Statistics should not be the gatekeeper.

Most people who need statistics aren’t statisticians. They’re researchers running an experiment, students wrapping up a thesis, engineers comparing two builds, analysts trying to make sense of a survey. They have a real question and a real dataset — and not enough time to learn which test is correct, in which scenario, with which assumption.

Decisionly was built for that gap. You upload a file. The engine looks at the shape of your data — variable types, sample sizes, normality, variance — and picks the test that actually fits. It runs it, reads it back to you in plain English, and hands you a clean PDF.

We care about getting the answer right more than getting it fast. The selection rules are rooted in the same textbooks a statistician would defend. The interpretations don’t hand-wave. The PDF doesn’t hide its assumptions. If a test would be wrong, the app stops you instead of letting you ship a misleading result.

We’re early. The product is free during beta and the roadmap is open. If you’ve broken something, want a test we don’t cover, or disagree with how we explain a result — we want to hear about it.