Over the last couple of years, the concept of data governance has rapidly grown in significance as businesses look to comply with relevant government and industry regulations, minimize costs, improve profit margins and deploy data-driven initiatives to turbocharge revenue. In line with this, CEOs and other C-suite executives are pushing for the adoption of data governance best practices that incorporate employees, vendors, systems and business processes.
The end goal of applying data governance best practices varies from one organization to the next. In most cases, it’s to improve the visibility of data assets in order to drive faster and better business decisions. For others, it’s to better the efficiency of enterprise data management. However, data governance could also be implemented to comply with regulations such as GDPR, HIPAA, SOX and Solvency II.
Whereas it isn’t difficult to define the objectives of a data governance policy, businesses often struggle to develop and implement a governance program that works. By applying proven data governance best practices, organizations can overcome the most common challenges when facing the implementation of a data governance framework. We’ll take a closer at some data governance best practices below.
Appoint a Governor and Governing Council
Like any organization-wide initiative, you need a clear and well-defined leadership structure that takes charge of the data governance program. Without such strong unambiguous leadership, the data governance program is destined to fail from the very beginning as tasks fall through the cracks due to data inconsistencies. Identify a ‘data governor’ who will be mandated by the CEO to drive and monitor the execution of data governance best practices, policies and rules across the enterprise.
Usually, this person will have another designation within the business (such as the chief operations officer or chief compliance officer). However, in large organizations, a chief data officer is entrusted with the data governance role.
Data governance is inherently a deeply political process that calls for consensus across a wide range of internal stakeholders. By having a data governor in charge of the process and who reports directly to the CEO, it’s easier to break down the silos between business units and get everyone working together toward the same goal.
The data governor will be the head of a data governance council that comprises representatives of key business units.
Develop a Strategy
The council should develop a vision of where it wants the organization’s data governance to be in the future. The strategy should ideally focus on the long-term goal but should also incorporate short and medium-term milestones that eventually lead up to the desired end objective.
Once this strategy has been finalized and signed off, the governance council can work backwards by defining the project plans and milestones that are needed to deliver on the strategic outcome. The plans and milestones are tracked by key performance indicators (KPI) and periodic reports submitted to the CEO and board to show progress.
Nevertheless, note that a great strategy is meaningless if it isn’t accompanied by great tactics. It’s in the tactical aspects of data governance that you translate the high-level strategic objectives into definitive deliverables.
Survey the Territory
With the data governance strategy and leadership established, the next step is for the governance council to embark on a detailed survey and inventory of the existing data governance policy across the organization. This exercise is important for two main reasons.
First, every enterprise has some elements of data governance already in place even before it sets out to formally develop a clearly defined and well-structured governance program. These can be incorporated into the governance program. They could even be a selling point for the program itself—employees will see the program as just an expansion of existing data management methods.
Second, no two organizations are the same. Each company has a unique fingerprint when it comes to its processes, procedures, personnel, vendors, risks and opportunities. By surveying the enterprise, you can have a real feel of where the organization is in terms of data governance.
With the knowledge gathered from the survey, the data governance council can establish a realistic road map of getting the organization where it should be.
Identify Pain Points
The purpose of the survey is intelligence gathering. During this process, one of the things you need to pay attention to are the pain points. These are the concerns that business users voice that demonstrate a need for a robust data governance best practices framework.
We’re talking about user comments around not trusting the data they receive, relying on their gut to make key decisions, difficulty in accessing the data they need, inability to interpret data and problems with matching the data they have to the problem they want to resolve.
By demonstrating how data governance can eliminate these pain points, the governance project is likely to experience enthusiastic support from the ground up. Often, employees see data governance as a relatively abstract concept that is far removed from their everyday realities. Matching governance to pain points makes it a more practical and relatable initiative.
For data governance to succeed it requires the support of all departments of the business. A modern large or medium-sized organization is a complex entity that comprises hundreds or thousands of moving parts. Ergo, attempting to incorporate every aspect of the business and enlist the support of all business units from the start can be overwhelming. It makes sense to start small and then scale your successes across the organization.
Even though the ultimate outcome of data governance is enterprise-wide compliance, breaking down the program into smaller milestones makes such a large undertaking feel more achievable. In addition, this staggered approach allows the governing council to concentrate their limited time and resources to the achievement of each milestone.
As data governance best practices and principles are applied to a select number of departments, business units or processes at the start, they are tested out and perfected so the council can determine what works best. As you roll out the governance program over a wider section of the organization, the entire process will be more refined, efficient and effective.
Price Your Business Data
If you cannot ascribe a specific value to something, it can be difficult to determine whether and what resources should be committed to it. Similarly, if a business doesn’t know the worth of the data it possesses, it is unlikely to sufficiently measure, protect and enhance the value of the information. Data is much like water—so critical to business operations yet so easily taken for granted.
To assign financial value, you have to develop an internal marketplace for business data that’s founded on data ownership, user entitlements and data management technology. That way, departments have to pay for the business data and data technology they use. With this transactional approach to data access, it’s easier to drive a data governance campaign as the various units can appreciate that loss of data quality, integrity and currency will cost them.
Understand Data Governance is a Program Not a Project
A data governance framework isn’t something you develop once and then forget about for good. It isn’t a project with a definitive end date and deliverable after which no further action is needed. Rather, data governance is a program. That means that it’s a combination of multiple projects, it doesn’t have an end date and is focused on an overarching outcome as opposed to a deliverable.
As some governance program projects are completed, new ones come up all in a quest to bring the organization closer to the strategic goals it has for its data. Having this understanding from the get-go is vital as it ensures that the board, C-suite executives, data governance council and the rest of the business adopts a long-term view.
Data governance isn’t just about the data but also involves the people who use the data. Without strong, coherent and consistent communication, a well-thought-out data governance program will not take off due to lack of operational efficiency. Communication plays a role from the very beginning of the program. In the early stages of governance implementation, convey the business benefits of the program to nurture buy-in.
As implementation progresses, communicate program successes via metrics that keep employees excited to participate. The data governance council must also clearly define what the role of each program participant is and what guidelines they must adhere to. Data governance involves change management so affected employees should be made to understand what changes are afoot and prepare accordingly.
Organizations are not static entities. They are constantly changing thanks to market demands, regulatory requirements, operational adjustments and new technologies. These changes can significantly alter the risk landscape. Unfortunately, many enterprises assess enterprise data risks just once a year. If the organization doesn’t have a data governance framework that includes building in data risk governance and monitoring into everyday business processes, then it runs the risk of constantly being behind the curve of the evolving enterprise environment.
Every major change to the organization’s technology and procedures should include a data risk checklist that evaluates how the change could potentially affect the ability to effectively apply the data governance best practices charter.
Data governance is a composite discipline that brings together data management, data quality, data policy, data security, risk management, compliance and business process management in order to effectively, consistently and coherently manage data assets throughout the entire organization. We’ve looked at the key considerations and what data governance best practices you can follow for efficient and effective implementation.
This is the third post in a five-part blog series on data governance. Make sure you check out the rest of the series to learn more about: