The migration will take six months. This is what the whiteboard says.
Boxes and arrows. Dependencies identified. Effort estimated. Leadership approves the timeline and allocates the engineers.
Eighteen months later, the migration is still running. The old system is still in production. The new system is partially deployed. Both require maintenance. The team is exhausted.
This is not a planning failure. It is a structural inevitability.
⸻
The Parallel Tax
Every migration creates a period where two systems coexist. This period was designed to be temporary.
But temporary has no enforcement mechanism. The moment both systems are running, every new feature must be evaluated against both. Every bug must be investigated in both. Every deployment must account for both.
The migration did not reduce complexity. It doubled it. The team that was supposed to be building the future is now maintaining the present and the past simultaneously.
⸻
The Last 10 Percent
The first 80 percent of the migration moves fast. The common cases transfer cleanly. The dashboards look like progress.
Then the edge cases arrive. The obscure integration that one enterprise client depends on. The data format that was never documented. The behavior that exists because of a bug that three downstream systems now treat as a feature.
This last 10 percent takes longer than the first 80. It is where the migration stalls, because the cost of completing it exceeds the tolerance of leadership for a project that was supposed to be "done by now."
⸻
The Zombie State
When leadership loses patience, the migration enters a zombie state. It is not cancelled, because cancellation requires admitting the investment was wasted. It is not completed, because completion requires more investment than anyone is willing to approve.
Instead, it lingers. The old system stays in production with a thin maintenance team. The new system runs for new customers only. The organization now permanently operates two platforms, paying the parallel tax indefinitely.
⸻
Before You Start
The migration will cost more than the estimate. It will take longer than the timeline. And the organization will lose interest before it is finished.
If you cannot secure a hard cutover date with executive commitment, do not begin. A half-finished migration is more expensive than the legacy system it was supposed to replace.
End.