Data Management Maturity Model takes things up a level
A Carnegie Mellon University off-shoot applies maturity model techniques to big data. Cross-group communications are a key. The result is the Data Management Maturity Model.
The CMMI Institute for software engineering last week released its Data Management Maturity Model, in an effort to improve data management practices in the face of onrushing waves of big data. Part of Carnegie Innovations, the institute is associated with Carnegie Mellon University.
"Everything is based on data today. We make so many decisions based on data, the integrity, governance and quality of that data needs to be better assured," said Lynn Penn, director of enterprise integration at Lockheed Martin Corp. She said she and her colleagues have begun to use the Data Management Maturity (DMM) Model as part of development programs at Lockheed-Martin.
With early adopters among such organizations as Lockheed-Martin, Fannie Mae, Freddie Mac and the Federal Reserve System, the new framework is meant to help teams align data management strategy with business goals of companies and institutions.
Penn said the DMM helps users create a common terminology for describing data and the framework focuses on data management at strategic as well as planning levels.
Like earlier noted CMMI Institute work, the DMM model sets guides for achieving different capability levels based on measured improvements in specific process areas of data management. The CMMI Institute worked with the Enterprise Data Management Council and other organizations during the DMM development process.
In a way, the new work is intended to bring some of the design rigor already familiar to many through the group's well established Capability Maturity Model Integration, or CMMI. It sets guides for building repeatable software engineering practices within organizations, and has been in use for more than 20 years.
The key to CMMI is a series of levels of process improvements that in turn represent levels of IT maturity. Users work through a process in which they evaluate current data management capabilities, uncover gaps and highlight strengths. The method helps leaders establish priorities and create a roadmap for process improvements.
While there are similarities between CMMI and DMM, there are differences as well. The original CMMI takes somewhat of an agnostic view on data issues, which DMM more directly addresses.
In the big tent of data
"This is CMMI for data management, we have applied the principle and the architecture and structure and level-gradated capabilities that can make a built-in path to improvement," said Melanie Mecca, program director for DMM at CMMI Institute.
Mecca agrees that some of the software architecture themes of DMM have been seen before. Data management experts have long said that projects stumble when the business side of the organization is not adequately involved in the data strategy.
But DMM looks to better enable this cross department communications, by outlining practical steps users can take to achieve that goal, she said. A balanced perspective is part of the process.
"It's about providing managers with guidance -- ways to bring about a balance between 'the forest and trees' for the business professionals, data management professionals and IT providers," Mecca said. "We do it as a big tent approach -- we put all the participants together."
Penn agrees, while emphasizing that DMM is about guides rather than rules. Users need to apply the correct effort too, she said.
"This isn't a hall pass or guarantee that everything is ok because of DMM," she said. "But the probability of success is increased by employing the [CMM] models as best practices."
Check out this video - A senior BI manager on DMM
Learn about CMMI and agile methods
Discover more data management best practices