Many people see the dictionary as a thing of unquestionable authority, with words and meanings that rarely change. Likewise, in corporate culture, business leaders often mistake the business glossary – a collection of data-related terms and definitions used within an organization – as being static, straightforward, and simple to develop and maintain.
Timothy Mulholland, senior of insights and analytics at Salt Flats, dispelled such misconceptions in his recent presentation at DATAVERSITY’s Data Governance & Information Quality conference. With a background in library science and digital asset management, Mulholland shared a wealth of field-tested strategies for successfully creating and governing a business glossary.
The Case for Business Glossaries
The business glossary may be one of the most underestimated elements of Data Management, treated as an afterthought or a necessary ornament for creating a corporate public face. In reality, a well-conceived and managed business glossary defines not only the terms contained within it but the company behind it.
“I think of a business glossary a lot like a torch,” said Mulholland. “It’s a symbol of continuity linking all your different groups together.”
Just as the Olympic flame symbolically unites the ancient and modern games, a business glossary links old and new processes, as well as potentially generations of users, in the life of a company.
Mulholland pointed out how glossaries support a corporation up and down a whole ladder of functions, from bread-and-butter operations to the visionary. Because it sets down nomenclature to be used throughout all areas of the business, an effective business glossary forges a shared understanding between various groups and departments in an organization, thereby facilitating communication and data literacy and preventing data silos.
As the meanings gathered within a glossary are implicit statements of values shared by the whole company, it can act as a sort of ambassador in onboarding new employees by bringing them into the company’s culture. This “speaking the same language” goes beyond improving understanding within the company, and it can even resonate with customers through branding and outreach. Finally, when a business glossary is decisively conceived and wisely governed, it ensures the lasting quality of all data systems, which in turn builds trust in those systems – and the company as a whole.
Dangers, Do’s, and Don’ts of Governing a Business Glossary
According to Mulholland, the biggest pitfall in developing a business glossary is the assumption that it’s the easiest entity to create within Data Management. After all, isn’t a glossary merely a formal transcription of the ideas and terms used in operations every day? But as in the old parable of the three blind men and the elephant, each person in a company may have a different relationship with the concept in question, and they may see – and define – it differently.
As companies increasingly rely on the cloud, seemingly simple terms may get muddied over time: With cloud data, “you have a lot of different people adding and subtracting information,” noted Mulholland. “And if you don’t have proper governance there, it could really erode its quality.”
Bigger companies face an additional risk of informational entropy since they’re built up from standalone solutions tailored for specific departments with their own manners of doing things. Finally, for companies that rely on blockchain technology, the decentralized nature of this new and often ill-defined territory can lead to further erosion of data systems.
Mulholland highlighted a few key guiding principles of governing a business glossary:
- Think of the business glossary as more than a dictionary. Business glossaries are not static lists; they are living, evolving entities that describe processes at an enterprise level rather than local, individualized systems.
- Avoid authoring business glossaries in meetings. Setting down terms should be performed definitively by a Data Governance council rather than through meetings in which the input of too many minds tends to erode clarity.
- Common practice is not necessarily common sense. Shared conventions and cultural divides can create wildly different perceptions of even the most baseline concepts. For example, Mulholland pointed to a multinational outfit whose “Health and Beauty” tag was interpreted by its North American branches to emphasize beauty, while Asian operations focused largely on health.
- Learn to filter out the noise and decide what matters. There can be a fine line between detail-oriented precision and needless nitpicking, so make sure every item warrants inclusion as a purposeful value-add.
- Govern with authority. The strength of a business glossary relies on its management, and a clear definition of where the buck stops in the data chain.
“You might build the greatest business glossary ever,” said Mulholland, “but if you’re not providing ownership, making changes, being proactive, the quality erodes, and it becomes useless. It’s kind of like the Golden Gate Bridge: You paint it from one end to the other, but then you have to start back again and paint it over again.”
The governance and lifeline of a glossary are crucial, but in the end, a glossary can be no better than the strength of its definitions – and bad definitions tend to commit the same pattern of sins. Bad definitions are likely to include highly subjective elements that are open to interpretation and fail to delineate the contexts and departments where the term sees its highest traffic. They are commonly posited in the wrong tense, giving the erroneous impression that the term refers to obsolete practices. Finally, inferior definitions fail to tag the appropriate ownership of the term, making it harder to track and correct when misusages arise.
On the flipside, Mulholland pinpointed a set of criteria that the most effective business glossaries meet:
- First, they are readily scalable for use in any department or context. To have lasting value, a definition needs to be as transparent in the boardroom as it is in conversations mong IT wonks.
- Next, they must be clear, concise, and – especially if the term will live in a highly searched data system – classifiable across contexts.
- Finally, a term must on the one hand be foundational enough to stand the rigors of varied usages across time, but on the other, be proactively conceived so that it can be modified when the realities of the business change. “You need to understand that you’re never done,” said Mulholland, “and that things can and will change.”
To extend Mulholland’s torch metaphor, the fire must be lit, but it also must be maintained. When changes filter into corporate usage without vigilant and responsive governance, the quality of the business glossary will erode over time. When changes are vetted appropriately, data stewards can facilitate growth rather than erosion, allowing data initiatives to mature and ensuring a more-than-solid ROI for the work required.
Because a business glossary is such a significant initiative, governance must be implemented and maintained through a bifocal lens that takes both long- and short-term management goals into account. In the short term, Data Governance leaders must be reflexive, responding to immediate changes with the understanding that not all changes will be permanently adopted. As Mullholland stated, “This means facilitating decision-making, speaking to subject matter experts and different departments to say, ‘This is what we’re building, this is why we’re building, this is what we have.’”
In the long view, Data Governance leaders should be resilient in the form of future-oriented strategic planning that presumes things will change in a way that the company has never done before. Because governance is an ongoing, perpetually corrective process, leaders must be ready to enable these change control processes. Long-term governance of a glossary must be “about engagement and fostering ownership within different groups, getting users to adopt your business glossary as a part of your data catalog,” Mulholland asserted. “And that also means monitoring the quality and the value of your business glossary throughout – making sure it aligns with other data initiatives that arise as your company evolves.” With all this in mind, when the governance council speaks through its business glossary, it will be in a language that the entire business – and its clients – can understand.
Want to learn more about DATAVERSITY’s upcoming events? Check out our current lineup of online and face-to-face conferences here.
Here is the video of the Data Governance & Information Quality Conference presentation:
Image used under license from Shutterstock.com