, units) plus a column or a known date marking when the change happened
A dated metric with a column marking before vs after the event. For example, daily revenue around a campaign launch:
Minimum 15 rows · Best with 60-1,000 rows: a few weeks to a couple of years of daily or weekly data, with a solid pre-event history
Standard-library analysis: interrupted time-series measurement of an event's impact. Mark which rows fall before and after the event (a launch, price change, campaign start); the tool builds a counterfactual from the pre-event trend and seasonality, projects it across the post-event window with 95% prediction intervals, and measures the gap — average, relative, and cumulative effect, each with honest uncertainty. Works on any dataset: map a date column, a numeric metric, and a before/after column.
The actual metric plotted over time next to the expected counterfactual path across the post-event window — where the points pull away from the expected series is the event's apparent impact.
The effect in three sizes — per period, relative, and cumulative — each with a 95% interval, plus how many post-event points escaped the prediction band.
The counterfactual's report card: how well the pre-event model fits, how noisy the series is, and whether post-event deviations differ statistically from pre-event noise.
Plain-English interpretation — what the numbers mean, what's significant, and what to do next.
Did the campaign actually work?
Mark your rows before/after the campaign start. The tool projects where the metric was already heading from the pre-period alone, then measures how far the post-period ran above or below that expected path — with an honest interval, not just a before/after average.
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.