7 Master Data Management Project Best PracticesMobile, the cloud, and the constant flow of information make managing the 'building blocks of content' within the enterprise more challenging than ever. Here's how to stay in control of your data.
Master Data Management (MDM) is the process of establishing and implementing polices, standards and tools for administering data that's most essential to an enterprise, including information on customers, employees, products and suppliers.
"When we talk about master data, in my mind we're talking about structured data -- the building blocks of content in any organization," said Arvind Singh, co-founder and CEO of Utopia, a Chicago-based enterprise data solutions provider. In a phone interview with InformationWeek, Singh said that unstructured data -- or information arriving from multiple sources at very high volumes and speeds -- is impacted by MDM practices as well.
- The Five Must-Haves of Big Data Storage
- Banking on Results: Turn an Avalanche of Data into Actionable Insight
- Big Data Analytics: Profiling the Use of Analytical Platforms in User Organizations
- Big Data: Harnessing a Game-Changing Asset
- Take the InformationWeek 2013 Database Technology Survey
- Security Implications of Big Data Strategies
"Unstructured data has to be tagged so that you can make sense of it. And you start connecting the dots between the unrelated pieces of unstructured and structured data," said Singh. He believes that organizations must first "get their arms around master data governance and management" before tackling unstructured data.
[ For more on MDM, see How To Manage Big Data. ]
Singh proposed the following 7 best practices for master data management:
1. State the problem you're trying to address. Every MDM project has to start with this question: What is the business problem we're trying to solve?
This problem, naturally, will vary based on the organization and its needs. The following examples, said Singh, are typical: "I'm not able to make the right inventory decision; I'm not able to make the right hiring decisions; I'm not able to forecast my revenue properly."
2. Determine the project's mission and business value, and link the initiative to actionable insights. It's important to clearly state the project's return on investment (ROI). This is essential not only for getting the product funded, but also for convincing the front office to buy into it. "The reason a lot of master data projects fail is because the justification, the ROI, is not necessarily linked to the business value," said Singh. Fixing bad data "because it's the right thing to do" isn't good enough.
3. Understand how the project helps your organization prepare for big data. If you manage structured data efficiently, you'll be in a better position to deal with big data as it arrives. "Structured master data is the most fundamental of all content in an organization," said Singh. "If you don't address fundamental issues around the foundational data of today, you're not going to be able to scale up to big data, which obviously has both structured and unstructured data, and at very high volumes."
4. Devise a good IT strategy. An IT strategy for your master data management initiative needs to be sensitive to two things. First, it must be sustainable. "In other words, the technology platform needs to be reusable. It has to be scalable," Singh said.
Second, real-time governance is essential. The IT tools must not only be capable of doing the usual cleansing, validation, integration, and enrichment efforts, they also must work in real time and be available for future use.
5. Business users must take full ownership of the master data initiative. Singh called this tip one of the most important of his 7 best practices. "In my mind, there is almost no doubt that the ownership of master data lies squarely with the business user," he said. Of course, the IT group is essential in providing the project's technology, support, platforms, and infrastructure. But the ownership of business rules and standards lie with the business user.
Why? "With all the analytics and reporting and dashboards, business users are now directly engaged in the process of information management," said Singh. He added, "Many years ago, IT produced and generated the reports. Today you have business users modifying, customizing, and creating reports in real time on their devices. That's a fundamental shift."
6. Pay attention to organizational governance and change management. You must have a very strong governance model that addresses issues such as change management and knowledge transfer. "Recognizing and addressing the culture in an organization is extremely important, and the governance framework fosters communication between the business users and the various parties involved, including IT," said Singh.
7. Choose the best technology platform. IT folks have a lot of options when it comes to architecture and software choices. And there are plenty of pros and cons concerning standardized versus best-of-breed software. "This becomes pretty significant when you start talking about what kind of technology platform to put in place to manage your master data," Singh said.
Today, master data management faces additional challenges because data is generated at all times and in a variety of places. "In the past, you were creating master data in a very structured environment, generally inside the office," Singh said. "Today, you're also creating data through mobile devices, or on the cloud."
InformationWeek is conducting our annual Outlook Survey to explore how IT leaders are planning their priorities and budgets for 2013. The results of the survey will appear in an upcoming issue as well as in an in-depth report. Take our InformationWeek 2013 Outlook Survey now and enter to win one Samsung Series 5 Chromebook. Survey ends Nov. 19.