An intro to Master Data Management for PerformancePoint Server
Much of the excitement, praise, and good publicity for PerformancePoint Server rises from its ability to be a comprehensive end-to-end business intelligence suite (mentioned here and here — and more from Gartner's opinion). If SQL Server BI offerings and BI applications such as PerformancePoint Server aren't enough, Microsoft purchased Stratature, a company that has focused on Master Data Management (MDM) software which will enhance the already great BI offerings from Microsoft. Roger Wolter mentions the impact the purchase will have on other BI applications such as Office SharePoint Server, SQL Server Integration Services, and others.
My post today provides a (very) brief description of MDM, the case for it, and how PerformancePoint Server will benefit from its implementation. More details can be found at Microsoft's MDM site. The MDM site also mentions the availability of a Technology Preview of Microsoft’s version of Stratature’s MDM in 2008, followed by a formal Technology Adoption Program.
What is Master Data?
I like the definition that is given in the white paper, The What, Why, and How of Master Data Management.
Master data are the critical nouns of a business and fall generally into four groupings: people, things, places, and concepts. Further categorizations within those groupings are called subject areas, domain areas, or entity types. For example, within people, there are customer, employee, and salesperson.
The white paper helps identify what data should be treated as master data with a list of criteria — you could call it a methodology. Maybe some of the methodology was inherited from Stratature. Some people might simplify the definition by making it analogous to reference data; data describing a physical or virtual object and its properties. Reference data are usually described with nouns. Typical reference data are:
- Physical: products, material, assets, customers, locations
- Virtual: cost centers, planned buildings
What is Master Data Management?
Again I refer to the white paper to define MDM. "We define Master Data Management (MDM) as the technology, tools, and processes required to create and maintain consistent and accurate lists of master data." The authors are also careful to point out, as is the Data Warehousing Institute, that the majority of organizations suffer from bad data for reasons that have nothing to do with technology. Causes include inaccurate reporting, disagreements over which data is appropriate, and incorrect definitions. Sometimes MDM becomes a company-wide effort and other times smaller efforts are made around key sources of Master Data until success justifies a stronger effort to include more of the organization’s data.
The white paper provides an abbreviated MDM project plan framework and methods for creating lists of master data.
The case for MDM?
A story is told in the article, Demystifying Master Data Management, of a manufacturing company that lost a key distribution plant to fire. To maintain relationships with customers, the CEO asked for customer lists from the company's CRM, ERP, billing, and logistics systems. Each application rendered a different list providing a different view of the customer. That is because most companies do not have a precise, unified view of their customers, products, and so on.
MDM resolves the mentioned complications by conforming the reference data such that inconsistencies are removed. This is illustrated by the following diagram.
Notice how reference data, such as Product, is viewed equally by each department. How does a company create a view of its customer, and other reference data, that is the same across organizations? According to another white paper, the Case for Master Data Management,
"The answer is to proactively manage master data—the report and analysis structures used by employees in every functional area—with a software solution called an MDM system. The MDM (master data management) system:
- Handles the necessary integration of transactional storage, and analytic systems
- Supports compliance with Sarbanes-Oxley [and other required compliance such as Basel II --see my Blog post about Basel II]
- Enables an auditable, repeatable process
- Captures and protects your company's human intelligence about report structures—this is knowledge which otherwise lacks a home."
My post only touches on the benefits of using an MDM system. To see more, go to the resources site of Stratature. Also, Stratature offers Webinars.
PerformancePoint will benefit from MDM
Finally, we get to how PerformancePoint Server products will benefit from an MDM product. Consider the cost of getting data integration right in order to profit from the BI tools that are now available such as PPS Planning and Monitoring. I remember wondering if I should mention the challenges (and best practices) for conforming data in the Planning data integration documentation, a step needed before loading data into the staging database that is created from the PerformancePoint Planning application. An MDM system will help simplify this large step by providing the prerequisite of clean, conformed master data (and all of the other benefits mentioned in this and other articles) prior to loading data to the PPS Planning application or using data for Monitoring and Analytics.
Consider the following image (from Stratature's article, The Human Factor) that helps you visualize the clean control of data an MDM system can help provide to PerformancePoint Server. There is much to look forward to from Microsoft and MDM.
[update] See also PerformancePoint and Master Data Management revisited
