Master Data Management (MDM)

March 1, 2012

The battle of Data

After getting dozens of credit card offers in the mail from a premier bank, I decided to get their credit card. I got the card and started using it. But, for the next 4 months, the mails kept coming asking me to sign up for the same thing! Apparently, the news hadn’t reached their product marketing department.

A few years ago, I worked for a company that had recently accomplished aggressive inorganic growth. They were very successful in increasing their customer base as well as product portfolio by acquiring competitors. But, in these acquisitions and mergers, the right level of system integration was not always achieved – primarily because high immediate costs and low short term ROI made full IT/Data integration a difficult case to sell to the shareholders. In subsequent years, IT maintenance costs became very high, as the company started facing data quality, consistency, and accuracy issues as well as operational inefficiencies. Numerous trouble tickets were being opened every day due to data problems and excessive money was being spent on maintaining and supporting the intricately interconnected systems.

The purpose of these two examples is to indicate how today’s organizations are fighting the battle with data and sometimes losing it. Now, let’s discuss what MDM is and how it can help resolve these problems.

MDM definition(s)

Master data management (MDM) is a data management discipline to actively “manage” data across the enterprise, rather than “maintaining” it in each transactional system. MDM is also described as a combination of technology, tools, and processes that enable the creation and maintenance of consistent and accurate ‘master data’. While other data management approaches attempt to improve data quality under various separate initiatives/projects, MDM seeks to improve data quality at an enterprise level by taking following steps –

(a) Improve data quality at both ‘the source of data entry’ and also during ‘data exchange among IT applications’ by instituting various data governance procedures and policies and implementing data quality monitoring tools

(b) Cause an enterprise wide behavior change where data integrity and quality is considered highly critical to business success

(c) Provide a ‘well documented centralized master database’ that can accommodate the differences from various disparate existing data sources and feed the master representations back into those different representations

(d) Provide ‘standardized software components’ to handle unique identification methods that eliminates the need for each application developer to design the process at an application level

(e) Provide various kinds of ‘master data services’ at the core and application layers that provide a means for unifying the enterprise application architecture, thereby freeing the developers to focus on application specific functionality.

Although, the selection of master data entities depends somewhat on the nature of an organization, ‘customer’ and ‘product’ are included in most MDM implementations.

The Business Case for MDM

The recent emphasis on regulatory compliance, SOA, and mergers and acquisitions has made the creating and maintaining of accurate and complete master data a business imperative. But, integrating master data is usually perceived as an IT-owned issue, and business stakeholders may be reluctant to engage in these initiatives. Hence, the ‘Why’ and ‘How’ of MDM are questions that many CIOs are trying to answer to make a business case for implementing enterprise wide MDM. Today, this trend is particularly visible in retail, finance, insurance and telecom industries that have large amount of customer data and smart business decisions greatly depend upon availability of ‘intelligent information’. For example, fragmented and inconsistent data affects cross-sell/up-sell ability, delays time to market, creates supply chain inefficiencies and weakens market penetration.

Who has the answers?

MDM is a relatively new and evolving field. IBM and Oracle rolled out their MDM offerings in 2008. Informatica entered the MDM space in 2010 with the acquisition of Siperian. Currently, IBM Infosphere, Oracle MDM Product hub, and Informatica MDM (aka Siperian Hub) are three of the main MDM solutions available in the marketplace. These companies are competing to win MDM contracts on the premise that implementation of MDM will help organizations create robust, reliable, scalable systems for the future and maintain enterprise data more efficiently, thereby reducing the ‘lights-on’ costs for data maintenance.

MDM risks and challenges

Given the limited availability of disposable funds and outside investment, it is imperative to make sure that the planning behind an MDM program is flawless. If a company’s data quality isn’t in top shape with solid data governance in place, MDM will not succeed. A common mistake is for companies to think that MDM is a quick-fix for all data issues, and a panacea for data quality problems. But, it is important not to forget that sloppily-executed MDM deployments can waste money and precious work time.


As data continues to grow rapidly and information intelligence continues to play in increasingly important role in driving business decisions, MDM is a subject area that is bound to get more attention in coming years. However, MDM has yet to mature into a standard framework, but if it does, it has the potential to unify the data management industry in a way similar to what Service Oriented Architecture did a few years ago for the IT industry.

By: Megha Goel


“Master Data Management” (The MK/OMG Press) by David Loshin


One Response to “Master Data Management (MDM)”

  1. Siddharth Sehgal said

    Great Article Megha, I had similar problems with my internet provider. I think MDM will continue to grow in the coming years.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: