Business glossaries provide a business data term list, which is an authoritative vocabulary that promotes a common understanding between stakeholders in an organization. The best implementations support a collaborative approach to managing information, such as descriptions of business terms, related source and reference data, contacts, and relationships between terms and processes.
How business glossaries can improve analytics
There are a number of ways that business glossaries make analytics better. Let's take a look at what I consider the top 5:
- Data preparation. Properly prepared data collects a lot of technical metadata describing the form and content of data that a business glossary links to terminology business users will understand. It's all essential to understanding analytical results.
- Data lineage. Lineage involves knowing not just where data came from but also whether it came directly from a source or passed through a process, system or application before analysis. The data lineage aspects of a business glossary supports analytics by providing an audit trail for post-decision defenses and post-outcome accountability. Regulatory compliance requires both.
- Data quality. Artificial intelligence (AI) applications are driven by analytics that require massive amounts of high-quality data to achieve their goals. A business glossary improves the quality of data by augmenting data with the tacit knowledge of users who understand what the 1's and 0's mean in the ABCs of the business.
- Data governance provides the guiding principles and context-specific policies that frame data-related processes and procedures. A business glossary is both a guide and a frame for the enterprise’s common conceptual understanding of its data and its many uses. This is especially important in light of how data feeds analytics. Governance as it relates to a glossary entails defining subject areas based on known business and IT terms, as well as clearly assigning those terms to a central definition. A business glossary is also one of the intersections in the convergence of data and model governance.
- Operationalizing analytics, such as operationalizing the model life cycle, is highly dependent on the foundation provided by a business glossary. For example, a business glossary aids in properly assessing the data sources and variables available for use in models. Further, it identifies the business metrics affected by the model. A business glossary also helps bridge the conceptual gap between analysis and insight.