Data has become all-important for success in today’s knowledge economy. Enterprises strive to derive insights from data and use it for competitive advantage. Driving new revenue sources, keeping customers happy, and ensuring efficiencies depends on harnessing data the right way. A robust data governance framework ensures the easy availability of enterprise data.

Enterprises that do not take data governance seriously waste resources and miss opportunities. Data processing and clean-up are time-consuming and can take up half of the time of highly paid data scientists. Poor data quality forces enterprises to perform non-value-adding tasks. Such tasks consumed 30% of total enterprise time in 2019.

Here are seven things to consider when designing an effective data governance framework.

1. Get the basic structure right

Data governance entails organizing data at the enterprise level. The scope of any data governance framework includes:

• Defining ownership of data
• Clarity on data collection, such as sources of data, method of data collection, classification of data, and so on.
• Method of cleansing and making data ready for analytics, and dealing with disparate data.
• Deploying infrastructure and fixing methodology for storing data.
• Policies and processes related to enterprise data.
• Restrictions cantered on the data, as in the users allowed to access or change each data set.
• Deploying technology. Technology alone cannot make a data governance framework successful. But not having the right technology can lead to failure to implement strategies or at the required pace.

There is no “best data governance blueprint.” A good data governance strategy is flexible to suit enterprise needs. But it remains comprehensive enough to cover these basic elements.

2. Align the data governance strategy to business needs

The best data governance strategy ties together people, processes, and technology.

An effective team aligns data management processes to business needs. 

Resilient enterprises:

• Approach data governance holistically. They consider all data assets but give priority to business needs.
• Link data governance initiatives with business use cases. The data governance team identifies the business need and the challenges in fulfilling such needs. Such challenges make up use cases. It may relate to growing the business, running the business optimally, or managing risks effectively.
• Give priority to use cases even if the solution is not perfect. Adopt a “needs-based” approach to governance. Consider a transformation project to upgrade analytics. In today’s fast-paced business environment, it might not be viable to wait until the entire data is ready. Rather, give priority to transactional data and product data. Logging sales data and establishing a clear hierarchy of products might get the job done.
• Link data governance projects to digital transformation efforts. Digital transformation projects already have the attention of the CEO and the involvement of the rank and file. Most transformation projects, be it digitization, omnichannel enablement, or modernization, depend on high-quality data.

3. Get top-management buy-in

Often, the C-suite executives don’t recognize the potential of data governance to create value. Data governance becomes a support function patronized by IT and ignored by others.

• Get top-down business-leadership support upfront. Convince them of the benefits. Quantify the benefits, if possible. Or else, illustrate the indirect benefits.
• Have clarity on funding. Often, data governance fails not because of incompetency or lack of a good plan, but due to the lack of resources to implement the best-laid plans.
• Involve the top management in data governance. Assign responsibility of business data to CFOs, CMOs, and other executive-level leaders.

Enterprise leaders become champions of data governance initiatives once they understand the benefits. Their support also pre-empts issues such as role clarity and conflicts over data ownership.

4. Embrace Agility

Data governance often restricts access to data and stifles innovation. Successful enterprises make the right trade-off between strong governance and accessibility. They:

• Use governance appropriate for the data. For example, apply light governance for data that does not go beyond the boundaries of the enterprise. Such data may also not need full metadata.
• Identify sensitive data, such as personally identifiable information. In most enterprises, critical data that require continuous tracking and restricted access is only about 10-% to 20% of all data. Uniform attention to all data is overkill. It leads to resource wastage and delays the rollout of projects.
• Set sensitivity level for each data to free up low-risk enterprise data. An Asian financial enterprise that took this route could free up about 60% of its data for access to all employees.
• Use data masks to ensure privacy.
• Make employees sign internal non-disclosure agreements (NDAs).

5. Implement data projects Iteratively

Successful data governance initiatives depend on an iterative approach. Doing everything in one go often ends up in confusion and wastage.

• When implementing new data-centric projects, work in sprints. Identify and implement priority areas based on the potential value.
• Establish strong master-data management for agility
• Establish data governance as a scalable service, to get incremental value and reduce costs.
• Implement robust collaboration systems to share data assets and resolve issues

6. Get Rank and File Support

Data governance initiatives need the active cooperation of the rank-and-file users, who use or process the data.

Leading enterprises motivate employees to ensure the right quality data at the source. They also attempt to convert skeptics. Some initiatives towards such ends include:

• Identify stewards from the rank and file to promote data governance initiatives.
• Training employees on how to improve the quality of data they handle.
• Recognizing and rewarding the creation of high-quality data
• Spreading awareness through events, publications, data art, and anything.

7. Address regulatory concerns

One of the biggest reasons for data governance initiatives is addressing regulatory concerns. The initial trust for data governance came from banks that had to comply with BCBS 2391 and other regulations. Enterprises with multiple businesses or businesses in various geographies often face conflicting regulatory obligations. The European Union General Data Protection Regulation (GDPR), for instance, mandates enhanced protection for the personal data of European consumers. GDPR applies to even businesses based outside the EU. Several other regulations apply globally, regionally, or locally.

• Identify data that needs regulation. Understand how and where the enterprise uses regulated data.
• Adopt technology and automate to reduce complexity in compliance management.
• Evaluate the risk to regulated data continuously.
• Tweak the governance model depending on the level of complexity. For instance, a global bank would need C-suite representatives in its data governance council. They would also need automated metadata tracking. In contrast, a tech company operating in single geography may only need less formal governance. The metadata track may even be in MS-Excel sheets.

The value of the data governance market touched USD 1.81 billion in 2020 and will be USD 5.28 billion by 2026, with a CAGR of 20.83%. Enterprises that invest in data governance unlock insights worth millions of dollars.

Here are the top priorities of big data analytics in 2022.