Untitled Document
Editor's note: Missed our SOA in Action Virtual Conference? Replay
sessions on demand here.
Over the past two years, the demand for master data management (MDM) has remained
strong, despite the economic downturn. This isn't surprising, since an effective
MDM implementation is one of the few IT initiatives companies can pursue to
realize near-immediate business process improvements across many different areas
within the enterprise.
However, the question confronting IT and business decision-makers is which
MDM solution and methodology will work best to resolve their master data challenges?
One possible route for many large organizations with extensive SAP implementations
is to make their SAP enterprise resource planning (ERP) application suite the
focus of their MDM initiatives.
It seems like an enticing proposition, given the possibility of capitalizing
on existing IT infrastructure, investment, institutional knowledge and so forth.
Yet SAP ERP systems are not designed to support master data management, and
are simply not the right place to master data.
Why not master your data in SAP ERP?
Clearly, many companies across multiple industries can benefit from MDM. But
the question is how can they implement an MDM solution to achieve both long
term value and advance their business initiatives? As we've seen in multiple
MDM implementations over the last few years, successful MDM projects hinge on
providing top-notch capabilities in four primary areas:
- Data modeling
- Data cleansing, matching and enrichment
- Consolidating data from multiple sources
- Managing complex corporate or product hierarchies
In a typical MDM implementation, data stewards should be able to automatically
merge duplicate records, and handle survivorship of data attributes into a single
"golden record." The hub should have integrated data quality capabilities
so that data cleansing continues after the implementation.
It's also critical that the hub vendor be able to provide multiple proof points,
demonstrating that its product can handle more than one area of data-not just
customer or product data, for instance. This kind of multiple data type capability
is crucial for achieving expected ROI, as expanding the use of the hub to solve
business problems beyond the scope of the original implementation area is almost
always required.
-1-