Data governance is a set of principles, standards, and practices that ensures your data is reliable and consistent, and that it can be trusted to drive business initiatives, make decisions, and power digital transformations. A successful data governance program enables you to do these things in a way that is repeatable, and which can scale and adapt as data volumes—and sources—grow and technologies evolve. In short, good data governance means you can use your data with confidence, both now and in the future.
Data governance requires an organization to understand and take stock of regulatory requirements, legal requirements, and business best practices which their data must meet, then establish rules and adopt automated and human processes to enforce the rules. The drivers of data governance are usually regulatory and legal requirements; however a governance rule can be any practice to which the organization wishes to adhere. Governance often dictates where certain types of data may be stored and codifies data protection methods, such as encryption or password strength. Governance can dictate how to back up data, who has access to data, and when archived data should be destroyed. Organizations can also set governance objectives around improving data quality or breaking down silos that isolate certain data.
You often hear about data governance “frameworks.” A data governance framework consists of the rules, people roles, processes, and technologies that work together to align everyone in the organization on your data governance strategy. If data governance is “what,” then a data governance framework is “how.”
This article defines data governance, identifies four core components of successful data governance, and provides five actions to begin your data governance journey.
A successful data governance program possesses four key ingredients: vision and business case; the right people; intelligent data governance technology; and efficient processes.
The vision spells out your broad strategic objective for building a governance program. The business case clearly articulates the specific business opportunity. Put another way, your vision is your destination, and your business case is your vehicle for getting there.
A vision statement, although broad, should be actionable, not abstract. It should look three to five years ahead. For example, “create a better customer experience by reducing the time to resolve issues, delivering more relevant marketing materials, and protecting sensitive customer data,” is a solid data governance vision.
The business case must also be actionable, but will be more pragmatic and hands on, specifying the actual people, roles, technologies, and processes involved in moving your data governance efforts forward.
You need to put the right people in place to support, sponsor, steward, operationalize, and ultimately deliver a positive return on your data.
First and foremost, you need an executive steering committee or data governance council. This group communicates, prioritizes, funds, resolves conflicts, and makes decisions about data governance for your entire enterprise. The steering committee is made up of the executive leaders of your organization. Sometimes they belong to the C-suite; they may also be vice presidents or directors accountable for the specific lines of business.
You might be lucky and find an executive sponsor immediately so that a steering committee gets off the ground right away. In many cases, however, some grassroots efforts are required before senior leaders are willing to commit. UNC encountered this situation and figured out how to navigate the pushback by starting with small governance projects and building understanding of their internal culture so they could move confidently when a growth opportunity suddenly emerged.
An effective data governance program should include the following roles in addition to the high-level people on your executive steering committee:
Executive sponsor. This is the C-level executive whose responsibilities span functional, line-of-business, application, and geographic silos. Identify your sponsor early, because this person allocates resources, determines staffing and funding, identifies high-priority business issues, and fosters cross-functional collaboration.
Data stewards. Data stewards are the business and IT subject matter experts (SMEs) who translate how your data governance framework affects your organization’s business processes, decisions, and interactions. Business stewards must be IT-savvy. Likewise, IT stewards must understand the business. Experienced business analysts capable of acting as communication bridges between business and IT can make the best business stewards, while data and enterprise architects and senior business systems analysts make strong IT stewards.
Data governance leader. This person coordinates tasks for data stewards, helps communicate decisions made by stewards to relevant stakeholders, drives ongoing data auditing and metrics that assess program success and ROI, and is the primary point of escalation to the executive sponsor and steering committee.
In the context of data governance, “technology” primarily means automation. Many technology solutions and platforms exist to help you automate different aspects of data governance, traditionally completed manually. But you must consider the full lifecycle of critical enterprise data, from creation to archival, when choosing technology to support your data governance efforts.
You should also focus on intelligent automation. Intelligent automation possesses four key qualities:
Automation: Leverage the power of artificial intelligence (AI) and machine learning to reduce redundancy and free up the time of data professionals—especially data stewards and your data operations team—and allow them to focus on more important matters.
Scale: New data types and sources are always emerging. The huge and growing volumes of enterprise data need to be assessed, curated, and protected so that anyone or any system can use them. The technology chosen for data governance needs to continually govern more data from more sources and handle more user requests without trouble, scaling up to the largest cloud data lake.
Extensibility: Modern data governance isn’t just about documentation and compliance. It delivers measurable value for the entire organization. Intelligent automation thus goes well beyond data governance project management to deliver a fully integrated platform for data management that includes data quality, data privacy, data cataloging, and stewardship capabilities.
Agility: An intelligent automation solution enables you to get started quickly without massive and costly custom integration efforts. and the flexibility within a proven intelligent data governance solution will also allow your organization to react swiftly to new regulations, new business models, or new competition.
Business policies and standards are critical for any data governance program. You need to agree on policies that span the enterprise rather than be confined to a particular line-of-business or departmental silo. Typical policies include data accountability and ownership; organizational roles and responsibilities; data capture and validation standards; information security and data privacy guidelines; data access and usage data retention; data masking; and data archiving policies. With the culture at each organization being different, there isn’t a right or wrong set of policies to consider as you map out your data governance program. The one area to watch out for, however, is to steer clear of being considered overbearing with red tape. Successful data governance today should focus their policy and procedure decisions more on collaboration than on control. Decide together on what’s best for the organization while also understanding that enforcement doesn’t have to feel restrictive. By making this pivot, you will shift your data governance program from being policy centric to value centric.
If you are just beginning to explore data governance, here’s a simple five-step roadmap to help you succeed.
Your first data governance initiative is critical. Get it right and you’ll have the opportunity to expand into an enterprise-wide program.
Very important: this first attempt at data governance must deliver a hard return on investment (ROI) – or at the very least return on effort - in a reasonable timeframe to show demonstrable value to the business. If possible, make it a project that will excite senior management. That means being able to provide metrics that show tactical success as well as progress on longer-term goals.
What do you want to achieve? This is not a rhetorical question. More governance programs fail because goals are too vague, or expectations differ. Here are some examples of the most common data governance goals:
Improve efficiency of critical processes that have been held back by low-quality data
Comply with regulations more effectively
Consistently use trusted data across the enterprise to drive every tactical and strategic decision
Data governance programs involve a lot of people. Even if your actual data governance team is small, your project will impact large numbers of employees, customers, partners—in short, anyone who depends on your data. Many of these people will have opinions, and some will voice them loudly. Don’t be fearful of this. Embrace their passion but make sure to organize it.
Use a responsibility assignment matrix like RACI (which represents roles for responsible, accountable, consulted, and informed). This ensures that the right people provide input—and approvals—at the right time, and that everyone understands their individual responsibilities.
The person responsible is likely to be an experienced project manager who manages schedules, assigns resources, and builds the case. The person accountable takes ownership of the major decisions and the results of the program. This is likely to be an executive-level person who owns the resources, and who has veto power. Consulted are the business and IT subject matter experts who will help you provide the necessary context to achieve your goals. And the informed are the people who will be affected by your data governance effort, but who don’t have a direct say in the direction of your initiative—something you’ll need to make clear from the start.
Your data governance teams need clearly defined, repeatable processes that are designed for the reality of the task ahead. There are four core processes that support every data governance program:
Discover: Identify and understand the data being governed
Define: Document data definitions, policies, standards, and processes. Assign ownership (a critical, often-overlooked step) and define your key metrics and KPIs
Apply: Operationalize data governance policies, business rules, and stewardship
Measure and monitor: Measure the value of your data governance efforts and monitor compliance with your policies
Data governance initiatives are always evolving. New internal data projects as well as regulations (and new risks) constantly appear. You need a technological platform that delivers value today but can also adapt and evolve as your requirements change. Here are some key considerations when considering your data governance technology:
Focus on flexibility and interoperability
Automate to accelerate processes, workflows, data discovery and reporting
Consider the cloud for scalability gains
Build a metadata repository
To learn more about the must-have elements of a modern, intelligent data governance solution, visit www.informatica.com/solutions/what-is-intelligent-data-governance.html
AIA Singapore offers insurance products and medical protection to individuals and businesses in Singapore.
Goal: AIA sought a deeper understanding of its business by identifying customer and financial data based on lineage and intelligent metadata. The objective: improve data quality to increase sales, improve decision-making, and cut costs.
Solution: AIA used Informatica Axon Data Governance to create a data governance framework, which automatically scanned and indexed metadata from core systems using Informatica Enterprise Data Catalog.
Result: AIA achieved a deeper understanding of customer data by tracking data movement and transformations. The Informatica solutions also maintain data quality, enabling AIA to optimize sales, decision-making, and costs.
McGraw Hill Education is one of the "big three" educational publishers providing educational content, software, and services for pre-K through postgraduate education.
Goal: McGraw Hill wanted to grow revenues in an increasingly digital educational marketplace. To achieve this, it needed to improve business intelligence reporting.
Solution: By deploying Informatica Axon Data Governance, McGraw developed a data governance management framework. It used Informatica Data Quality for data profiling and to track data quality.
Results: Today, McGraw Hill is seeing strong digital growth in the higher education market, increased profitability, and improved decision-making through a better understanding of sales trends through trusted data.
For first-hand insights into building successful data governance program, there’s no better source than your peers. Learning from people who have charted their own governance journey gives you not only best practices, but a chance to learn from their missteps and build on their experiences. The Informatica Data Empowerment Experts Series is a monthly webinar series that does just that: Bringing together people from a variety of industries to share the stories and real-world lessons they learned while empowering their organizations with clean, well-governed data. Register now for the next webinar or catch up on to past sessions on demand at www.informatica.com/dataexperts.
If not done diplomatically, data governance may sometimes perceived as just more red tape and corporate controls by the people it affects. That’s why it’s important to take your first successes and evangelize them. A little internal marketing goes a long way to publicize the value you’re bringing to the organization. And by encouraging people to understand it and even participate in data governance activities, you’ll help them see it less as a rigid sort of control and more as an exercise in collaboration for advantages that will benefit everyone, such as:
More efficient and comprehensive reports that draw on un-siloed data
Better collaboration between business and IT, thanks to shared responsibility for improving data quality
More precise predictions and planning strategies
More accurate analytics driven by democratized access to trusted data
An intelligent approach to modern data governance that includes the right people, processes, and tools is key to the success of your organization’s digital transformation journey. Whether you’re pursuing greater customer centricity, better analytics, or improved regulatory compliance, an enterprise data governance program can ensure that the data driving your initiatives is trustworthy, high-quality, available, and accessible to everyone who needs it.