The engine is not magic; the quality of your question determines the quality of the panel. Three habits improve every answer:

1. Name the stake. "Should we acquire ACME?" is sharper than "What about ACME?" because it tells the panel what kind of answer to produce.
2. Name the constraint. Budget, timeline, irreversible commitments. The Ground seat does its best work when the constraints are explicit.
3. Name what counts as success. "Increase margin without losing top quartile customers" is testable. "Grow" is not.