It's important to manage your metadata well so that people can easily find, understand, use and share your information assets.
You can improve the way you manage your metadata by conducting a metadata management project.
Conducting a metadata management project
A metadata management project is a planned, iterative approach to improving how your agency manages its metadata. It is an important tool for helping your agency achieve the aims of its information management strategy.
Follow these steps to:
- target areas where metadata quality could be improved
- improve metadata management across your agency.
1. Determine the scope
Decide whether your project will be enterprise-wide or limited in scope.
There are benefits from improving metadata management at the broadest practical level. However, if the scope of your project is too ambitious, it may be unsustainable or harder to implement.
You may start with a broad, enterprise-wide view, but plan for incremental improvements in areas of greatest need.
In deciding the scope of your project, ask:
- how many users would this change benefit?
- what resources are available?
- how much improvement can it actually deliver?
- is it practical?
- is it sustainable?
- what are the priorities for action?
You could include in your project scope:
- the whole enterprise or agency
- a new system or systems being acquired or developed
- all business areas or user communities with a related line of business
- areas of greatest business risk or value
- areas where metadata implementation requires most improvement.
Your project scope may target specific information assets on your agency's information asset register.
2. Confirm support and resources
Obtain senior management endorsement for your project.
This will ensure you can engage stakeholders to establish contact points and support roles.
3. Consult key stakeholders
Consult key stakeholders to better understand:
- the current state of metadata implementation
- any existing issues related to metadata
- possible improvements
- user needs.
Key stakeholders could include:
- staff who use information systems
- business owners of information systems
- information system administrators
- information technology support staff
- data or metadata analysts
- data stewards or custodians
- senior information or data management staff, for example a chief data officer or chief information governance officer.
Some of these stakeholders can also help you undertake the project.
4. Understand existing practices
Work closely with your stakeholders to understand what practices and schemas are already in place and could be built upon.
You should look at informal and ad hoc practices and schemas as well as formal or official ones.
- existing system configurations
- schemas in use or proposed for use
- data models
- common business language
- controlled vocabularies.
When engaging with stakeholders, you may discover issues that are beyond the scope of the project. You will need to decide whether to address these issues in the current project or note them for future action.
5. Identify best-practice schemas or standards
Identify any available schemas or standards recognised as best practice, especially those which set baseline requirements relevant to your business (for example, geospatial, medical or scientific).
Compare these standards with your existing metadata implementation. Identify whether your existing implementation conforms to these schemas and standards or needs improvement.
Our Data interoperability maturity model can help your agency:
- assess its maturity in managing metadata according to standards
- take steps to improve its maturity.
You may need to adapt schemas and standards to meet the unique requirements of your business or work within system configuration constraints.
Remember that adapting or deviating from recognised standards can be risky. Assess whether the risk is acceptable.
- which schemas and standards you have decided to adopt
- how they meet the business need
- any adaptations made.
6. Develop a metadata framework
Write a framework for metadata implementation based on the user needs, schemas and standards you have identified.
You could incorporate your framework into an existing framework or strategy, such as a data or information governance framework.
At a minimum, your framework should outline:
- the schemas and standards to be used
- roles and responsibilities for using and managing metadata
- systems, processes, procedures and rules to help maintain the metadata implementation
- evaluation and monitoring processes
- review cycles
- approval processes.
Seek endorsement from senior management for the framework.
Once endorsed, promote the framework to users within all relevant business areas.
7. Implement metadata
Implement metadata according to your framework.
Create a data dictionary or model
Consider creating a data dictionary or model to represent the schemas identified in your framework.
A data dictionary or model shows the relationships between metadata properties and values.
Data dictionaries and models can help users understand how schemas describe information assets. They can also help systems administrators configure information systems.
Configure metadata in agency information systems, according to the requirements of the schema wherever possible.
Workarounds, such as cross-mapping across systems, may be required to meet requirements.
Establish a metadata registry to help users understand and agree on definitions for metadata properties.
You may wish to look at standards to better understand:
- the purpose of a metadata registry
- how to coordinate metadata registration.
Useful standards include:
- ISO/IEC 11179 Information Technology – Metadata Registries
- ISO/IEC TR 20943 Information Technology – Procedures for Achieving Metadata Registry Consistency.
Help users understand their responsibilities in maintaining metadata as outlined in the framework, with reference to roles, procedures and business rules.
8. Respond to change
It’s important to ensure your metadata continues to meet the changing needs of your users over time.
Revisit your metadata implementation regularly to ensure it:
- effectively enables people to find, understand, use and share information
- complies with current standards
- meets the changing needs of your business environment and users
- works with your business information systems
- is automated wherever possible.
Make sure you check in with your users often to ensure they:
- contribute to, adopt and understand agreed metadata terms and definitions
- are creating good quality metadata using tools such as the metadata registry.