Most data governance implementations are conducted through implementing roles, committees, policies, and maybe the odd metadata repository or data dictionary. While this approach may interest the data crowd, it neither engages with nor brings much value to the day-to-day concerns of most business users.
The main challenges concerning data for the average business user are the lack of overall visibility and understanding. They face issues such as:
- Where should I source a given data element from?
- Are fields A and B used for the same purpose across the different systems?
- How does my data stay in sync?
- What is the provenance of my data?
Think about your organization for a moment. Chances are there is no go-to point where you can quickly answer questions like these – at least not yet.
Data Landscape
Data does not occur in a vacuum. Nor is data intrinsically complex. It is the environment and business reality in which data operates that makes it challenging.
The typical data landscape at a large organization is fragmented across a few hundred systems, thousands of Excel and Access data sources and several thousand business processes. If only you could make this fragmented, largely obscured landscape more visible and better understood as a whole, then you could begin to exploit data better and reduce your management costs. It turns out that you can.
Shared Understanding
You build a better data landscape by leveraging and incentivizing business users. The key to doing better with data is creating an improved and shared understanding of data and its usage beyond the specialist data crowd. Shared understanding brings people closer together, builds engagement, fosters collaboration, and lowers barriers to changing behaviors.
Imagine data, its connections, usage, dependencies, and broad business context being described in business-understandable terms and recorded in a single, easily accessible knowledge repository for all to utilize and contribute to. This isn’t such an insurmountable task, as many committed and imaginative groups are proving every day.
Collaboration Universally Proven
In the consumer world, millions of people are collaborating together on valuable open source projects, wikis, networking groups and community portals without ever meeting or even speaking to one another. Meanwhile the corporate world is falling behind. Colleagues working for the same firm continue to have regular face-to-face meetings, happy to write up data requirements in Word. They’ll follow up with some email ping pong, wrestle with “track changes” and then bury the document on a project network drive. How can we hope to leverage this slow, uncertain and fragmented approach or form an integrated, accessible and insightful whole?
Connect Corporate Knowledge
Up-to-date knowledge on data and its usage is present in your organization right now. But this knowledge is segmented and trapped in the heads of individuals across many functions and disciplines. This knowledge needs to be documented and shared across the business, rather than staying with – and eventually leaving with – the people who have it. What’s needed is a framework and platform to liberate, collate and connect understanding.
Understanding requires context. The business context of data and how it relates to business processes, reports, projects, regulations, and so on, can be built up collaboratively by repeated small contributions from the business community within the organization. Matching the distributed nature of data and its usage with a distributed and collaborative approach to building shared understanding lets us harness the knowledge which is already there and paid for.
In return, the community gets an increasingly rich and comprehensive living body of knowledge which enables people to make sound, joint, informed decisions. With a collaborative, systemic approach to data governance, knowledge about data is captured, made available, and continually enriched.
Take Action
To help you implement some of the ideas in this article, I’ve posed some questions below which I use to help organizations gauge potential areas for improvement. Feel free to copy and adapt them to your situation.
- What data knowledge assets exist in your organization; e.g. data dictionaries, business glossaries, logical data models, data quality registers? How joined-up and widely adopted are they?
- What other business knowledge assets exist; e.g. process modelling tools, policy directories, change project registers? Do these assets enable business users to construct a connected data view?
- Do data owners and other responsible parties in your organization feel they have a sufficiently integrated understanding of data and its usage? Is there an integrated view of critical data elements and the business contexts in which they are critical?
- How engaged is the business around data? Does the business have a standard way of communicating data requirements?
Ownership Critical
As part of a data governance initiative, the business community needs to take ownership of its data. Having people take ownership and stay engaged on the basis of a shared understanding is not only more productive, it also enables them to make far more informed and responsible decisions. Our business is data, but let’s remember our data is about the business.
