Understanding Master Data Management

The objective of MDM is to provide and maintain a consistent view of an organisations core business data; this may involve data that is scattered across a range of applications. Typical examples include Customer, Product, and Supplier. It is in reality the reincarnation of the problem of how to manage the consistency and integrity of the myriads of data assets that exist across the enterprise. Systems such as ERP, SCM, CRM, BPM, ODS, data warehouses/marts, legacy applications, ECM (unstructured content like emails, documents, etc), portals and various home-grown applications. Each of these applications generates and works on a data set which partially or completely overlaps with the others. The same Customer, Product etc is often represented in two or more applications, sometimes differently. The objective of MDM can be therefore summarised as providing a consistent view of an organisations dispersed data and its associated definitions.

It is a natural evolution that the enterprise data model has become a bedrock of companies MDM strategies. To implement a MDM strategy successfully a company needs to understand all its enterprise level data as master data is enterprise by definition.

The enterprise data model is the foundation that MDM is built upon.

Master data is often one of the key assets of a company. It's not unusual for a company to be acquired primarily for access to its customer master data. There are also a number of companies that have created highly lucrative business models based on selling data about specific marketplaces (examples include Reuters, Bloomberg and IMS Health).

The Silk Road Story

To explain the MDM challenge in non IT terms I have included below the story of ‘The Silk Road’ as it provides an approximation of the issue but set within a historical context.

The Silk Road is an extensive interconnected network of trade routes across the Asian continent that connects Asia with the Mediterranean and Africa. The description of this route as the `Silk Road' is somewhat misleading. Firstly, although the word road implies a continuous journey, very few travellers traversed the route from end to end. Typically, goods passed through a series of agents. Secondly, no single route was taken; crossing Central Asia several different branches developed. Extending for over 4,000 miles this route was not only a conduit for silk, but also for many other products.

The route was extremely treacherous. The Chinese monk, Faxian, gives us an inkling from the accounts of his travels along the road at the end of the fourth century:

"The only road-signs are the skeletons of the dead. Wherever they lie, there
lies the road to India."

Faxian, 399 to 414 BC

As has been mentioned silk wasn’t the only precious commodity traded. Caravans heading towards China carried gold and other precious metals, ivory, precious stones, and glass, which were not manufactured in China at that time. In the opposite direction furs, ceramics, jade, bronze objects, lacquer and iron were traded. Many of these goods were bartered for others along the way, and objects often changed hands many times. There are no records of Roman traders being seen in China, nor Chinese merchants in Rome. In essence silk passed from China to Rome passing through many agents meaning that:

• The Chinese never knew where the silk went
• The Romans never knew where it came from

This is the crux of the MDM problem:

• The Data Producers never know who is using their data
• The Data Consumers never know where their data comes from

The skeletons of the dead mentioned in the quote from Faxian can relate, in our context, to failed projects that didn’t consider the enterprise nature of data.

The above blog is drawn directly from my new book: 'The Enterprise Data Model'.