Data Governance: The Foundation a of Self-Service Analytics Deployment

Data Governance: The Foundation a of Self-Service Analytics Deployment

It’s all about protecting your data and enabling enterprise-wide adoption of data analytics

What is Data Governance and why is it Important?

Data governance forms the guideline for organization-wide data management practices and processes that make efficient use of trust-worthy data possible.

With the volume of data and data sources exploding, combined with our business functions working towards becoming increasingly data-driven, the IT teams are simply overwhelmed. This has given rise to the need to make everyone in the organization data-literate so that they can recognize patterns in data themselves and inform their decisions without waiting on the IT teams. We call this phenomenon self-service analytics.

Once you recognize the need for self-service analytics, you will want to start laying the groundwork for data management and here’s why it is the natural next step: when everyone in the company is looking at your data, it is not just the best practice to ensure that the data they’re looking at is most up-to-date, accurate and relevant but common sense. Think about what would happen if Business Development, Sales, and Marketing were looking at different versions of sales and revenue numbers. Think about the repercussions your organization would face if sensitive data landed in the hands of an executive who could misinterpret it without sufficient context. Disaster!

If you’re thinking, most organizations have some form of data governance for individual departments and applications, you’re not wrong. The challenge, however, is that it’s not necessarily institutionalized. A systematic introduction of data management practices is therefore a transition from informal rules to formal control that is consistent across the organization. Without good governance, investments in data and analytics will fail to meet key organizational demands such as revenue growth, cost optimization and better customer experience. Hence, data governance serves as the foundation of self-service analytics.

Common Data Problems Observed in an Organization

  • Unclear ownership and accountability of data - not knowing who to turn to for data clarity
  • Low data discoverability - unable to find the data when needed
  • Unclear change management process – not knowing what to do after making changes to the data, how to document or communicate the changes
  • No single source of truth - unsure of which are the latest version of data sets users should refer to
  • Missing data definitions and formulas – unable to understand what the data represents, in what context or how to use it effectively
  • Unclear roles and responsibilities - who oversees the maintenance of a data source, what are the expectations from data owners, etc.

Advantages of Data Governance

  • Consistent data and uniform processes that lead to better decision support
  • Increase in scalability of the IT landscape at a technical, business and organizational level
  • Optimization of data management costs through central control mechanisms
  • Achieving compliance guidelines
  • Improved security for internal and external data
  • Transparent communication, increased efficiency and shorter turnaround times
  • 3 Components of a Data Governance Model

After understanding the commonly observed data life cycle problems, benefitting from data governance requires rethinking its organizational design. An effective governance structure comprises of three critical components:

Data Management Office (DMO)

Led by a Chief Data Officer (CDO), a DMO is typically responsible for defining the governing policies and standards. It empowers the data leaders within data domains with tools, playbooks, training, etc., and ensures coordination and consistency across key roles in the data life cycles.

Data Domain Leadership

Data leaders set and execute strategies of their domain that include taking initiatives to train executives, employing external sources to fill the gaps in information, etc. Their job is to understand the needs of data consumers and find ways to meet them. They own and manage the data – assessing and improving data quality, updating data definitions, etc.

Data Council

Chaired by the head of DMO, the Data Council owns strategic directions and principles of data governance in an organization. Its participants are the DMO and Data Domain Leadership and act as an interface between them. It is responsible for proposing DMO structure, defining domains, and assigning leaders. Confirming adherence to standards and policies, reviewing progress on initiatives, serving as funding gateway and resolving issues are also the duties of the Data Council.

Best Practices for Data Governance Excellence

Once you have the three governance components in place, McKinsey, recommends following certain best practices to ensure your governance creates value.

Taking a top-down organizational approach
The first step in implementing successful data governance requires the DMO to engage with business leadership, understand their needs, highlight the data challenges, explain the role of data governance, and get their buy-in. The next step is to form a council with senior management which will give the governance strategy the right direction that is aligned with business needs as well as oversee and approve various initiatives to drive improvements.

Linking governance to organizational transformation themes
To ensure that governance efforts create value, integrating them directly with on-going transformation efforts that already have management’s attention makes it easier to shift data responsibility and ownership towards product teams. Thereby, incorporating the right governance mindset at the point of production and consumption.

Prioritizing data assets and aligning data leadership
A common mistake an organization makes is looking at all data assets at once. Not only does this overwhelm them but also prevents them from focusing on efforts that are directly linked to business needs. To succeed, data assets should be prioritized in two ways: by domains and by data within each domain. The transformational themes will guide the data council to prioritize domains and create a roadmap for deployment. Depending on the business needs, the level of governance sophistication can be decided.

Getting people excited about data
There’s no denying the fact that data-driven decisions yield higher success rates. To get people habituated to the data culture, leading organizations believe in taking a case-study adopted first by the senior leaders. They become role models in driving change, advocating the impact of quality data, building data supporters and converting the skeptics. When leaders use data and statistics as part of their everyday lingo, it percolates through layers in the organization.

Adopting an iterative approach and focused implementation
To keep the excitement going, getting quicker victories is crucial to showcase the impact sooner. Using iterations leads to faster adoption and helps data leaders prioritize use cases and domains. It enables them to apply need-based governance that maximizes the benefit to the business as priorities shift. Long term development can occur once the value of the governance program has been proven.

Getting Started

To implement data governance that delivers value, the process begins with making crucial decisions or requirements for the people and policies governing your data. Here’s a good place to start:

  1. Understand the opportunity cost of not getting data governance right in terms of missed upside, extensive time lost in manually cleaning data, or incorrect and suboptimal business decisions.

  2. Choose your data stewards and administrators who will oversee the program. Ask yourself what it would look like to elevate the conversation to top management.

  3. Identify the domains or parts of domains that need governance efforts the most.

  4. Define the processes and policies that work for your enterprise.

  5. Implement an iterative working cycle that gets you off the ground as soon as possible and allows you flexibility to future improvements.

  6. Choose a data governance solution that can work with your native architecture as easily as it can be customized to suit your evolving business needs.

  7. Design and plan your training program to onboard people and continually educate them on best practices.

  8. Data governance is critical for organizations looking for transformative opportunities in this hyper-connected world. The key to achieving success is to make governance activities a natural part of your organization’s everyday operations rather than thinking of them as frameworks.

If you’re interested in learning more about how Dundas BI can play a vital role in your data governance efforts, we’d love to chat. Get in touch with us.


About the Author

Dan Crowther


Dan Crowther is a Customer Success Advisor at Dundas Data Visualization Inc. He supports clients at every stage of their BI journey to help them achieve their goals with lasting competitive advantage.

Follow on Linkedin