Data can change over time. Historical data will remain stable, whereas recent data can change frequently. Therefore, we recommend you to have a data maintenance process in place. This process consists of maintaining data and metadata regularly, checking URIs & URLs, checking user feedback and continuous improvement and evaluating your success.
Data maintenance process
Maintaining data and metadata regularly
Both data and metadata can become outdated. Updates, changes or other influences can cause metadata to become obsolete. This will result in low discoverability and lower quality results for re-users in their search for valuable datasets. Therefore, updates should be made regularly. Depending on how the data is generated, the relevance of updates and the re-use of the data set (frequency, quality), the data set needs to be updated with a particular frequency. Perform a thorough assessment of the data and discuss the policy within your organisation.
When an update is made, a key recommendation is to mention the date of the update in the metadata!
Checking URIs & URLs
The World Wide Web is dynamic. Therefore, it is important to regularly check if all your URIs and URLs (from and to datasets) are still working. If the data set’s URI or URL changes, a website that refers to your data set will redirect users to non-existing pages.
Checking user feedback and continuous improvement
User feedback will increase the quality of your data publications. Users can provide feedback about the data on any aspect and by incorporating feedback into your processes, the usability and discovery of the data can be improved. Consider adding an option to receive feedback as a potential improvement of the portal.