The riskiest sentence in enterprise IT is 'we'll rewrite it from scratch.' The graveyard of big-bang rewrites is full of projects that were 80% done for three years while the business froze around them.
Grow around, not replace
The strangler pattern inverts the rewrite: the new platform starts as a thin layer around the legacy core, absorbing one capability at a time — reporting first, then a workflow, then a module — each behind an interface the rest of the estate already understands. The legacy system doesn't get switched off; it gets hollowed out until switching it off is an anticlimax.
Parallel-run or it didn't happen
Before any capability cuts over, both systems run it simultaneously and the outputs get reconciled — every invoice, every balance, every edge case the old system handled in ways nobody documented. We've had month-end close on both systems during transitions. That's not caution theatre; that's how you find the undocumented behaviour before it finds you.
Knowledge rescue
The scariest legacy risk isn't the code — it's the retirements. The people who know why the system does the strange thing on the 4th of April are leaving. Modernization is your last structured chance to capture that knowledge into tests and documentation. Treat it as such.