This question came up today while visiting a customer and it’s a question that is been on the mind of many end users. As they consider some ERP project, perhaps a migration, or some upgrade, they wonder, “Should I concern myself with MDM before I do this project, or should I worry about MDM afterward?”
The answer is simple, but there is a level of complexity that is implied by the question that needs to be explored first. I posed what might be a simple question – MDM before or after ERP – but the real issue is what is meant by “ERP”? For some firms ERP can be a business application strategy (build the business applications on one common, sharable data and process model) or it can be a sourcing strategy (acquire as much technology from one vendor), or it can be both.
But for some users ERP is also a “suite” strategy, in that the vendor they work most closely with offers several application suites that are not of the same data and/or process model. This is not a single vanilla ERP offering, but since it is offered by a single vendor, it is often called an ERP strategy by the user – and the vendor!
So the question really needs qualifying: given that a firm is moving from a complex heterogeneous environment to a simpler but still reasonably heterogeneous environment, toward fewer data and/or process models, should MDM be of value – before or after? The simple answer is yes!
The idea is this: in any migration effort there will be a data clean-up process. This data clean-up process can be quite complex, from identifying data sources, to mapping and merging data, to normalization and enrichment – all before it is populated into the new systems. Much of the technology used in this work is what would be used with any MDM effort. So from a technical view point “doing” MDM before hand, that is, achieving single view of master data would be good preparatory work for any large scale system migration.
Where MDM adds much more value is in the ongoing process, the discipline side of MDM. The migration of any number of business applications to some newer, even fewer business applications, does not necessarily mean that the business will adopt new, updated, and effective governance routines for creating and sustaining master data. But if an MDM routine was adopted as part of the design of the large system migration, a new process would be established that can persist after the migration. In other words adopting the discipline of MDM before ERP, or as part of the migration activity, means that post migration the business oriented routines would be established such that master data consistency will be assured – ongoing. Adopting MDM before ERP can help simplify an ERP initiative or any other large scale system migration.
Any thoughts to the contrary? Any ERP’ers want to argue “for” ERP over MDM?