Mastering Data Management

More in Business Process Management

• New ebook: Understanding Big Data
• 37 Definitions of “Employee ID”
• Defining Terms: Agree, or Doom your Project
• Establishing a Valid Business “Need to Know” for Sensitive Data

More in MDM

• Best Practices in Test Data Management: Discovering and Subsetting
• TDWI Chicago Preview: Data Warehousing, Analytics & More
• Who’s Afraid of the Big Bad Data?
• MDM for Chief Privacy Officers: A healthcare provider example

More from Lawrence Dubov

• MDM Vs Data Governance: History and 8 Predictions
• What’s Harder: Starting or Sustaining MDM?
• Conclusion: Quantifying Assets
• 3 Essential Questions for Data Quality: Benchmarks, Penalties & Latency

Share this Post

Recent Comments

• ICMCC News Page » Where East Meets West: ehealth in China and the US on Where East Meets West: ehealth in China and the US
• Kim’s Training Podcast on Training as a Safety Net: Kim May
• TDWI Chicago Preview: Data Warehousing, Analytics & More | Mastering Data Management on New ebook: Understanding Big Data

Initiate's Larry Dubov explains why understanding relationships and hierarchies among your data is essential

The next stop on our MDM Roadmap involves examining two very important domains of MDM: relationships and hierarchies, and third-party data sources. When moving beyond a single domain, it becomes increasingly important to address these very common scenarios.

Relationships & Hierarchies

The value of MDM greatly depends on what the program does to manage relationships between master entities. For a more complete discussion of this topic, try these posts:

• Larry Dubov, Master Data Management: Importance of Relationship Management
• Adam Isler, MDM and Relationship Linking
• David Linthicum, The Importance of Relationships Management with MDM

Any mature MDM solution includes strong support for relationship management. A holistic and panoramic view of a customer includes a comprehensive view of relationships. As we discussed in the previous section, various enterprise functions require different views and definitions of entities. The same applies to relationships.

Understanding relationships between parties and products, as well as product hierarchies, is critical for enterprises. It is quite common for an MDM program to begin with one or a few master entities and then to evolve in a multi-domain MDM with a high focus on relationships.

At a high level, there are three techniques used to build entity relationships and hierarchies:

• Use an external trusted source of relationships (or multiple trusted sources) and build relationships and hierarchies by comparing in-house master data with external trusted sources
• Define the rules that will be used to algorithmically infer relationships. These rules can be based on common attribute values, (e.g. people sharing the same home address are defined as a household). The attributes can be matched exactly or probabilistically by applying the same probabilistic and fuzzy methods that were used earlier for entity resolution.
• Link relationships manually by using graphical interfaces. This method requires additional data stewardship resources and therefore additional costs associated with data stewardship

Each method has its inherent benefits, areas of application and limitations. In full-scale MDM implementations, enterprises can benefit from using all three methods to create and maintain a comprehensive view of relationships.

The use of each of these methods can, on the one hand, increase the value of the MDM solution while increasing the MDM implementation cost

Third Party Data Sources

An MDM solution can greatly benefit from the use of third party data. A data hub matches enterprise master data with the third party data, which helps to improve the quality of the match and enhance entity and relationship resolution.

Third party data also helps in the development of customer and product hierarchies. External trusted sources can improve data quality and enrich enterprise master data.

There are, of course, costs associated with third party data. This includes the costs of data subscriptions and the costs of integration with trusted sources. Specifics of the third party source or sources, their usage and integration modes will impact the total cost. But regardless, third party sources are extraordinarily important as they can contain information essential to doing business.

(My colleague Ian Stahl has been writing about how a shipping company is leveraging this third party data.)

Next week, we’ll discuss decommissioning systems and applications and its impact on your MDM roadmap.

This is part of a series, Building an MDM Roadmap. For other posts and a complete index, view the Table of Contents.