So Jim and I thought it would be fun to tackle this dilemma in a Star Wars-themed debate across each of our individual blog platforms, with Jim taking the position for “Agility” as the Rebellion and me taking the opposing position for “Bureaucracy” as the Empire.
Note: Yes, most conversations between self-proclaimed data geeks tend to result in Star Wars or Star Trek parallels . . . and I lost the coin toss. Thankfully, I found StarWars.com to help me with some of my rusty Star Wars facts!
Disclaimer: Remember, this is meant to be a true debate format, where Jim and I are intentionally arguing polar opposite positions with full knowledge of the reality that data governance success requires effectively balancing bureaucracy and agility. To view a video related to the debate, click here.
Please take the time to read both of our blog posts, then we encourage your comments below and your votes in our poll.
Let us begin . . .
Hi there – Rob “Darth” Karel here. I’m Chief Terror Officer of Galactic Empire Operations (and active member of the “Dark Side Leadership Council”) and here’s my position: In order for data governance to succeed, organizations must define, implement, and comply with strict cross-enterprise guidelines, policies, and standards. All employees (and citizens) are accountable, and noncompliance will lead to strict penalties. No, not death – at least not for first-time violators. But a bad review or loss of funding may be in your future.
We here at the home office recognize that data governance in and of itself is a cost center – an expensive one at that. So why do it? Because there exists a variety of burning Death Star (aka “burning platform”) issues that executive management must tackle to ensure we succeed in our corporate vision:
Galactic domination, destruction of the Jedi Order, and successful entry as a viable No. 2 competitor to Apple in consumer electronics. Failure is not an option!
Our core business drivers are similar to any business and include:
- Revenue growth. For the Galactic Empire, primarily taxation revenue. Thanks again for sending us 80 percent of your annual income, and yes, it is necessary for the tax collectors to carry blasters.
- Improving operating efficiencies/reducing costs. For example, cloning stormtroopers doesn’t come cheap. DNA sequencing errors can delay security deployment. Worse, cloning mistakes can lead to embarrassing and costly rogue mutant outlaws that distract from core operations.
- Reducing risk. For example, pesky Rebellion uprisings can delay or even destroy Galactic expansion efforts. And even the top executives at the Galactic Empire must comply with reporting and transparency edicts from the Emperor himself. (He just loves perusing his Galactic Domination Weekly report.)
- Strategic differentiation. We constantly have to squelch – I mean, re-educate – citizens that believe the upstart Rebellion’s efforts will make their lives better. Just because they’re fast and agile doesn’t mean they’re right.
All of these objectives for our business are compelling, and for an operation as complex as ours, agility is not an option. We’ve got millions of employees dominating – I mean, serving – tens of billions of Galactic citizens, and even every regional governor must manage complexity within his/her/its own siloed operation.
So how can you get all of these regional governors, all of their employees, and all the citizens they oversee to ensure the data they capture and consume is of the highest quality and trustworthiness? Only with formal and strict implementation of foundational data governance components such as:
- Policy definition and enforcement. Corporate policies around data accountability and ownership, organizational roles and responsibilities, data capture standards, information security and data privacy guidelines, and data access and usage policies are all examples of policies that must be agreed upon and enforced in order to ensure high-quality data and appropriate usage.
- Business process alignment. Data governance must focus on more than just after-the-fact stewardship and data quality validation and mitigation processes. Data governance must influence and improve the upstream processes that capture, update, transform, and/or derive data, as well as the downstream processes and decisions that consume the data. The ability to govern the entire cross-Empire data life cycle from creation to destruction is impossible to accomplish via a small grassroots data governance program.
- Architecture and technology enablers. Yes, we at the Galactic Empire love to build things, ranging from light sabers and droids to Imperial Star Destroyers and Death Stars. But even our tech-happy engineers, designers, clones, and slaves (someone needs to do the dangerous manual labor) recognize that tools are only as good as the value (or destruction) they will enable. So data governance ensures that the technology follows the Emperor’s vision and not the other way around.
Sure, we here at the Galactic Empire would love to be more agile. But we recognize the cold, hard, bitter reality that progress without at least some foundational discipline (call it bureaucracy if you must) is impossible. OCDQ-Wan Harris, representing our primary competition – the overly ambitious Rebellion – will try to confuse you with a shiny bauble of hope. He will imply that data governance doesn’t have to be slow and bureaucratic, and that with a little discipline and a lot of passionate evangelists, you can in fact govern your data effectively. Don’t believe him. On the Dark Side of data governance is the fact that bureaucracy is a necessary evil.
So give yourself to the Dark Side. It’s the only way you can save your data governance program.
Data governance can be designed to achieve this same level of intuition. When a new product or service is being considered people immediately begin to outline what information they will need to make the product as success. Where will this information come from what are the qualities of this data (not quality but qualities). This knowledge is captured in all conversations. Of course what is needed are some tools and techniques to facilitate this process like in project management where a simple tool such as a spreadsheet can be used to create a project plan. You don't need Microsoft Project to create a plan. Sometimes simple is better. Likewise other techniques can be deployed to ensure governance in the design development deployment and production of data.
My other suggestion for ensuring that the bureaucracy does not become a permanent burden is to consider the bureaucracy as having a limited life span. With one client I suggested that the bureaucracy for data governance be established for a period of up to 36 months and then summarily destroyed. If by that time the data governance practices have not been successfully deployed and become intuitive abandon ship. Data governance has failed. No need to maintain the facade anymore. At that point celebrate your success or failure as the case may be and try something else. Most people I suggest this too are shocked. But in today's world if a major change like data governance cannot be accomplished in less than 36 months it's a waste of time. That's not agile but it does limit the costs and impacts to a finite degree.
And with regards to Agile data governance I suggest give up. You can pretend to be agile by showing how you've addressed the low hanging fruit but as the tree begins to topple over many will realize your agility only provides you with the opportunity to run away from the impending doom of the tree toppling. Leave agile in the world of fiction.