How Master Data Management Can Help Tame the Data Governance Mayhem
While it is widely accepted that data is driving the digital workplace, there is a problem. Enterprise data lives in a range of silos that makes it just about impossible to get a single view of the truth. While this clearly causes major problems for sales and customer management, an often-overlooked problem with this is its implications for data governance and compliance.
What Is Master Data Management?
There have been many attempts to tackle this problem over the years, one of the most notable of which is master data management (MDM). Gartner defines MDM as a technology-enabled business discipline in which business and IT work together to ensure the uniformity, accuracy, stewardship, governance, semantic consistency, and accountability of an enterprise’s official shared master data assets. In its Magic Quadrant for Enterprise MDM (Subscription Required) of last January, Gartner outlines several characteristics that an MDM platform must offer. They include the following:
- Support the global identification, linking and synchronization of master data across heterogeneous data sources through reconciliation
- Create and manage a central, persisted system or index of record for master data
- Enable generation and delivery of a trusted version of one or more data domains to all stakeholders, in support of various business initiatives
- Support ongoing master data stewardship and governance requirements through workflow-based monitoring and corrective-action techniques
In sum, MDM is a technology-enabled business discipline, in which business and IT work together to ensure the uniformity, accuracy, stewardship, governance, semantic consistency and accountability of an enterprise’s official shared master data assets.
Related Article: Is a Single Source of Data the Way Forward for Data Governance
MDM as a Practice
However, MDM is more than just technology. Darko Ivanesko founder and CEO of ESTAForm an online travel authorization platform that manages huge amounts of sensitive personal data, argues that master data management is a practice more than a tool. MDM is meant to maintain a single point of reference for every instance of company data. Without accurate data management, discrepancies in data will appear which cause delays and run-on costs. This could mean incorrect billing dates on your invoicing software, mistakes in stock records, or even setting up the wrong meeting date. This is why MDM is dependent on every department.
“Using MDM principles helps develop data management strategies. Because it is a broad-spectrum practice, MDM builds a multi-disciplinary approach,” he said. “People can be assigned responsibility for certain data (Data Stewards and Owners) to increase accountability. Data managers must determine standards for how information is captured and where it is necessary for similar sets of data to be categorized."
There are also several tools managers have access to that automate data management. These tools can track record sources, centralize data, and use smart AI to standardize data sets. All these tools are useful and certainly necessary for large corporations, but MDM is possible to maintain without them.
There are other advantages too, Patty McDonald, global solution marketing director at Addison, TX-based Symphony RetailAI, added. Disparate systems cause inefficiency and create inconsistencies, making it difficult to compete in an omnichannel world, Master Data Management, on the other hand, gives visibility to data, via a single source, that enables accurate, agile and confident actions for all business users. "In fact, MDM has been found to lead to a 10x increase in productivity," said McDonald.
With retail as an example, a business can streamline data across an organization and govern all channels with real-time visibility — including data like products, prices, locations, vendors and customers. Collaboration can also be improved outside the organization, across suppliers and partners.
Logical Source of Data
Is it possible to create a single source of truth? Dave Mariani is one of the co-founders of AtScale and currently its Chief Technology Officer. He points out that creating a single physical source of data is not realistic or feasible but creating a single logical source of data is.
Many organizations, he said, are piping data into centralized cloud data warehouses today and that is a good thing to do. However, data will always exist in silos outside of the data warehouse or even the data lake. For example, most organizations use a variety of SaaS applications which all lock customer data away behind proprietary APIs.
On top of that, there is a host of data that is stored in operational databases that might not make it into the data warehouse. All this means that hosting data in a single physical location is not feasible. That said, data virtualization powered by a business-friendly semantic layer can deliver on the “single source of truth” promise.
Jun 22 11:00 AM PST
How to Modernize Your Intranet and Avoid the Build or Buy Headache
Join Workgrid’s Rob Ryan and Frank Pathyil to discuss the challenges in building or buying an intranet.RegisterON DEMAND
Making the Complex Possible: How to Accelerate Your Digital Transformation
Hear how leading companies are reimagining their digital transformation projects and identifying new opportunities for growth.Watch Now
The Race for Digital Transformation: Employee-Centric IT Against the Odds (with Paddy Power Betfair)
In this webinar, we’ll explore real-world use cases that illustrate the transformational benefits of employee-centric IT.Watch NowON DEMAND
How to Use Space Reservation Tools to Return to the Office Safely
Explore the innovative tools that help make the transition back to hybrid and in-office work seamless.Watch Now
“Data virtualization works by exposing a logical view of data while leaving the physical data in the respective data stores,” he said. “By avoiding the physical movement of data, organizations can deliver a consistent, governed view of data without needing to re-architect their data platforms or invest in costly and time-consuming data engineering projects that can never keep up with the growth of data or the explosion of new data sources.
Single Source Of Truth
The importance of finding this single source of business “truth” cannot be overestimated, according to Mike Johnson, Senior Manager at Parkavery said. “It is important for all areas of the organization have a common language — in this case, master data. Master data serves as a foundation that allows the entire organization to understand the key components of their business and develop strategies based on the same information,” he said.
To maintain control of master data, there are a few key elements that ensure the desired consistency is achieved. First, it is important that master data is owned by a single, neutral governance group.
Second, the governance group responsible for master data should have representation from all functional areas within the business to ensure common understanding and input. Having a single control point for master data decisions with input from all functional areas will establish the continuity required to support concise master data management.
Single Source Alternative
But what if you cannot get a single source of truth? Chris Bergh, CEO of Cambridge, Mass.-based DataKitchen, a DataOps consultancy offers an alternative. He suggests using DataOps to create an integrated set of workflows with automated governance, which would be ideal for SMBs who might not have as many resources.
Data teams using inefficient, manual processes often find themselves working frantically to keep up with the endless stream of analytics updates and the exponential growth of data. If the organization also expects busy data scientists and analysts to implement data governance, the work may be treated as an afterthought, if not forgotten altogether. SMBs who are strapped for resources and using manual procedures need to carefully rethink their approach to governance.
With automation, governance can execute continuously as part of development and operations workflows. Governance automation is called DataGovOps, and it is a part of the DataOps movement.
Governance is, first and foremost, concerned with policies and compliance. Some governance initiatives focus on enforcement — somewhat akin to policing traffic by handing out speeding tickets. Focusing on violations positions governance in conflict with analytics development productivity. Data governance advocates may get farther with positive incentives and enablement rather than punishments.