Every outbound campaign is an experiment you get to run once. If the reply rate is wrong, you discover it after the send is gone. The Digital Twin flips that: it runs hundreds of Monte Carlo passes over your real lead mix and your sequence, then reports the p10–p90 range of meetings you should expect.
The interesting output isn't the median. It's the bottleneck. When a twin says the constraint is show-up rate, not reply rate, you stop rewriting subject lines and start fixing the booking flow. That is a change you can make before the campaign costs you anything.
We cap it with a verdict — GO, FIX or KILL — so the decision is explicit. A model that only ever says 'looks promising' isn't a control system; it's a horoscope.