How it works

1. You describe

Your project in your words. A language model only extracts the facts: age, level, time, goal, horizon. It never estimates anything. When a fact is missing, we ask.

2. We find your comparables

Cohorts dated at their start and followed to their outcome: chess players from monthly rating lists, crowdfunding projects from monthly snapshots, companies from public registries. No survivor lists, no success stories.

3. You get the chance

A statistical model computes the chance, its range, and what moves it among people like you. The evidence and validation available vary by model; consult the result and its sources.

What we do not do

We do not guess when the data is too thin. We do not show causes we have not proven. We do not name anything about you that is not a lever you can act on.