First, they assume the careful method is the slow one. It is the fast one. O’Hare went from selection to go-live in about four months, at the busiest airport on earth, with no operational disruption, precisely because nobody bet the operation on a single cutover weekend. Domain by domain, validated, reversible. Nothing had to be redone, so nothing slowed it down.
Second, they buy from whoever has the most consultants in the room. Wrong instinct. A company that sells manpower needs your migration to stay complicated. A company that builds products needs it to end. Ask which one is pitching you.
Replace the AODB. Do it the careful way. Buy it from people who build products, not people who bill hours.
The full method is here: The Great Migration: how to replace your legacy AODB without operational risk.
Two questions worth asking any vendor pitching you an AODB replacement:
- At any point in the migration, can you tell me exactly which data domains the new system controls, and exactly how to undo the last one?
- When this is finished, do my own teams run it, or do yours?
If the answer to the first is “we’ll be in a war room that weekend”, and the answer to the second is “ours, indefinitely”, you are not buying a migration.