Data Wars: A Governance Saga

I originally posted this on May 4th on LinkedIn but thought I would share here as well...enjoy!

Today I decided to visit the thesaurus and look up the term governance. I was not at all surprised to find many results of synonyms for governance, all of which sounded rigid and overbearing. Terms like bureaucracy, domination, authority, control, law, politics and rule were all listed along with many others. None of these words are very welcoming, nor do any of them invoke ideas of creativity, collaboration, innovation or insight. And yet, all of those things are traits our data journey should include. At no time should a practice be implemented in your organization that stifles the progress of gleaning insight from data. It is no wonder, then that the sheer mention of the term data governance invokes reactions ranging from groans and eye-rolling to straight-up negativity and refusal to engage. In fact, given that today is May the 4^th, I would even argue that most people would liken data governance to The Galactic Empire rather than The Rebel Alliance.

               Queue heavy breathing…

               A long time ago in a corporation far, far away, I got my first glimpse of what could have been considered data governance by most standards. I was confronted with a rigid set of rules and/or guidelines on how and when I could use certain data. I was even provided steps on how to use said data. This was a large organization and although I won’t mention the name, they no longer exist. Just as in another galaxy far away, the oppressive nature of this set of rules that existed for the good of the corporation was put in place to support a belief that was thought to benefit the majority. In reality, it suffocated inventive thinking and deterred many in the organization from exploring their data which in many cases leads to accidental insights. Those are some of the best kind!!

The concept of creating a program to govern data is a noble one and when done correctly can result in the enablement of some very innovative and creative data solutions. Accurate, timely and consistent data will only strengthen the confidence an organization has in their data and the ability of their data team to deliver valuable insights. No organization will fight you on a desire to make this happen, but when we approach leadership with a plan to govern our data, visions of a data police-state begin to take shape. There is a profound fear of the practice right from its very inception and it starts with the name. Data Governance sounds not only boring but overly official and it evokes ideas of rigid structure and seas of red tape. Who can blame anyone for not jumping on board that pleasure cruise?

It took me a long time to get here and that journey consisted of some very heated discussions. Had there been lightsabers at the time I may be just a pile of brown robes urging my son to type this article through whispers of “use the Force, Aidan.” The problem with data governance concepts for me was not only the rigid idea of rules and policy that accompany any successful governance practice, but the required pre-work it seemed was necessary to stand up the practice. I am inherently lazy and don’t want to do anymore documentation than is absolutely necessary. That said, as time went on and more charming personalities prevailed, I began to see the light.  Data governance does not have to be intrusive, oppressive or stifling. It can be something that enables data creativity, promotes value to the organization and aligns with your strategic initiatives.

Using the Force…

Imagine for a moment the difference between the Sith and the Jedi. The Sith are ruled by emotion and control great power. Their immense hatred and anger are an endless resource pool with which to wield powerful lightsabers and should you cross one be ready for some choking and/or shocking. Similarly, some organizations have a wealth of data at their disposal to draw endless insight from. They wield the power of this data with little or no preparation and insights can be tainted by the noise of data quality issues which are just like emotions. Unclean and rife with errors, this data is not aligned with much other than the single source it represents and has the ability to skew conclusions inaccurately in one direction or another. Blinded by the various challenges that have followed this data from source to insight, business decisions can explode forth from an organization like an emotional outburst, taking out whatever is in their path and/or wake. 

Conversely, when one learns the ways of data governance and employs these for good, a different story emerges. This is a story of consistency, calm organization and an absence of the chaos that would otherwise accompany the data. The organization benefits from focus and preparation to reap the rewards of the insights that are created. Data has been profiled for inconsistencies and cleansed of those troubling outliers. Great understanding of data has taken place through the use of taxonomies and ontologies to map data from various sources to one another and provide a consistency that resembles an almost meditative state. Reference data is accurate, informative and is strengthened by a lineage which inspires confidence in the Data Jedi who wields said data. Calm, collected and devoid of the emotional noise that originally tainted this valuable data, the saber of insight swings smoothly and accurately and all strikes are timely. The battle is easily won as all destructive insights are countered by those born of a robust data management process.

“Do or Do Not, There Is No Try” – Yoda

So how do you get started? I say just start. Begin by surveying the landscape. Know your enemy (bad data) and the battlefield (systems and data sources) and document what that looks like. Every warrior needs a map of their battle location and Jedi’s are warriors. The more you know about what you are dealing with the better. Remember that there is no failure in this exercise. Every move you make to establish a data governance practice is a step forward, whether it works or not. If you have read my book on business intelligence then you will know how highly I regard failure as a data point. Failure is one of our greatest tools for learning. Another thing to consider as you implement your program is that data governance is less about technical know-how and skills and more about behavioral or cultural shifts. It is about shifting the mindset of those working with data to seeing the benefit, or rather necessity of a data governance practice. Just as money used to be the most valuable asset of an organization, data has become the new king of that realm. The ability to deftly wield your data assets in an agile manner is the ability to generate revenue much more quickly and reliably than your competitors.

Given the reliance on behavioral shifts, it is important to socialize your data governance efforts with those in your organization. Building up enthusiasm and energy around the data governance program is a big first step to insuring its success. You should be conducting interviews with data owners, business leaders and technical subject matter experts alike. Understand what makes your data tick, what it means and how it is similar to other data in the organization to begin building a data overview for your organization. Once you have a good understanding of what is out there, you can determine which data is most important and which is more supportive of those important elements. What is your reference data? What are the data domains you need to define? That which follows will be steps to prioritize, profile and cleanse these elements and you will be well on your way to being one with The Force!

You can do this. You should do this. May the Force Be With You, Always!