The five key principles of a data governance framework
27 September, 2022
Principal Data Scientist
at Decision Inc.
Save millions, avoid a data disaster, and accelerate your speed to business value by applying these five principles of a data governance framework. This strategic move will help you get the most value from your data while ensuring your competitive advantage.
A data governance framework is the collection of rules, processes, and role delegations that ensure privacy, structure, and compliance in a company’s enterprise data management. These work in unison to enable data to be well organised and maintained, searchable, and meet all required regulations.
For a business, having a data governance framework in place increases speed to value. If the data is organised, searchable, and validated, organisations can begin analysing it more effectively at a much quicker pace. The framework provides decision-makers with the confidence that they are using the right data at the right time.
Many businesses are aware of the importance of having a data governance framework in place, but few understand how they need to go about setting one up. New laws, like the General Data Protection Regulation (GDPR) have provided much needed impetus for organisations to ensure they are compliant. Not to mention the immense pressure put on organisations to protect customer data. Yet, most companies are still reactive when it come to implementing such a framework.
Navigating the roadblocks
There are, of course, many obstacles between wanting to implement a data governance framework and actually executing one. This is exacerbated when companies try an organisation-wide approach to data governance.
Firstly, this approach requires too many stakeholders to be effective. Invariably, a business might spend years and millions of rands on the project only to abandon it because there seems to be no progress.
You might be interested in our article: Accelerate Growth and Scalability with Strategy-Based Data Governance
Adding to this is the complexity of the technology landscape. With continually evolving technology to contend with, a business never really has a clean slate to start such a project. The company must find ways of integrating data governance into their legacy environments while these environments are undergoing changes and new technologies are being included. This poses great challenges in trying to fix a data system while it is in a constant state of flux.
From a people perspective, a large number of team members must be available to embark on an enterprise-wide initiative and prioritise it. If there is not sufficient buy-in from principals or directors and they do not make this a priority within their teams, it will not be successful.
Similarly, if the data governance framework is not linked to business outcomes, progress will be slow as users do not see the value of time invested, with the risk that the entire project might be abandoned.
Finally, cost plays a major role in the success of a data governance project. If the scope is too big, to begin with, a large business investment is required, and the return on this might be too slow.
To lessen these hurdles, companies should consider a small-scale approach to set up quick wins as cost-effectively as possible. This involves starting with only the critical data first and getting the data governance approach right on this subset before scaling out to the rest of the data.
5 key data governance principles
Getting started with a data governance framework requires a business to apply five key principles.
1. It begins by having a data governance strategy and vision in place. This must address why the business is pursuing the framework and what the thinking is behind it. Strong input from principals and directors is required as this provides clear validation why this project is important.
2. Secondly, there is the people aspect that encompasses the team structure and individual roles to consider. Without key roles in place, the strategy will not progress. People are critical as they will be responsible for implementing the processes on a day-to-day basis to ensure progress happens.
3. The third principle is that of policies. A company must understand what the actual governance policies are that it must abide by. For instance, what is the minimum data quality, who accesses data, where is data stored, for how long is the data stored before it gets deleted, and so on.
4. Fourth is the execution of the data governance framework. This requires the business to put in place all the things related to the policies. For instance, a data catalogue is necessary to organise data and search for it, as well as storage of meta data for the business to know about all the different data sources. Having a glossary of data terms is essential as this ensures everybody in the business understands what is meant by certain phrases and everyone is using the same terminology about the data. Other considerations are data quality metrics, classification of data, profiling the data, and being compliant with data regulations. Essentially, this principle is about making sure the company complies with all the policies it has put in place.
5. The final principle is the enablement of all aspects of the framework. Effectively, the company must ensure that there are processes in place to manage every step of the data lifecycle. As an example, someone must be responsible for profiling any new data, putting security in place to keep the data safe, and keeping track of when to archive and terminate data. A data steward must also be appointed to manage any exceptions, such as if the data quality does not meet the required minimum standard. Fortunately, there are many technology tools available that can help organisations execute on their data governance and manage the process of developing the framework.
Ensuring quality data
Fundamentally, companies must begin the process at the point where data is created. The adage ‘garbage in, garbage out’ applies here. If the data capture is not carried out by someone who understands the importance of having accurate data, then the entire process risks falling flat.
Education is vital in this regard. However, this must be supported by rules to assist users in entering the data correctly. For instance, the data entry system should automatically detect that a phone number must have a certain number of digits in place. These guard rails help to ensure the company has high quality data from the onset.
A data governance framework is a continual process for any company. The business must have an end state in mind to know what it wants its data to achieve. Knowing what good data governance looks like and identifying where the critical issues are, will help a company focus on the relevant data that its key performance indicators are based on. Simply put, it must start with the most critical data first.
Access to good quality data makes it easier for any business to achieve more effective decision-making. Knowing all the data is accurate, clean, and accessible leads to a more productive team and makes it possible to use data as a competitive tool in today’s connected environment.
For more information about our Data Governance Solution:
Decision Inc. Pursuit of Progress
Decision Inc. helps organisations develop and deliver the Technology and Digital Operating Model to support their strategy. We create the capability for clients to innovate and compete by connecting the technologies, platforms and data they need to thrive in the Modern Era.
Decision Inc. brings a deep business and technology understanding and provides functional insights, embedded in the technology platforms of the industry, to help clients make an impact. We help the best businesses in the world to power their business, enable their people and accelerate digital progress.