Analyzing the most common mistakes companies make when implementing Data Governance

Experience is simply the name we give our mistakes.
                                                                        Oscar Wilde

“The 9 biggest mistakes that companies make when implementing data governance (and how to avoid them all)”, is excellent report written by Nicola Askham that allowed me to reflect on these points and their solution, making a re-categorization of these mistakes and providing some more comments based on our experience in this type of initiatives.

These mistakes could be grouped in 3 categories.
Need for the initiative: what drives the implementation of a Data Governance initiative.
Leadership of the initiative: who are the ones that promote and carry out the implementation of the Data Governance initiative.
Implementation of the initiative: what is the approach taken to carry out the implementation of the initiative.

Need for the initiative

There are usually two major needs for which an organization plans to start a Data Governance initiative: either by complying with regulations imposed externally (a regulatory body) or internally (audit observations), or by the acquisition of technological tools that include capabilities to support projects of data quality, data glossary, security, etc. These needs can lead to some mistakes when implementing the initiative that can lead to failure and loss of an excellent opportunity to have the organization’s attention and resources on these issues.

In the first case, the need to fulfill a requirement can lead to the Principle of least effort to satisfy the regulations or the checklist approach of executing isolated tasks that fulfill the request. In these cases, what happens is that there is no access to the benefits in the medium and long term that come with the correct governance of data in an organization.

They tend to become tedious tasks that wear down the concept of Data Governance, which causes that the change at the cultural level of the organization, necessary to improve its practices, is not achieved. To avoid this, it is crucial no to forget that “every cloud has a silver lining” and to try to satisfy the regulations obtaining benefits for the business, taking advantage of the opportunity as a facilitator of facilitating cases that allow the organization to begin introducing the Data Government in it.

In the second case, if the initiative starts with the impulse to use the tools that are available, there are several risks to face. The first is to build the initiative around a technological solution, which will not allow the true cultural change of the organization at the business level, since it is more difficult to involve it when the Data Government is only thought of as a technological solution.

Another risk is to take the implementation of a Data Governance initiative as a project and not as a comprehensive process that affects people, processes and technologies. For example, if it is thought that Data Governance is implemented when implementing a data quality project for the improvement of any problem of the organization, there is a risk that the construction effort will become obsolete in the medium term, since once the specific project is finished, the solutions become obsolete due to lack of maintenance.

The clearest example is seen with glossaries or data models. Many times, the effort is made to create them; but then, if the organization has not made the change in people and in the processes to incorporate such practice, glossaries become difficult to maintain, no matter how much technology is available.

To avoid these problems, it is recommended to never start with the tools, but to understand what you want to do and why, in terms of conceptual frame work and strategy regarding Data Governance. After that, if the tools are already available to help achieve the objectives, faster results can be obtained.

Leadership of the initiative

Another mistake that usually happens is related to who leads the implementation of the Data Government in the organization. This point is one of the most difficult since, in general, the first ones that begin to promote these issues within an organization are from the IT area because they know the data problems they have and how they impact on their daily tasks, or for the knowledge of the tools that we mentioned in the previous point.

The problem here is that we have to make the organization understand that IT, though it is an actor with a dominant role in what has to do with support in the implementation of technological solutions and architecture, cannot lead the initiative if you want to succeed in a Data Governance implementation. They must be the business areas, which are the owners of the data and who must be responsible for seeing the data as an asset of the organization ensuring its quality and safety from its creation to its destruction.

Think of data quality, although IT can identify and correct data based on what the business areas can tell you, but the processes of how data is captured are changed, quality will never improve.

The fact that the business does not get involved can also lead to Data Government not being within the strategic objectives of the organization, which can cause Data Government to lose relevance and failure of its implementation. The way to avoid these problems is to first have an internal sponsor (person interested in the implementation of data governance) that has experience over the business areas, raise awareness through workshops on responsibilities and benefits of data governance for the business and present results that show that the implementation of these practices help achieve the strategic objectives of the organization.

Way of Implementation

Another key factor to have a successful implementation of Data Governance is the way in which said initiative is implemented.

The first option, when the organization is convinced that they must incorporate Data Governance, is a Big Bang approach. With said approach the risk is: wanting to incorporate numerous changes at the same time, that people feel overloaded and that the results take time to be noticed.

In practice, with this type of approach, successes it is not possible because this requires a cultural change that requires the conviction of people through the achievement of results and that they can change their behavior in order to adapt the processes necessary to maintain the benefits over time.

The suggestion to avoid these problems is that of an iterative approach through facilitating cases with quick wins for the organization, which allow us to show tangible results to convince and thus carry out the necessary organizational change. Another way is to make an evaluation of the maturity of the organization, in terms of Data Governance, to determine the level of training the organization has in order to incorporate the different aspects of Data Governance.

These will allow us to work at different levels at the same time. Prioritize the problems that allow us to obtain results and where the organization is able or feels safe to move forward, while preparing the organization to work further on the other aspects.

In conclusion, there are several known mistakes that can be avoided if we focus on the three aspects, detailed throughout my publications, taking into account the recommendations provided. The first step is always the hardest and no implementation of this type of initiatives, because of the type of change it requires, is exempt from encountering many pitfalls, but there is nothing like achieving results to gain momentum and achieve organizational change that can maintain the achievement of the objectives over time.

Eng. Gustavo Mesa    @gmesahaisburu

Business Analytics & Information Management specialist

You also could read this articles:

Data Governance boosts Analytics and Big data

Data Governance in Practice

Leave a Reply

Your email address will not be published. Required fields are marked *