For years, companies of all sizes have focused on business process efficiency, network management, infrastructure optimization and outsourcing. Only in the past few years has the concept of “data governance” begun to get the attention of senior management. So, what is data governance?
Data governance is more than data quality, master data management or archiving and retrieval – it is the fundamental process of managing data across the enterprise: the policies, rules and systems that govern how data is managed and used throughout an organization. Data governance affects both the business and IT functions in an organization, and “unites” them at a policy level while monitoring them at an operational level. This closed-loop process provides a framework for continuous improvement across the enterprise.
Until recently, data governance has been at the bottom of the compliance checklist for most business executives. As long as data was backed up and stored offsite, the company was covered. But on a daily basis, operations struggled to share information about common customers, orders, suppliers and partners
Enterprise data governance has its challenges, but the rewards far outweigh the risks. By embracing enterprise data governance, organizations can:
- Improve business decision making through data consistency and accuracy;
- Improve data quality through defined and documented processes;
- Dramatically reduce the cost of data integration;
- Reduce risk by enforcing compliance, and;
- Establish a clear view of data to take advantage of system consolidation or outsourcing opportunities.
Ultimately, both cost and risk are reduced, contributing to the overall health of the organization.
So what are the challenges of implementing a data governance program? There are four key areas of focus:
- First, adoption can be difficult if there is no central repository for definitions, rules, policies and data quality metrics. Much of this information ends up in paper reports, making it difficult to share and syndicate across the enterprise.
- Second, enforcing policies is difficult to achieve without a layer of automation that enables the processes. Even if governance processes have been agreed upon, they are often not followed, especially as changes in the organization occur.
- Third, accountability can be a major issue. Without reliable metrics, it can be a challenge to drive improvements in data quality across the organization without some way to track enforcement.
- And finally, given the current global economic climate, coming up with hard dollar savings can make it difficult to justify such a program. Strong measurements and metrics are required to show the real return of a data governance program and give it the momentum required to expand beyond its initial scope.
Every business has to think about the value of their data when considering how it should be managed. What is the role of that particular piece or set of data? Is it mission critical? How does it affect other data or business processes? Think about each business process in your company. What if one in every five transactions had to be re-worked? What if you couldn’t complete 20 percent of your call center interactions because of a lack of high quality data? What if your name and address information had to be updated monthly across 300 systems? How could you possibly tell if all those systems were in compliance? The cost of bad data is real and the only way to fix it is to embrace data governance as the first step in driving a new age in data management.
This post originally ran on CIO.com