Data Governance and Digital Transformation: Two Sides of the Same Coin
Why not “boil the ocean”?
One of my colleagues, a senior data modeler, challenged me upon reading Part 2 of A Framework for Defining the Scope of Data Governance Strategy Projects to be more explicit about why, when developing a corporate data governance strategy, it's a bad idea to try from the outset to “boil the ocean.”
I was recommending against an overly-ambitious effort to identify, document, and model all the organization’s data and associated systems and processes. My preference is to focus initially on engaging the organization’s data to help solve a specific high-value and immediate problem.
Avoiding paralysis by analysis
Part of my caution about ocean-boiling is related to fear of getting trapped in an expensive and time-consuming “paralysis by analysis” exercise. As a longtime consultant and project manager I have seen over-ambitious development efforts collapse under their own weight before any value was derived by the client.
On the other hand, it's also not a good idea to rush in and grab the first “low-hanging fruit” data governance challenge without also considering longer-term implications related to standardization, security, privacy, data stewardship, and program sustainability.
Balancing strategic and short term interests
How you balance the strategic with the short term will depend on what you want to accomplish with better data, for example:
- Are you trying to reduce the amount of time it takes to organize and create real time status reports and thereby speed up how important decisions are addressed?
- Are you trying to avoid miscommunication and mistakes due to lack of data or terminology standardization?
- Are you trying to improve predictions by maintaining cleaner and higher quality data?
- Are you encouraging innovative data use by staff and management by providing better tools and support?
- Are you as a manager recognizing a need to clearly specify and manage data governance responsibilities?
Each of these can be profiled in terms of “what needs to be done now” versus “what can be addressed in the future.” For example, even when we adopt an agile framework to manage early delivery of improved data analytics, we also need to be considering how to expand our approach when, based on early successes, we need to expand scope to other problem areas.
The last point above - - clearly specified responsibilities - - is especially important regarding how data are standardized, transformed, stored, maintained, or analyzed.
Defining management responsibilities
Defining such management responsibilities at the enterprise level can be a daunting task, especially when siloed legacy systems and processes are involved. That's why it makes so much sense, even in an enterprise-level data governance strategy, to first focus on well-defined and important problems that can be improved through better data, analytics, and business processes.
Which brings us to a discussion of the relationship between “data governance” and what some people refer to as “digital transformation.”
Both involve more than technology
In general, I don't think one can be successful without the other. They both have the potential for touching on and impacting every aspect of the organization. Yet, each of these terms really does mean a lot of things to different people and can impact – and be impacted by -- a range of concerns including data quality, data modeling, security, privacy, data literacy, master data management, and data stewardship. Also, each involves much more than technology.
Given how data are critical to the operation of any organization, any effort to control and improve data usefulness is bound to touch on a wide range of systems and processes. Yet, a too-ambitious or top-down approach to governance or digital transformation, as suggested above, is bound to run into challenges, starting with the most likely challenge of all: the need to accommodate constant change, which can come in many forms:
- Change in government regulations impacting tax accounting or financial reporting practices.
- Organizational change brought about by reorganization, acquisition, or divestiture.
- Increased scrutiny on data governance generated by a severe data security breach.
- Increasing competition in key markets requiring new and better data.
- Staffing or management changes that directly impact data governance or digital transformation efforts.
We all know that change happens. No matter if what you're doing is “strategic” or related to day to day operations, you must be ready to accommodate change.
That's just part of management and all the more reason to have both short and long-term goals in mind when embarking on targeted or prototype data governance or digital transformation efforts. If you really think through any data governance effort that’s intended to have long-term impacts, you can't avoid broader “digital transformation” efforts. Eventually both system and business processes will be impacted by a push to improve the impact of data on efficiency, decision making, and the organization's bottom line.
Prototype to program
How management and staff adapt to these changes cannot be left to chance. Even if an initial foray into improving data governance attacks a well-defined and specific problem, it’s important not to underestimate the need for change and transformation when it comes time to turn these efforts into an ongoing program.
You need to be prepared when asked to expand your efforts to other areas of the organization. Documentation of data and metadata, along with an understanding of related systems and business processes, will be invaluable when working through an expanded data governance scope; more on this in Part 3.