The “Right” Data Governance Approach

In the first two blogs of the series, I wrote about staying non-invasive in your approach to data governance, and how data governance should not threaten work culture. In this blog, I will address how to select and adopt an approach to data governance that best suits your organization.

Data Governance – Right from Wrong

This series started with a brief description of the three approaches to data governance. I mentioned the command-and-control approach that dictates that thou shalt govern your data. I wrote about the traditional approach that hopes that people gravitate to the program you have designed and deployed. And I laid out the non-invasive approach that starts with the premise that you are already governing your data, but are likely doing so informally, which leads to inefficiency and ineffectiveness. These approaches are detailed in an earlier Alation blog titled Data Governance: Stay Non-Invasive in Your Approach.I often get the question, “What is the right data governance approach for my organization to take?” People expect me to say that the non-invasive approach is always the answer, but the truth is that is not always the case. The best approach is the approach that will activate your program, engage the data stewards, and lead to the demonstration of business value. Without business value, the opportunity may be short-lived. Management has to see the value of governing data formally for the program to be allowed to sustain and operate effectively.

Let’s review the features of each of the approaches and consider what it will take to activate the program for your organization. But before I discuss how to activate your program, however, we need to define what it means for a program to be active.

Active Data Governance and the Bill of “Rights”

The Oxford Dictionary describes the word “active” as “characterized by busy or lively activity.” To me, active means fulfilling what I refer to as the Data Governance Bill of “Rights.” The Bill of “Rights”, with the word “rights” in quotations, means getting the right people, involved at the right time in the right way, using the right data to make the right decisions, leading to the right solution.

The Bill of “Rights” succinctly states what makes a data governance program active: getting people to become busy or lively when it comes engaging in data governance. I will address considerations for activating people toward the end of this blog.

Selecting What is Right

Organizations typically follow the approach that imitates their existing organizational culture. The problem with matching the approach to the culture lies in accurately assessing the present culture and then distinguishing the path that will be met with the least resistance and has the greatest potential for success. It may surprise you to learn that there is not a single “right” approach that can be adopted by every organization. That is why it is important to become educated in the similarities and differences in the approaches.

The similarity is singular as all three approaches have the same goal in mind. The goal of data governance is to execute and enforce authority over the management of data. The approach you take matters less than the goal. The goal should always be to formalize accountability for data such that people become active in assuring that the definition, production, and use of data follows the rules and provides a path toward appropriate decision-making that directly benefits the organization.

The command-and-control approach is the most invasive. In this approach, roles are assigned as being new and all processes are described as being new processes. Command-and-control sets a demanding tone that people will follow the rules, people will be given new titles, and that people’s progress will be slowed by data governance, and tools are purchased early as a large factor in the solution.

The traditional approach is less invasive as the path to governance is clearly defined, and it is expected that people will stay on course. In this approach, people are identified into roles, there is a single governing process, people are told that they should follow the rules, return is demonstrated through improvements in data quality, and existing tools are leveraged first as the means to flush out true requirements for technology that will enable the program in the future. This approach can be slow and passive in achieving results if people don’t buy in and the governing process is tedious.

The non-invasive approach is the least invasive to the work culture and the present levels of activity associated with governing data. In this approach, people are assigned formal roles based on their present relationship to the data, governance is applied to existing and new processes, the discipline is communicated as something that is already taking place, metrics are effectively mapped to business outcomes through key performance indicators (KPIs, success is associated with operational efficiency and effectiveness, and tools are utilized to enable the success of the program.

It is common that organizations render governance solutions that are a hybrid of the three models described above. Often components of the program are already in place: management states that a governance solution must be put in place, people have been assigned to new roles, tools have been acquired, and/or the assessment has been completed and the plan has been specified. In these scenarios, it is up to the people administering the program to gain the confidence of the organization, which in turn results in people becoming actively accountable for defining, producing, and using data. This becomes the best method for moving from a passive to active data governance program.

Activating People Toward Formal Data Governance

Alation describes active data governance as a people-first, business-led, and value-driven method for governing data. Active data governance focuses on the needs of the business, delivering trusted data, and driving adoption by embedding governance in people’s day-to-day business workflows and activities.

An active program requires accessible data documentation that enables people and teams to efficiently and effectively locate the data they require, confidently understand the data that is part of their daily routine, and coordinate and collaborate actions across the organization that result in data-driven decision making. People, as the stewards of the data, lie at the heart and soul of successful data governance. The data will not govern itself and requires that people of the organization become actively engaged in improving the definition, production, and usage of data.

Here are examples of what is meant by “activating people”:

• Activating people requires understanding their existing relationship with data and formalizing this relationship around roles and responsibilities.

• Activating people requires connecting policies to meaningful actions in their day-to-day responsibility like data consumption and analysis in contrast to disconnected governing policies and processes.

• Activating people requires that staff recognize themselves as stewards of the data and become formally responsible for managing data assets for the organization.

• Activating people requires that they be given the tools and know-how to benefit from the data and address opportunities to leverage and improve the value that is achieved through the data.

Active data governance programs demonstrate value by getting people to accept accountability for the data as part of their job and drive opportunity to improve organizational and personal effectiveness.

Summary

Selecting the right approach to data governance must focus on what gives your organization the best chance to be accepted and operationalized, activate your data stewards and demonstrate business value. In the next blog of the series, I will address how everybody in the organization can be a data steward if they are held formally accountable for their relationship to the data. Your best chance to implement a successful data governance program that covers the entire organization includes enabling everybody as a data steward and activating them with the appropriate information to help them with their job function.