A commitment to what engineering will know by a specific date — not what will be built
Specific questions this phase will answer. Each should have a clear, observable answer — not "learn more about X."
What will we decide based on what we learn? Define the conditions upfront — including the conditions that would lead to stopping.
This is a learning commitment, not a delivery commitment.
The decision gate is the most important section. It transforms discovery from open-ended research into a structured experiment with clear outcomes. By defining what would make you continue, pivot, or stop before you begin, you remove the temptation to retrofit conclusions to match expectations.