Business glossaries are comprehensive lists of common business vocabulary that enable a common language between business and IT.They provide the foundation for data governance initiatives that brings together cross-functional teams to make interdependent rules or to resolve issues or to provide services to data stakeholders.
By giving business context to technical artifacts (technical metadata), the business glossary makes it possible to catalog, govern and use valuable corporate data in a trusted, consistent and efficient manner.
While it would be hard to find someone to disagree on the value of a business glossary, I’ve heard mixed opinions on the tactical implementation of it–particularly in the context of large enterprises. Specifically, whether or not having just one enterprise business glossary to govern valuable corporate data is practical and timely enough to implement and quickly benefit from.
The two camps we’ve heard about and documented are:
Single Enterprise Business Glossary – For the purists and also smaller, less complex organizations, this approach is best and can eliminate any redundancies and ambiguities. This “boil-the-ocean approach” does require a lot of discipline and time to get the definitions agreed upon and approved. For example, reconciling the varying definitions for the term ‘customer’ may run into many roadblocks especially with Finance, Sales and Marketing.
Multiple Business Glossaries – The multiple glossary approach can add business value sooner than a single enterprise business glossary approach as it allows quicker implementations of smaller business glossaries in a controlled and planned environment. How different glossaries are divided up is dependent on each organization’s needs (business unit, function, product, etc…).This approach may create some manageable redundancies such as the definition for ‘customer’ in the example above, but it allows you to narrow the usage to specific contexts that’s pertinent to the respective users.
While having the ability to categorize terms in hierarchical tiers may also work to answer the ‘one glossary for all’ question, it can potentially result in a glossary that is too deep with business terms becoming too hard to find, causing user frustration.In addition, if security is a concern, administering and managing privileges will become more resource intensive when compared to the multi-glossary approach.
While there is no one-size-fits-all solution, it’s good to have the flexibility of having multiple business glossaries to address all the needs of enterprise data.
How does your organization tackle this issue?
