You need a Business Glossary! We bet that you have heard that one before. It makes sense – employees need a glossary to explain exactly what (invisible) data is, where it resides, who can access it, and what it does. Sounds great, doesn’t it? We agree that it sounds great, but why are so many Business Glossaries ignored or reviled? We think we know the reasons.
“For the past several years we have watched with increasing dismay at the increasing chasm between information technology (IT) groups and their business counterparts.” –Harvard Business Review
Business Glossaries, these days, seem to have been written by anybody but businesspeople. They are often the product of data architects, database administrators and other IT specialists who may be experts in their respective fields, but who come to data from such a different perspective, that they speak their own language. When thinking about data, they imagine databases, technical requirements, schematics and the optimization of such things. They are calculating the amount of storage space for a given field or they are thinking about how long the Extract Transform Load process will take. They are comfortable with SQL statements and C++ scripts.
Are you still there? Still awake? We know that technical language can obfuscate what should be clear, practical and useful. The Business Glossary 2.0 should begin by answering two simple questions: what data do we have and how can I use it. We believe that the Business Glossary needs to be written for businesspeople in the language of business. The technical specifications should be filled in around the business use and purpose. Usually, the process is reversed – the business definition is appended to the technical specifications, where it is never to be seen or heard from again.
What we advocate goes beyond a better Metadata Repository. We think that the Business Glossary 2.0 should be the central hub from which data is managed. It should be the master of the data domain, and not a throwaway by-product.
Who is the ‘Business’ anyway?
Current Business Glossaries are usually developed in the following process: In a good-faith effort to bridge the IT-Business gap, IT attempts to get useful feedback and guidance from ‘the business’. Unfortunately, all too often, this process has turned into a perfunctory chore – get approval from ‘the business’ in order to check it off the list. The task is so onerous that ‘the business’ and IT have come up with an entire role devoted to it. This new layer of bureaucracy has the sole purpose of approving IT work from the so-called business perspective. Unfortunately, the person in this role is not a sales executive who needs better customer lists or a financial analyst who wants up-to-date sales numbers or a data scientist who needs reliable data for forecast models. Instead, ‘the business’ is represented by a professional intermediary who isn’t really embedded in the weeds of business, and may never have been.
In an organization where this intermediary role does not exist, then the process of getting ‘the business’ perspective consists of a short interview or series of emails that aim to define terms and capture use-cases. While this process sounds good, it tends to be an exercise in futility. If the email chain does not get entirely bogged down, each side wants to expend the minimal effort in order to satisfy the requirements. They will compromise on a document that nobody really likes in order to end the process, and congratulate each other on finishing the job.
From our perspective, this well-intended process misses the target. Data should always serve the business, therefore the Business Glossary should serve the business (by that we mean actual business people, current business processes, etc.). To be clear, the goal is not to create a Data Dictionary that satisfies IT and their checklist. Rather, the business perspective will be the center around which all other considerations revolve. For this to happen, we advocate for anchoring the Business Glossary around the Business Data Element.
What is a Business Data Element?
The Business Data Element (BDE) is the fundamental conceptual unit of data from the business perspective. It encompasses how a non-technically-oriented person in the organization thinks about data. Take the BDE ‘Address’: For the businessperson, an address is the mailing or ‘street’ address of a person or an entity. It typically consists of a street name and number, an apartment or suite number, a municipality, a state and a zip code. The practical value is that if you write the address on a letter and affix appropriate postage, you will have a reasonable expectation that it will be delivered to the intended party. That is the use-case and value of ‘Address’. To keep matters simple, we won’t introduce the potentially vital distinction between ‘Shipping Address’ and ‘Mailing Address’.
From the IT perspective, ‘Address’ consists of 4 to 6 related pieces of data that must be individually stored, and also related. The seemingly simple BDE ‘Address’ illustrates some of the complexity that is involved in maintaining data. ‘Address’ may seem obvious to the business person, but represents a cluster of concepts to the IT specialist. This complexity is compounded when internal systems are updated or replaced, or when new systems are integrated, as often occurs in a Merger or Acquisition. Tracking where and how data is stored becomes a significant challenge and, if this information is not adequately known and documented, it becomes a nightmare.
We need to appreciate the complexity of the task and the level of IT skill required to manage data, but that does not mean that IT staff should be in control of writing the Business Glossary.
The Business Glossary needs to be written by the business. But who wants to take on such a task? The answer is everyone and no one. The Business Glossary needs to leverage the hive mind in order to extract and document the perspectives of those who live and breathe closest to the data. The Business Glossary should offer a Wikipedia-like service where the knowledge of all, whether extensive or limited, is integrated into a legible, useful, practical format. Depending on the sensitivity of the BDE, this can be truly open-source, or can require gatekeepers to ensure legally compliant and reliable information. To create this, you will need real-time communication tools, preferably integrated into the Business Glossary. You will also need a functional process that is efficient, repeatable and reliable.
We have pioneered a process for extracting this tribal knowledge from the minds of those in the organization called Controlled Chaos. In short, it employs fast-paced collaborative sessions in order to initiate interaction, structure discussion and disagreement, and distill useful definitions and other knowledge. This information becomes metadata – the data attached to BDEs that enables efficient governance.
In addition to capturing insights into the BDE itself, the Business Glossary 2.0 should capture the data’s relationship to business processes and visualize this relationship. This can be done by cataloging how each business process creates, changes or draws on particular BDEs. Then you can map the BDE to the business process, and vice-versa in order grasp how and where your business depends on data, and how you can improve the data supply chain. These visualizations give you tremendous power to improve efficiency, make decisions and manage data – in other words, to reduce your Data Debt. Data debt is the cost of not collecting, storing, moving and using data efficiently.
The Business Glossary 2.0 becomes the center piece of data governance. IT and data governance must start from the Business Glossary. Changes to the system must stem from changes to the glossary. This reverses the usual process, and puts the business in charge of data.
The Business Glossary 2.0 brings to fruition the promise of the original Business Glossary. Rather than being a tack-on solution, the Business Glossary 2.0 effectively charts your entire business. Data is the glue that connects people, process and technology. The Business Glossary 2.0 thus gives you unprecedented perspective on how your business is running (effectively or otherwise). Creating such a powerful Business Glossary 2.0 is, alas, not a perfunctory task. It requires an innovative technological solution and a smooth process to gather the information. The upside is tremendous. The Business Glossary 2.0 can help you minimize your data debt, and get your data working optimally for your organization. Seize and maintain control of your data.