In a recent Forrester Muse post, Michelle Goetz, talks about a common cause of tension between IT and business when looking at master data management.
For IT, MDM is often considered to be just another technology component, and the approach is defined around a technical goal, such as Single View of the Truth.
This approach may be supported by business consultants, who are paid millions to build a business case to justify the technical goal. In my experience, these business cases, which typically focus on the basic technical architecture required to support the IT goal, fail to meet business needs.
Master Data Management is a massively political initiative. The principle objective should be to achieve reuse of core data so as to reduce the risk of error and, possibly, to reduce the operational cost of maintaining core data by reducing the number of points of capture.
Achieving reuse of data for multiple purposes means that these different purposes need to be understood and that conflicts of purpose need to be managed - the principles of lean data governance. It is no surprise that data governance is listed as a critical success factor for MDM.
MDM will also meet political resistance from IT. Any project which aims to achieve reuse of information is going to have an impact on the budgets and objectives of multiple projects that had previously incorporated elements of the MDM requirement into their charters. MDM may divert budget from these tactical projects, as the cost savings of reuse come into play. This is sure to drive resistance from project owners. MDM may also have different time lines to these tactical projects, leading to the valid concern that tactical goals may not be meet timeously.
Business are unlikely to buy in to a technically driven business case and approach, and, if they do it is not likely to succeed. The business case for MDM needs to be built based on an alignment to the strategic business goals - which typically require a multi-domain, interactive approach that focuses on key business processes. For example, if the business goal is to "reduce time between order and deliver" then the solution will have to support elements of Customer, Product and possibly even Supplier and Sales. To support additional business goals, additional elements or domains will have to be brought into play.
A sensible approach may also look at building a full MDM capability in a series of steps, by addressing key requirements early and adding to these. So, an initial business case could be based on implementing the lean data governance required to accurately define the business requirement, while the next phase would be to analyse and improve data quality in support of meeting this business requirement.
Achieving these objectives helps to increase the maturity of the organisation to the point where an MDM project is likely to succeed, as well as providing a solid technical and process infrastructure that can be reused within the larger MDM environment