Developing and implementing a metadata strategy has always been a tough proposition, as most of the senior management do not understand the importance of the subject. Let alone metadata many IS community members don’t even understand the importance of data.
Metadata in most simple terms means “structured data about data” or “data that describes something (that may or may not itself be data).” Examples of metadata include: definition of the data element, business names of the element, systems abbreviations for that element, the data type and size of the element, source location, data steward, alternate alias, alternate spelling etc.
In other terms, metadata is any information used to aid identification, description, characteristics, location of, access to, data elements and information. So the question is, why is metadata important? Meta-data serves as a bonding agent that ties various tools and technologies together at an enterprise level.
A sound metadata strategy ensures seamless sharing, exchange and integration of tools and repositories. Developing a metadata strategy for an enterprise requires co-operation from both technical and business area members in an enterprise. Technical metadata provides the description of the data in an IT infrastructure where as business metadata defines where the data came from, who owns it and how it gets transformed.
To have an effective metadata strategy some essential building blocks include:
- Identify the technical members most knowledgeable about the data
- Identify the business area members most knowledgeable about the business data and business processes
- Have business users identify the data stewards and data advocates for data and processes
- Determine sources of record for data within an enterprise
- Determine the usage of metadata
- Identify all sources of metadata including data models, rose models, CASE tools, silo repositories, data dictionaries, glossaries, third party dictionaries, business rules, data mapping documents, business process mappings, data flow diagrams, corporate abbreviations, excel spreadsheets, catalogs etc.
- Determine the overall architecture for metadata storage. A metadata storage architecture defines an organizations overall strategy for metadata storage and retrieval of metadata. A centralized repository pulls in all the metadata in one location, a distributed architecture creates many small focused repositories and a hybrid architecture combines the two for best results
- Determine integration points and processes necessary to consolidate and integrate the metadata
- Determine metadata reporting and dissemination strategies
- Evaluate full lifecycle metadata management tools
- Standardize and document metadata sourcing processes
- Determine how to keep the metadata up-to-date. This has always been a challenge in organizations. It is important to keep the metadata refreshed as obsolete or stale meta-data can result in wrong decision making
- Standardize and document the metadata change management process and procedure
- Define metadata security needs
- Define components of the meta model
- Identify issues and constraints
- Form a metadata steering committee
- Discuss and confirm the metadata management strategy with the management.
Remember, an enterprise metadata solution cannot be done with a big bang approach. The metadata solution project needs to be divided into smaller more manageable chunks. An advantage of delivering a metadata solution in a phased approach is an early buy-in from the business areas and project champions.
In conclusion, development and implementation of a Metadata Strategy enables an organization to begin to measure the value of the information assets under their control.