A modernization plan can contain a detailed delivery schedule and still leave the hardest questions unanswered. Which behavior must be preserved? Who decides whether a replacement is ready? I want those questions discussed before the team becomes attached to a migration date.
Plan for evidence, not just activity
Each milestone should produce something that helps someone make a decision. A reconciled report, a measured recovery exercise, or a small production workflow can be more informative than a percentage-complete estimate. These outcomes also expose assumptions while there is still time to adjust the approach.
A plan becomes useful when it tells you how to recognize progress.
I leave explicit room for discovery and keep the rollback path understandable. That does not remove uncertainty, but it makes the uncertainty discussable. A team that can explain its next decision is usually in a stronger position than one that can only repeat its final deadline.