Kick-start your data governance program with a business glossary
26 October 2020
No one likes misunderstandings. We've all been in a "the Italian man who went to Malta" situation before. Annoyingly enough, this situation happens in a lot of businesses too. Many discussions between departments have the same root challenge: different people speaking different 'languages'. In what's next, we will set the stage to take on this challenge.
The “why is my data different” question
Say you're working at an airport. You're looking into a report on flight delays from the operations department, but you're confused. The report you're looking at seems to tell a very different story than the data you work with. Delay times in the report differ from your data by more than 20 minutes! What happened here? A lot of phone calls, meetings and emails later it becomes clear: the people in operations have a different understanding to what the word "flight duration" means. Your data shows flight durations measured from the moment the plane leaves the ground at departure, to the moment its wheels touch the ground again. The guys at operations define this differently; for them the flight starts from the moment all passengers are onboarded. Quite a big difference.
Now, it's not only the aviation industry that is affected by this kind of issue. It happens everywhere. In banking, lots of discussions arise on the definition of a customer. Is this someone who has an account? Or is an insurance enough? Can a customer be a company, or are we only talking about physical persons? Are prospective clients considered to be a customer as well?
These kinds of issues lead to a lot of inefficiencies in businesses, but also to a lack of cooperation, a lack of trust in data, and maybe even wrong business decisions (imagine if the aforementioned airline company believed its delays had shot up, and acted upon this misunderstanding...).
Luckily, there's a way out: you can centralize and align organizational specific language in a business glossary.
What is a business glossary?
A business glossary is a centralized inventory of business terms. As the name already says, it contains definitions of terms used by the business, agreed upon everyone that uses them. It makes sure that the different taxonomies that are used in an organisation are clearly defined. It acts as a single source of truth for these definitions. Each one of these definitions is also owned by a specific person, who can act as a single point of contact for this definition, should questions arise.
Lets go back to the airport. Now imagine there was a business glossary that you could have consulted. Instead of being puzzled with the difference in data you're seeing, you would have just consulted the glossary and seen what "flight duration" is defined as in the company. Confusion over! You now know that your data shows something different than what is agreed upon under "flight duration". And you know to take this into account when analysing the data. No phone calls, no emailing, no alignment meetings with Operations. Easy as that.
Business glossary in data governance
A business glossary is crucial in any data governance strategy, not only for the value to the company discussed already, but also because it is the starting point for a lot of companies' data governance efforts. Now that you've made a business glossary of definitions, you'll surely want to know more. These business terms, do they have synonyms or acronyms that I need to know of (we all know we soon start talking about BISOs, KPIs, ROI and stuff like that very quickly in a business)?
I might also want to know: what are the allowed values this business term can have? Say you're keeping data about countries, are we then talking about US, United States or USA? Reference data is a next step that can bring a lot of value to your business glossary. And even more, do these business terms represent actual columns and tables in databases? If so, where can I find the data on this business term? These are important questions that can be triggered by starting with a business glossary.
How to create a business glossary?
This all sounds easy, right? Implement a glossary, problems solved... Although it's an excellent start, there is a bit more to it. You'll probably encounter some challenges when designing this glossary, but they shouldn’t stop you from getting started.
Take the following questions into account when starting a business glossary:
- How do we choose the right granularity of business terms?
- Who decides what a business term's definition is?
- How can we approve definitions?
- Other enterprises/departments are using different terms for the same data, how should we tackle these 'synonyms'?
- Who is the owner of the business term?
The ownership of the business glossary is very important. There's nothing useful about a glossary that's not being maintained: then it's just another list stowed away somewhere in the archives. This is something that needs to live in the company. It is absolutely crucial that the ownership of each business term is clearly defined. This person, the owner, will be able to tell anyone who needs it, more about this particular business term. Also, he's the one who has a final say in the actual validation of a term.
Without this ownership and accountability, a lack of trust in these definitions will remain, leading to a lack of adoption of this glossary, and this is to be avoided at all costs. And this is often a challenge in businesses. This sounds like common sense right? Well... based on our experience we could guarantee you that the vast majority of businesses struggle with this. Take your time to design and implement processes that ensure the proper management of your business glossary.
Need help creating your own glossary?
We’ve encountered the challenges that go with implementing a business glossary at clients in a wide variety of sectors. In our next blogpost we will share our key takeaways on building a business glossary.
If you find yourself stuck or don’t know where to start, don’t hesitate to reach out to me or the team. We’re here to help.