Gartner Inc. has plenty of advice for both beginner and more advanced master data management (MDM) practitioners...
By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
that need to kick their MDM maturity levels up a notch.
Gartner analysts speaking earlier this month at the firm's annual MDM Summit in Los Angeles advised attendees to remember that all MDM programs should be driven by clearly defined and measurable business objectives. Beginners should research and closely adhere to widely accepted MDM best practices -- and closely track the business process changes that result from MDM initiatives.
A definition of master data management (MDM)
Gartner Inc. defines master data management (MDM) as a technology-enabled discipline in which the business and IT sides of an enterprise work together to ensure the uniformity, accuracy, stewardship, semantic consistency and accountability of shared master data sets.
Intermediate to advanced MDM practitioners should remember to reuse proven MDM-related data governance and data quality processes wherever possible as they expand their programs throughout the enterprise. However, the Stamford, Conn.-based IT research group cautioned that different departments within an enterprise may require different MDM tools. In other words, just one MDM software vendor may not be "good enough" for the entire enterprise.
"How do you make progress with MDM?" asked Dimitris Geragas, Gartner's senior director of consulting. "It depends on where you are. It depends on where you want to go. It depends on where you've been. It depends on a whole lot of things."
Transition to multi-domain MDM remains a challenge
Gartner reports that many companies are still struggling to expand their MDM programs beyond a single domain such as product or customer -- and MDM newbies Ganesh Prabhu and Paru Mahesh can certainly understand why.
Prabhu, an IT manager, and Mahesh, a data quality manager, attended the conference to investigate the possibility of launching an internal MDM program at their company, San Francisco-based Riverbed Technology Inc., which specializes in helping organizations optimize their IT infrastructures and wide area networks.
Prabhu, who represents the IT side of Riverbed, and Mahesh, who works for the business side of the company in sales operations, are currently in the process of making the business case for MDM to company higher-ups. If the initiative moves forward as planned, they will begin by focusing on the customer master data domain and later expand the program to cover product master data as well.
The two expect that the process of expanding to multi-domain MDM will be relatively difficult, mainly because of all the complex integration work involved.
"Integration between the transactional systems and the multi-domain MDM [will be a challenge]," Mahesh said. "How do you build those insights? And how do you model the data so you can grow those insights?"
Mahesh expects to overcome those integration concerns by planning for the expansion well in advance and working closely with team members to anticipate potential problems.
Prabhu added that he is confident the proposed initiative at Riverbed will gain approval because executives there seem to recognize the benefits of well-executed master data management, such as improved customer service and inventory management. Prabhu also had a tip for other MDM newbies that need to build a business case.
"One key thing is to find the real operational efficiencies that you will get out of this project, and some of the key metrics that will be available once the solution is in place," he said. "Once you know the key benefits the business will have, it becomes a much simpler business case to make to executives."
More tips for improving MDM maturity levels
MDM is often thought of as an expansive and complex discipline, but that kind of thinking is largely counterproductive, according to Geragas. It is better, Geragas said, to think of MDM as a simple program -- or a series of incremental programs that speak directly to the business value they provide.
"The path of improving your maturity is about killing MDM as a discipline and starting MDM as a program," the analyst explained. "MDM as a discipline only holds the promise of value, but the way you unlock this value is by deploying the discipline as part of the program. A discipline cares about doing anything and everything. A program only cares about whether this is something that makes sense to the business."
A discipline has to be monolithic, whereas a program is incremental and involves progressing gradually through phases. The management of a discipline is very big and cumbersome, while the management of a program entails specific tasks.
A company that looks at MDM as a discipline might say, "We are going to improve data quality," Geragas continued. But a company that looks at MDM as a program would be more specific and say, "We are going to de-duplicate our customer data by 60%."
"A program does not do everything under the sun. A program does just enough to satisfy the business," he said. "A program doesn’t talk about anything and everything that could possibly happen. It talks about just the things that will happen."
For more on the Gartner MDM Summit
Get Gartner’s take on the connection between MDM and “big data”
Find out how Sony Pictures advanced to the next level of MDM maturity
Learn why office politics and MDM can be a dangerous mix
Gartner reveals five levels of MDM maturity
Organizations that want to figure out just where they currently stand -- and where they need to go from there -- can refer to Gartner's five levels of MDM maturity.
According to Geragas, organizations reach the first level of MDM maturity when they finally realize that something needs to be done to improve the quality of master data and other forms of information.
Level two rears its head when individual departments within organizations begin to get fed up with the inefficiencies caused by inconsistent data, and begin grassroots efforts to start fixing the problem. The danger at this level is the tendency to address only the symptoms of inconsistent data as opposed to the root cause.
At level three, organizations start to see "top down" support for MDM as company bigwigs become more concerned. Level four begins when an organization launches an enterprise- wide MDM program. At level five, enterprise MDM is up and running and supporting the business nicely.
"Funny things happen when you reach level five," Geragas said. "One is that MDM will go away as a name because it is so ingrained in the business that people will not think there is a special name for it. People will think it is just business."