Share this article
Establishing a well-defined data governance framework is essential for a data governance initiative. But it’s tough to know where to start on a project with such a massive scope. Here’s a roadmap you can use to get started.
What is a data governance framework? #
A data governance framework, also known as data management framework, is a defined structure that directs the implementation of guidelines, protocols, processes, and rules for data in an enterprise. It serves as the foundation of a data governance program.
See How Atlan Simplifies Data Governance – Start Product Tour
According to the DGI (Data Governance Institute), the data governance framework is “a logical structure for classifying, organizing, and communicating complex activities involved in making decisions about and taking action on enterprise data.”
In other words, a data governance framework is “the how — a blueprint for enforcing governance.” It should provide a clear visualization of how to ensure the quality, integrity, security, discoverability, accessibility, and usability of your data assets.
Table of contents #
- What is a data governance framework?
- Advantages
- Pillars
- Examples
- How to create an enterprise data governance framework
- A bottom-up approach to data governance
- Atlan: Effortless data governance for the modern data stack
- Related reads
How do you benchmark a good data governance framework? #
According to this paper on the proposed Data Governance Framework for Small and Medium Sized Enterprises (SMEs) by the Minnesota State University, a solid enterprise data governance framework should:
- Enable better decision-making
- Reduce operational friction
- Protect the needs of data stakeholders
- Train management and employees to adopt common approaches to data issues
- Build standard, repeatable processes
- Reduce costs and increase effectiveness through coordination of efforts
- Ensure transparency of processes
What are the advantages of an enterprise data governance framework? #
A data governance framework enables enterprise-wide collaboration to manage all data assets, thereby aligning them to your organization’s overall corporate objectives.
Here’s how a data governance framework helps:
- You have great visibility of how to synergize your data governance efforts
- You have the plan to control and have visibility of your entire data estate
- Transparency becomes inherent to how you manage data
- Monitoring your data consumption or use is efficient since it follows the blueprint
- A framework also sets a strong foundation for regulatory compliance practices
In short, less chaos → more trust → more value.
What data chaos looks like, in the absence of effective data governance. Source: Atlan.
This paves the way for true data democratization, effective collaboration across teams, and compliance with data protection laws and regulations.
The three pillars of any data governance framework #
If you look up the pillars of any data governance framework, you’ll find responses that include standardized policies and procedures, data security and access, compliance, risk mitigation, etc.
However, these can be seen as the components of a data governance framework template.
The pillars must reflect the essence of governance for the modern data stack. That includes making data flows traceable and data-related processes transparent so that you can understand your operations, improve your performance, and achieve your goals.
That’s why the following three pillars form the crux of any enterprise data governance framework for the modern data stack:
- Governance encompassing all data assets
- A practitioner-led, bottom-up approach
- Governance practices embedded within daily workflows
Let’s take a look at each of these pillars in turn.
Governance encompassing all data assets #
Everything from dashboards and code to data science models is a data asset. The data governance framework should take into account all data assets, i.e., data and analytics governance.
A practitioner-led, bottom-up approach #
As the number of data users and consumers keeps rising, making a few people (data stewards or engineers) accountable for data governance isn’t a sustainable approach.
Everything from dashboards and code to data science models is a data asset. The data governance framework should take into account all data assets, i.e., data and analytics governance.
A practitioner-led, bottom-up approach #
As the number of data users and consumers keeps rising, making a few people (data stewards or engineers) accountable for data governance isn’t a sustainable approach.
A decentralized, bottom-up data management framework that makes every data creator responsible for data governance is the way forward.
An example of a decentralized, community-led approach is the data mesh. The data mesh design proposes a federated computational governance model, where every organization is a federation of business domains. Domain owners fully manage the data they create.
However, each domain still follows a set of global (or federal) rules on data definitions, standards, processes, and discovery.
Governance practices embedded within daily workflows #
Data governance has always been associated with compliance, control, and risk mitigation. However, it is a business function that can support strategic decision-making by ensuring that everyone has access to accurate, relevant, high-quality, and trustworthy data.
That’s why data governance can’t be an afterthought. Instead, it should be embedded within the daily workflows of data practitioners.
Data governance framework examples #
There are several established, tried, and tested data governance framework examples already in use, such as:
Let’s briefly dive into each of these models to understand the differences and similarities.
1. DGI #
The DGI framework comes with ten universal components that address the why-what-who-how of data governance.
The DGI data governance framework. Source: DGI
Let’s look at some of these components:
- Goals, metrics, and funding are all about elaborating how the data governance program would increase revenue, optimize costs, and ensure business resilience despite risks or disruptions.
- Controls are for risk management and can be preventive or corrective. They can be applied at various levels of the framework to support the goals of the data governance program.
- A DGO (Data Governance Office) oversees the entire governance program, collaborates and liaises with other stakeholders, aligns data-related policies and standards, and maintains detailed records on the program.
DGI divides each of its components into core areas — rules, people, and processes — to simplify the concepts.
The ten components of the DGI data governance framework. Source: DGI
2. DAMA DMBOK #
DAMA DMBOK is another popular data governance framework.
DAMA-DMBOK data governance framework. Source: DAMA-DMBOK
It pictures data management as a wheel with data governance at the center (the hub) surrounded by nine knowledge areas. Data governance is considered to be the high-level planning required for effective data management.
Each knowledge area explores an avenue of data governance. For instance:
- Data architecture management represents the overall structure of data and how it connects with each application within the data ecosystem.
- Data development is all about data modeling, requirement analysis, design, implementation, and maintenance of data storage elements, like databases.
- Metadata management involves collecting, categorizing, integrating, and maintaining high-quality metadata.
The framework further defines environmental elements that provide structure to each knowledge area. They define the underlying processes, roles, technologies, and deliverables that guide the planning and execution of each area.
They also cover how an organization’s culture must evolve for data governance initiatives to work.
The seven environmental elements guiding each knowledge area in DAMA DMBOK. Source: DAMA-DMBOK
3. McKinsey #
McKinsey believes that rethinking the entire organizational design is the starting point for ensuring success with data governance. Their data governance framework template includes three core components:
- A data management office (DMO) defines policies and standards, trains and guides data leaders, and ensures that data governance is connected with every other function within the organization.
- Domain-based roles manage the day-to-day execution of the data governance program.
- A data council heads the overall strategic direction of the data governance program. It brings the DMO and domain leaders together to review progress, authorize funding, and resolve issues and roadblocks for effective governance.
4. Eckerson #
The Eckerson Group has six layers and 39 components in its proposed data governance framework. Let’s look at some of the layers:
- Goals and standards address the why and how of implementing a data governance program.
- Processes ensure that the data governance initiatives meet their end goals. This could be anything from ensuring quality and accuracy to cataloging metadata.
- Culture fosters an environment of collaboration, data democratization, and transparency without any conflict.
The USP of this framework is that it puts people at the heart of data governance by defining roles such as data owners, stewards, curators, and stakeholders to outline their roles and responsibilities when accessing, using, and changing data.
That’s because:
“The reality is that we don’t govern the data. We govern what people do when working with data.”
Eckerson Group’s modern data governance framework. Source: Eckerson
5. PwC #
The PwC enterprise data governance framework takes conventional models such as DAMA DMBOK and DGI a step further to account for next-gen data landscapes.
The PwC enterprise data governance framework. Source: PwC
PwC includes five components in its data governance framework standards. It starts with a data governance strategy, followed by a management layer encompassing all the aspects of a data ecosystem.
The lifecycle management layer covers all the policies required to ensure a streamlined flow of data throughout its lifecycle.
The stewardship layer focuses on enforcing governance and the governance enablers account for the people, processes, and technologies involved in ensuring effective governance.
6. Deloitte #
According to Deloitte, the data governance of tomorrow is about “maximizing the value of data for operational effectiveness, decision making, and regulatory requirements, and minimizing the risks associated with poor data management.”
Deloitte’s recommendation for a data governance framework consists of elements, such as:
- Policies and principles guiding data governance and data management
- Organizations establishing data governance roles and responsibilities
- Processes outlining how data is created, modified, and maintained
- Tools and technology chalking out the tooling, modeling, and data architecture implementation
- Governance controls define the metrics to measure the effectiveness of data governance
Deloitte also suggests continually monitoring and improving the data governance framework.
Deloitte’s data governance framework. Source: Deloitte
How to create an enterprise data governance framework #
Here are five steps you can follow to create an enterprise data governance framework:
- Revisit your definition of data governance
- Identify and define data domains
- Identify domain data owners and consumers
- Validate and document everything about the data
- Conduct data security and risk assessments for each domain
How to create a data governance framework? - Image by Atlan.
Step 0. Outline your data governance framework #
Here’s what popular data governance frameworks have in common:
- They start with the ‘why’ — the goal of data governance
- The goal is followed by the ‘what’ — what data gets governed
- Then comes the ‘how’ — how will that data get governed, and what are the processes, people, and tools involved
Before you develop a framework, answer these questions for your organization. If you want to learn more about these data governance best practices, this blog covers them in more detail.
Step 1. Establish your definition of data governance #
Data governance is an ever-evolving project, which is why you should establish your idea of data governance before formulating a data governance framework.
Snowflake mentions: As data volumes grow, new data streams emerge, and new access points emerge, you’ll need a policy for periodic reviews of your data governance structure — essentially governance of the data governance process.
So, here are some questions you should be asking while developing your definition of data governance:
- What is the purpose of data governance?
- Does it cover all data assets across the organization?
- Does governance also foster organization-wide data sharing and collaboration?
Step 2. Identify and define data domains #
Since the data management framework should cover all data assets, the next step is to identify and standardize data domains across your organization. You can have domains such as finance, marketing, sales, etc. corresponding to each function generating data.
Here are some questions you should be asking:
- Which are the prominent data domains in our organization?
- What data do they generate?
- Where is that data now?
- Who consumes that data?
Step 3. Identify domain data owners and consumers #
A key tenet of a data governance framework is shared responsibility for data. So, each domain creating data is responsible for managing it and ensuring its security, integrity, and privacy.
That’s why the next step is to assign data owners to each domain and understand its data consumption pattern to ensure that the right people have access to the data they need.
Here are some questions to get you started:
- Who is creating data within each domain?
- Who is consuming that data, and how? What do their daily workflows look like?
- What are the current dependencies to get access to domain data?
Step 4. Validate and document everything about the data #
By this stage, you should have a clear idea of data flow within your organization. The next step is to standardize data domain definitions, data flow rules and workflows, access policies, and more by documenting everything.
The documentation should address the following:
- Where does data originate from?
- What does it mean?
- How does it flow through your organization?
- Does it help domains meet their goals?
- Does it support your organization’s business outcomes?
A great way to document and share all that information is to set up a data workspace that uses active metadata to keep your documentation relevant, fresh, and useful.
Step 5. Conduct data security and risk assessments for each domain #
To complete your data governance framework, set up processes to conduct frequent data security and risk assessments for each domain. That’s because enabling data governance is a journey, rather than a one-time project implementation.
You should start by asking yourself:
- What are the existing data access policies and security checks for data from each domain?
- Who is allowed to access what data and why?
- Do these policies mitigate risks without creating data discovery, access, and collaboration bottlenecks?
Once you have followed these steps, you should be able to get started with building a decentralized, community-led data governance framework that works for everyone in your organization.
A bottom-up approach to data governance #
Implementing a solid data governance framework requires a substantial change in how organizations create, perceive, and use data.
Data leader and CEO of Cognopia (a company that advises Fortune 500 companies on data-driven business transformation) Neil Burge, calls governance an enterprise change program:
“If you are about to launch a Data Governance initiative, spend some time learning about the people that you need to engage to actually set and enforce rules around data. You are changing people’s behaviour around data, the processes they use to do their day jobs, and the tasks you ask them to undertake on a daily basis will change.”
That’s why it’s essential to develop a bottom-up, practitioner-led data governance framework and keep improving it with periodic reviews and assessments.
Atlan: Effortless data governance for the modern data stack #
The entire data management space is going through a paradigm shift.
The data world is slowly converging around the best of the tools for processing large amounts of data, a.k.a the “Modern data stack”
Data governance for the “Modern data stack” needs a rethinking.
This is where Atlan comes to your support — a data catalog and data governance solution built for agility, trust, and collaboration.
Related reads on data governance framework #
- What is Data Governance? It’s Importance, Principles & How to Get Started?
- AI an Data Governance Manifesto
- Federated Data Governance: Principles, Benefits, Setup
- Snowflake Data Governance — Features, Frameworks & Best practices
- Open Source Data Governance Tools - 7 Best to Consider in 2024
- Data Governance Policy: Examples, Templates & How to Write One
- 7 Best Practices for Data Governance to Follow in 2024
- Benefits of Data Governance on AWS: What’s Available and How Can You Build On It?
- Benefits of Data Governance: 4 Ways It Helps Build Great Data Teams
- Key Objectives of Data Governance: How Should You Think About Them?
- Data Governance Roles and Responsibilities: A Quick Round-Up
- The 3 Principles of Data Governance: Pillars of a Modern Data Culture
- A Guide to Gartner Data Governance Research — Market Guides, Hype Cycles, and Peer Reviews
- 5 Popular Data Governance Certification & Training in 2024
- 8 Best Data Governance Books Every Data Practitioner Should Read in 2024
- Automated Data Governance: How Does It Help You Manage Access, Security & More at Scale?
- Data Governance and Compliance: Act of Checks & Balances
- Data Governance vs. Data Management: What’s the Difference?
- Enterprise Data Governance — Basics, Strategy, Key Challenges, Benefits & Best Practices.
Share this article
