Insights | Analytics & Insights

You Want Data Science? Then You Need Data Governance

Data governance done right can lead to better analytics and insights to drive your business

closeup shot of female IT engineer working in monitoring room


Stick with me – I want to make a quick analogy that might help you better appreciate data governance.

Let’s say you are training for a race.  Day one, you may not be in the best physical condition.  Maybe you haven’t even run, walked, or crawled since your school years. But you know this will be good for you if you can just get started. You just have to start putting one foot in front of the other and hit the pavement.

The same is true with data governance in an information-driven business.  Many, if not most, organizations aren’t in the best data governance condition — their muscles are weak and their stamina is poor. Data governance is the conditioning and training of the data analytics and insights capabilities of an organization.  Sure, you can do it without governance, however the results will likely be poor, it will be incredibly difficult and inefficient, and you may even pull a muscle.


What is Data Governance vs Data Science?

Before I offer a definition of data governance, I want to stress that data governance is first and foremost about on-going quality control. When your data is good, secure, and usable, it improves your quality control throughout the company. That’s how important it is.

Data Governance

Now with that said, I define data governance as: the alignment of people, process, and technology to maximize and protect the value of organizational data assets. Sadly, it’s an often-underappreciated discipline. Sad because when it’s done right, businesses can realize the benefit: more accurate, reliable and insightful business intelligence, reporting, analytics, and advanced analytics, such as data science.

Data Science

A quick definition of data science is: a multidisciplinary field that uses scientific methods, processes, algorithms, and systems to extract information and insights from data. Two categories of data science include:

  1. Predictive analytics: forecasting, predictive modeling, and simulation.
  2. Prescriptive analytics: prescriptive actions and recommendations.

Data Governance & Data Science: Where is the Intersection?

Like running that race, the race time (results of predictions and prescriptions) is only going to be as good as the training and preparation (original data quality). In other words, data science doesn’t produce results that can be trusted without good data science governance.


Why You Should Care About Data Governance In Data Science

Unfortunately, many companies don’t treat data like the asset it is. Among the many bad outcomes, this can lead to uncertainty in decision making and an inability to measure and drive business objectives. However, with proper data science governance, your company can:

  • Decrease the cost of doing business.
  • Achieve transparency and alignment across the organization.
  • Get reliable data for traceability and audit needs.
  • Empower employees to concentrate on higher value work.
  • Differentiate, compete, and win in the market with data.

To achieve these benefits, leverage a data governance framework for big data that includes: strategic alignment, directives, organization, measurement, change enablement, and technology. Where you start depends on the size and maturity of your business.


It May Seem Complex, Yet It’s Important to Get Started

If this process seems burdensome, you’re not alone. For many, the steps to create a governance framework appear heavy and academic; measuring the impact of preventing bad data is like determining the value of insurance you never use. Still, it’s worth the effort. And the goal is not perfection. You can train for a personal best time or a spot on the podium; in either case, you will improve your results, as long as you train.

How do you get started?

Consider setting a 90-day plan to:

  • Identify a data domain worth governing.
  • Assign data stewards and create an accountability model.
  • Define KPIs to measure success.
  • Execute some initial governance activities and monitor.
  • Develop a long-term data science governance approach and roadmap.

Making the time and investment to build the right processes, tools, integrations, and organization structures will help you better manage, share, and attain business value from your data. Just get started.


Robert Sunker of RevGen PartnersRobert Sunker is passionate about data and how it can drive value for organizations. He leads the development of offerings and solutions at RevGen that can enable our clients to navigate disruption and thrive.

Subscribe to our Newsletter

Get the latest updates and Insights from RevGen delivered straight to your inbox.