The idiom of eating the elephant one bite at time is often used when trying to vision a roadmap for Master Data Management (MDM).
It’s a bit of a contradiction to look at it that way, because the essence of MDM is an enterprise wide single source of truth eventually for all master data domains.
But it may be the only way.
Using a cliché MDM is (as any discipline) about people, processes and technology.
In an earlier post called Lean MDM a data quality and entity resolution technology focused approach to start consuming the elephant was described, here starting with building universal data models for party master data and rationalizing the data within a short frame of time.
I have often encountered that many organizations actually don’t want an entity revolution but are more comfortable with having entity evolution when it comes to entity resolution as examined the post Entity Revolution vs Entity Evolution.
The term “Evolutionary MDM” is used by the MDM vendor Semarchy as seen on this page here called What is Evolutionary MDM?
The idea is to have technology that supports an evolutionary way of implementing MDM. This is in my eyes very important, as people, processes and technology may be prioritized in the said order, but shouldn’t be handled in a serial matter that reveals the opportunities and restrictions related to technology at a very late stage in implementing MDM.
Liliendahl on Data Quality 