Metadata is information about data within the enterprise. Data is of great value, but only if it is managed properly. Metadata management is key for building a data driven organization. Without a proper metadata management system, business user would not know how to use enterprise data for any kind of business use case.
Metadata can be mainly bucketed into four categories:
Business metadata includes the definition of data elements, the context of data within the business, the ownership of data within the organization, access protocols, confidentiality protocols etc.
Business metadata is used to search for the right data elements and to understand of the data elements are indeed the ones that the users need. It is also used to identify owners for remediation of data quality issues. It is also used to understand how data can or cannot be used, what are the restrictions on the usage of data and what are the privacy and confidentiality issues associated with the data elements.
Technical metadata defines the lineage of data to the source data elements. It also tells the exact place where data elements reside and can be accessed. Technical metadata provides information about the quality of data, volume, discrepancies etc.
Technical metadata is used to understand how to access the data. It also gives information about the quality of data, so that business users can decide if the data can be used for their requirements. Technical metadata gives information about the source system, which helps users understand the context in which the data was collected.
Operational data gives information about when data was acquired, how was it collected, what operations have happened on the data etc.
Operational data helps users to understand the validity and relevance of data, depending on when and how it was acquired and processed.
Usage data gives information regarding who have access to the data, who are using the data, for what purposes, at what frequency, from what locations etc.
Usage data gives an indication on the demand for the data elements and the type of users who would like to use this data. This information can be further processed to proactively recommend datasets to potential business users depending on their search patterns.
Usage data is also helpful to monitor possibilities of data security breaches, data leakages, frauds and query cost distributions.