While the concept and technology are not new, master data management (MDM) is experiencing a resurgence in demand which is expected to continue as companies seek out strategies to handle the burden of big data and information overload.
Master data management refers to practices and automated processes designed to create and maintain a common, single view of business entities--be it customers, products, locations or accounts--shared across systems, explained Daniel-Zoe Jimenez, program manager for enterprise applications and business analytics at IDC Asia-Pacific.
With MDM, organizations can ensure data quality and obtain a "single version of the truth"--the master data--which ultimately improves decision making, Jimenez said.
Andrew White, research vice president of supply chain management at Gartner, said: "[MDM is] to ensure the uniformity, accuracy, stewardship, semantic consistency and accountability of an enterprise's official and shared master data assets."
He pointed out that this concept of a "single version of truth" is not new but has been buried among other information management programs such as business intelligence (BI) which gained more prominence.
White added that MDM has always been "clearly and narrowly defined", focusing solely on master data. And today, it is at the center of much enterprise effort spanning big data, social media networking and cloud, he said.
Resurgence amid data bulge
Jimenez noted that MDM is gaining relevance and seeing "renewed attention" due to challenges faced by organizations today, including the onset of big data, the need to consolidate IT infrastructures to decrease system complexity, regulatory pressures, risk management, and the booming volume of audits and legal disputes.
"Customers, products, suppliers, and employees are all key assets of an organization. As a result, data and data ownership about such business entities must be maintained in a coherent and consistent way," he said.
For data-centric organizations seeking to reestablish control over data that needs to be shared across the company, MDM plays a critical role in providing the ability to map and track data assets, and reconcile data formerly constrained in silos, Jimenez said.
For example, he noted that for a company that sells a large number of products, failing to establish a methodical process to manage and maintain shared data across applications and across the company, could potentially leave brand managers unable to access accurate information about the performance of products and financial managers unable to measure profitability.
The more products the company adds to its portfolio, the worse this situation gets, he said.
Colin Tan, Asean regional manager of information management at IBM Software Group, noted that typically within a company, data is managed in multiple places, resulting in information being duplicated and therefore unreliable.
"Master data management helps high-value common data [provide] the single version of truth that businesses can leverage to derive business value and reduce risk," Tan explained. "And in today's era of big data, social data, Web 2.0 and cloud, MDM becomes more important."
Governance and data mess prevention
According to White, governing an organization's master data for reuse is critical for any business process that consumes data, inside and outside the company. In the case of big data, the analysis produced from understanding large amounts of complex and varied data needs to be semantically consistent with the data that comprises the business decision being adopted.
"If MDM is not sustained, nonsense will be made of the analytics of big data, meaning, big data will just become a big mess," he cautioned.
Similarly, in the case of social data, the sentiment analysis on such data also needs to be meaningful and must make sense to the user. Otherwise, they would not understand the data and make erroneous business decisions, he added. Ultimately, MDM is critical in enabling valid use of the data, he said.
IDC's Jimenez added that the technology is an "evolving discipline" that has expanded beyond the practices of rationalizing data elements across an organization, to incorporate or integrate with broader goals related to data governance and enterprise information lifecycle management.