Data migration is a complex process.
If you’ve conducted a data migration, or you’re about to do one, you can expect—at somewhere, some point in that project—to hear one of these nine different statements:
- “Profile the data? Why? It’s just going to slow us down.”
- “We’ll rely on the legacy systems data quality; IT cleansed it.”
- “Everyone knows the business rules; we don’t need to redefine them.”
- “Define the target-system purpose? Why? We already know what that is; we just need to upgrade or migrate the application.”
- “Data governance? Everyone works on it; it’s gotten us this far. Works just fine.” “Do that in extra coding?”
- “That function’s built into the ETL tool. It’s automatic.”
- “The business subject matter experts are too busy to work with us on this, on building out the business rules. IT will represent them.”
- “Data validation—we can do that at the end. We have the time.”
- “We can test with a subset of the data. We don’t need to load all the data until the very end.”
