Data Glossary

What is a Data Glossary?

A data glossary (or business glossary) defines and organizes the different business terms used to describe data within an organization.

Stored in an easily accessible repository, it provides a common language or vocabulary so that everyone understands what particular data or terms cover, avoiding confusion and incompatibility and driving consistency. For example, there is a unified understanding of what a “customer” means, so that data from different departments (sales, marketing, operations) is consistent and interoperable. Data glossaries are created and managed by the business, and updated in line with its needs.

Without a unified data glossary, data governance and sharing will be difficult. For example, data producers may enter incorrect data in a field while data consumers may believe that data in a field represents one concept, when it actually refers to something similar, but different.

What are the differences between a Data Glossary and a Data Dictionary?

It is important to understand the difference between data glossaries and data dictionaries. The concepts behind both are similar and they are both essential to effective data governance.

The biggest difference is that while a data dictionary is technical, and is owned and updated by IT, a data glossary comprises business terms, and is owned and updated by the business. It is therefore possible to have multiple data dictionaries for different data sources, but there should only ever be one data glossary across the entire organization.

A data dictionary will define the technical terms and specifications in a dataset – so for example how many characters are allowed in a particular column type or what format a field should be in. This ensures accuracy and consistency when collecting, entering, managing, or using data. By contrast, a data glossary will define what the contents of each field actually mean at a business level.

Read our entry on what a data dictionary is to learn more.

What are the benefits of a Data Glossary?

• A data glossary provides a single source of truth around how your data is described and categorized. It delivers five key benefits:
• It improves trust in data as users are confident that terms are consistent, helping them understand and reuse data effectively. This drives greater data democratization through self-service access to data.
• It safeguards data governance by standardizing terms and definitions, scaling them across the organization. It supports consistent overall policies and processes when using data.
• It enables easier search and data discovery, through consistent terms that make it obvious what a data asset contains. Users can quickly find and access the data they need to work with.
• It allows greater collaboration as users from different departments can work together on the same data assets, using a common language, breaking down departmental silos and enabling innovation.
• It drives greater transparency across the organization as reports and analysis are based on the same terms. There is no inconsistency between departments or business systems, helping decision-making and overall reporting.

How can you create a Data Glossary?

Creating an effective data glossary is a multi-stage process that needs to span the business:

• Start by assembling business users of data from across the organization. Ideally, these should be from from your data governance council, if you have one in place.
• Collect all the terms they and their teams use to describe data, including definitions of what they actually mean and cover.
• Look at how you can support the process by using existing industry standard glossaries – these both give helpful common definitions and ensure your data glossary will be compatible across your wider ecosystem.
• Compare all of these terms, eradicate any overlap, and publish your data glossary where it can be accessed by all. If, for valid business reasons, different departments use the same term to refer to different concepts, rename one or all of them to avoid confusion.
• Apply the data glossary terms to your existing data assets by working with data owners. Make sure that it is easy to link from data assets to relevant data glossary entries.
• Keep your glossary updated as new data and terms enter your business lexicon or new systems are added.