PART I
Introduction to Master Data Management
CHAPTER 1 Overview of Master Data Management
CHAPTER 2 MDM: Overview of Market Drivers and Key Challenges
CHAPTER 3 MDM Applications by Industry
Our civilization has evolved into a modern society by continuously acquiring and developing new knowledge and creating innovative ways to improve personal and business conditions. This evolution in large part is based on our ability and ever-growing need to collect and understand information in order to run businesses; predict the weather; analyze market performance; manage personal finances; define medical diagnoses in order to prescribe proper medication, and, in general, to do both mundane and new, exciting things.
Over the course of history, we have collected a huge amount of data; learned how to interpret it and transform it into useful, meaningful information; and even created a number of extremely sophisticated information theories and branches of information science. One interesting observation about the way we collect and use data is our reluctance to discard data that is either old or no longer relevant. For example, you may have a collection of old professional books, and most of them have some value. You have to decide which books are redundant and need to be discarded and which books should still be retained because some chapters or sections are still valuable. You might also be thinking that it is a good idea to create a catalog of books with pointers to their locations. Otherwise, you will not be able to find the book you want promptly when you need it.
Other examples of data that we tend to collect and keep include items of family history and sentimental value, such as photographs, letters from parents and grandparents, and certificates of awards we received in school—things we want to keep for a variety of reasons, all of which make the data more valuable as time goes on. Of course, there are other types of information we may have to keep in order to comply with the law (such as income tax returns) or for personal protection (paid promissory notes or signed legal papers). Finally, some of us like to keep documents that are no longer valid and are replaced by newer, more accurate versions—for example, old resumes or outdated wills. Keeping this old “stuff” eventually becomes a storage problem, and this is when people move the boxes with old documents into their attics and basements. As the amount of data stored this way grows, the task of finding the right document stored somewhere in the attic becomes a challenge. This can become more than just an inconvenience: Storing outdated or inaccurate documents may have some interesting and even unpleasant consequences. For example, if there are two significantly different versions of a Last Will and Testament document, using the wrong version during the settlement may have drastically different consequences from those intended by the owner of the will. And, of course, data is different from other “stuff”—you cannot have a garage sale to get rid of data you no longer need!
The business world has certainly experienced this dramatic data-growth phenomenon as well. And this trend has become even more pronounced as we have entered the digital age and now have access not only to the ocean of data naturally created by or stored in the computer systems but also to data that previously existed only in a paper form but has been digitized to be managed by computer systems and applications. This data proliferation in the business world has not been driven only by traditional organizational structures and processes where various application areas and lines of business created and managed their own versions or views of the data; other business and technical factors contributed to this phenomenon as well:
• The Internet and the World Wide Web have made enterprise boundaries “porous” or “fuzzy” in order to attract large numbers of customers and to enhance the customers’ experience.
• The cost of disk storage dropped significantly in the 1990s. This prompted enterprises to collect increasingly vast amounts of data and develop complex data models with hundreds and thousands of entities at multiple levels of aggregation. Enterprises started storing ancillary data, providing much more context about the transaction or event.
• The number of entities and attributes managed by enterprises increased significantly. Enterprise architecture and data governance lagged behind and were poorly prepared for the tremendous increases in the volumes and complexity of the data. This has resulted in a significant accumulation of data issues over the decades.
In this new digital age, an agile enterprise can become competitive only when it has access to relevant, accurate, and complete information about business conditions and performance metrics, enterprise customers, prospects and partners, products and markets, and a myriad of other things. Given the ever-growing amount of data that is collected and managed by the business units of global enterprises today, we can draw an analogy similar to searching for old files in attics and basements. Enterprises have a hard time creating, finding, and managing data that is complete, accurate, relevant, and uniformly available to all businesses and users that need this data to run their businesses. In addition to the technical challenges of creating such a master data facility, there are several organizational and political obstacles to cleaning and rationalizing existing data stores. A common example would be the individual business unit’s desire to hold on to its version of data because it is deemed unique to the business unit’s goals, or because it helps eliminate dependencies of data management across business units. Some of these reasons are perfectly valid, but they do not eliminate the need to have an enterprise-wide, accurate, and complete view of the key entities and relationships of critical enterprise data.
The issues, approaches, concerns, and applications of Master Data Management and its strategic “sibling,” data governance, are the subject of this book.