Data governance: Best practices for a human approach

Almost every business now runs on data—every successful one, anyway. Getting (and keeping) a competitive advantage requires transforming data into actionable insights and delivering those to the right stakeholders as quickly and accurately as possible.

The challenge lies in the data itself: Your company captures massive amounts of data, constantly streaming in from different sources across your organization. The challenge: to collect, keep, and, above all, make sense of it all without your system becoming a giant, untraceable, out-of-date data quagmire that’s impossible to scale.

The solution for managing this deluge of data is straightforward: data governance. This guide examines best practices for the technical elements of data use and management, along with those addressing the human side of data governance.

What is data governance?

Data governance is an overarching term for the policies and practices your organization has in place to know where your data is, how it’s being used, and whether or not it’s secure and protected. A solid data governance program helps ensure the reliability and trustworthiness of data throughout its lifecycle in your organization, making sure it’s consistent and isn’t being misused or mishandled.

Data governance best practices

Good data governance is important for any business focused on making data-driven decisions. Unfortunately, many organizations approach governance strictly as data hygiene and control and not as a critical business capability on which key business outcomes rely.

Data governance best practices must address both the technical and the human aspects of how your organization deals with data.

Technical best practices for data governance

Technical best practices for a strong data governance program should focus on ensuring the consistency, security, and accuracy of the vast array of data across your organization. From standardization and automation to data classification and protection, these practices provide the structure for keeping your data reliable, accessible, and protected.

Search for dark data: A surprising percentage of your org’s data can actually exist outside your defined data pipelines. This “dark” data can live locally on staff machines or in the cloud; it can be found in files, folders, and as shared assets on team drives.

It’s important to gather this scattered and unstructured data for two reasons. First, this data, if properly governed, can actually be incredibly valuable data for your business. Second, mystery data is often at more risk than your managed data.

Standardize, standardize, standardize: Set format standards for every flavor of your data, from relational to lake to blob. Use tools like dbt Cloud to enforce those standards during post-processing and data ingestion.

When data that has been gathered for a variety of reasons is streaming in from a multitude of different sources, data normalization is essential not just for efficiency, but also your sanity.

Classify and tag it: Standardization is a major pillar of data governance, and you can’t achieve solid standardization without data classification and tagging.

Best practices here include establishing a framework for applying your org’s core data policies to any and all data assets. Define strong metadata standards that include data classes and tags that correlate with how your business ultimately utilizes data. Clear and thorough classification and tagging make data more easily findable and understandable, so everyone in your business can easily use (and reuse) it.

Prioritize automation: Automate everything possible—workflows, approval processes, data permissions and access requests, and any other common activities that are required to make your data governance initiatives efficient.

Automating aspects of data governance pays off in two big ways: your teams save time and resources when relieved of manual tasks and processes. Automation also ensures that your organization is consistently and continuously adhering to its standards and policies.

Secure and protect: When you have data, you have to protect it from forces both inside and outside your organization. Controlling and policing internal access to valuable (IP) or sensitive (personal) data helps your org mitigate the business and financial risks of external breaches from hacking, ransomware, and other malicious entities.

Establishing thoughtful data governance policies around accountability, transparency, and control reduces compliance risks. Such policies are crucial for complying at scale with regulations like GDPR and DORA in the EU, as well as sector-specific regulations such as HIPAA for healthcare and PCI DSS for financial organizations.

Data retention and disposal: Data governance policies also help manage the life cycles of your organization’s data. This includes how long you retain active data assets as well as policies for archiving or disposing of data.

Measure what matters: Metrics for data governance are how you comprehend and manage your data estate. Measurements vary widely from business to business, but some common best practice metrics to gather include:

• Amount of data retained and data utilization rates (i.e., what data is in active use vs. sitting idle?)
• Data freshness
• How much sensitive data you’re creating
• Data classification coverage
• Number of data breach or sensitive information leak incidents in a month/quarter

Organizational best practices for data governance

Effective data governance relies as much on people as it does on technical management. Best practices for the organizational aspects of data governance encourage accountability and make data integrity and security a whole-organization mission.

Designate a data czar: Data is essential and should be recognized and treated as such. Recognizing this, many companies are appointing Chief Data Officers (CDO) for executive oversight of the organization’s entire data estate, the way CIOs and CTOs are in charge of its technology.

Not every company needs a formal CDO. However, every company does need someone responsible for managing—and achieving—your data governance goals.

Identify business goals for governance: Everyone agrees that data is important. However, everyone is also already busy with their own day-to-day roles and responsibilities. When your data governance policies focus only on the data itself, there will likely be low engagement from stakeholders.

In other words, you have to tell your teams why data governance is important. This means explaining the connection between your company’s data, its governance, and the ways this new program is going to help them do their jobs better. It’s also important to tie these objectives back to company goals and OKRs.

Data guardians: A related best practice is to make good data hygiene an organization-wide priority. Strong data governance means creating—and communicating—a company-wide mission that everyone who works with data is responsible for both its accuracy and its security.

Clear roles and responsibilities: Any data governance program needs to map specific roles and policies to specific teams or even individuals, making it clear who is responsible for each part of your governance program and thus closing any potential gaps.

Create a dedicated data governance team: Designate those who directly touch data, no matter their role in the org, as Data Owners. Data Owners are closest to the data they create and manage, and they are overseen by Data Managers who provide guidance and handle communication with leadership, other Data Managers, other teams, etc. Your data governance team has to be empowered to enforce data standards and policies across the organization, too.

Continuous improvement: Data landscapes—the data your org collects and how that data gets used—are constantly evolving. An important data governance best practice is a formal program for continual evaluation of governance practices so you can update and improve them as necessary.

Keep it simple: Data governance is important. However, it’s also not the primary job for most people in your organization. Having a dedicated data governance team goes a long way - but getting org-wide buy-in requires keeping things as simplified as possible.

Your governance strategy should minimize the impact on individual contributors and teams as much as possible by making data-related tasks clear and easy to follow. Implementing easy-to-use data governance tools and making them a part of everyone’s daily workflows goes a long way here.

Data governance is a team sport

A well-designed and well-run initiative can turn data governance from a compliance task into a strategic advantage.

Aligning governance with business objectives and building a team that enforces and maintains governance standards across departments will signal to everyone in your organization that data is one of your most valuable resources. When data is accessible, accurate, and secure, teams can confidently use it to drive insights and innovation.

In an era where data is a key competitive asset, sound, human-centric data governance not only mitigates risks. It also primes your company for long-term success.