Published by the people who ran them
Case studies
Every case study here is a real validation: someone tested an idea on a simulated crowd, scored that verdict against real-world responses, and published how closely the two matched. A score can come from a hosted survey or from numbers the owner entered. Each card says which.
No published case studies yet. Run a simulation, score it against a hosted survey or numbers you already have, and publish yours from the report page.
In the meantime, the accuracy page names survey alignment and known-outcome backtests. Unscored runs stay unscored.
Pay once for one full simulation: 500 people, 20 rounds, 3 platforms. Validate a simulation against real data, then publish it here from the report page. See how the score works.