What if you could test a decision before you make it?
Run the decision before you make it.
Test a decision against ten thousand simulated futures — across fifteen domains, from ICU capacity to supply chains to portfolios. Every public run is replayable, its assumptions visible, and graded against what actually happened.
Latest run
If the carrier suspension on the Asia–EU lane holds six weeks, retail lead times breach SLA in 62% of runs
Median delay +9.4 days (CI 7.1–12.8). The largest driver is tier-2 supplier rerouting capacity, not port congestion. All assumptions are visible in the run.
Recent runs
Under a rate hold plus oil spike, P99 drawdown widens 3.1 points for 60/40 portfolios
10,000 runs · CI on every metric
A summer flu wave at +18% admissions pushes ICU diversion risk past 20%
10,000 runs · assumptions visible
A comp freeze raises 12-month attrition by 4.7% in 71% of scenarios
10,000 runs · fork and re-run
Track record: every public prediction is graded when reality arrives — including the misses, with post-mortems. The calibration score is published and updated after each grading.
See every prediction →Two ways in
Make a decision
Browse runs in your domain, read the findings, fork the assumptions, and run your own scenario. If a consultant sent you a verified finding, the replay link shows you everything they saw.
Browse by domainRun decisions for clients
Put a verified, replayable simulation behind every recommendation. The mark certifies the math; your assumptions stay yours — and stay visible. Findings your client can try to break, and can't.
For consultantsWhat this is
An observatory, not an oracle
Decision Observatory is a simulation instrument. Describe a decision, and the engine runs thousands of parallel futures — each with different sampled inputs — and reports the distribution of outcomes: the median, the tails, the confidence intervals, and which assumption moves the result most.
It does not predict the future. It quantifies exactly how uncertain the future is, given what you believe today — and it publishes its own grades so you can judge how well-calibrated those quantifications have been. We don't ask you to trust us. We show you the record.
Before launch
Join the waitlist
The observatory opens with its first live, graded runs. Leave your email and you'll get two things: a note when the instrument opens, and first claim on one of the 100 founding-member seats at $99/month, locked for life. Nothing else — no newsletter.