Do not miss new blog posts! Subscribe to new posts, news, and updates.

Thank you

Metadata hierarchies in SharePoint Online

Metadata in SharePoint Online is used to find and manage files. The metadata may be free-form text or numbers entered by users, or it may have pre-defined values and vocabularies that users pick or are set as default on the storage container, e.g., company name, department, country, process, document type. The latter makes it easier to find information in and across SharePoint sites. It can also be used to as search results refiners like in the example below. This does requires mapping in the search administration. Advanced search allow users to search using metadata.

Search with metadata refiners

Metadata is often just a list of values, e.g., companies, processes, document types. If the list is long, it often makes sense to group it into a a tree or hierarchy, e.g, processes with sub-processes, document types with sub-document types. Most metadata should be set as default on the storage container when the site is set up, e.g., using the free Infotechtion Teams Hub. If users have to manually enter any metadata, then change management is key. Users will then have to see the value and importance of adding the metadata, and you need ongoing monitoring and governance to identify and address those users that don´t do it.


If users have to manually add some metadata, they can do this by writing what they think is the correct term and SharePoint auto-suggests the available tags.

SharePoint metadata auto suggest

Users can also press the tag symbol to see all available metadata values.

SharePoint metadata metadata options

If the list of values is long, e.g., countries in the world, then this should be grouped to make it easier for users to find the correct value, e.g. countries per continent.


The challenge is metadata hierarchies that should enable users to tag, filter and find files at any level in the metadata hierarchy. Let me share a simple example of this and your options.


You have some documents covering continents, and others covering countries within the continent.

  • Document 1 about Europe

  • Document 2 about the Netherlands

Critical success factor: If I search for Europe, I should find documents both Europe and countries within Europe like the Netherlands


You can achieve this by two different approaches. The best approach depends on use case. For continents and countries, we often end up with option 1. For document type and sub-types, we often end up with option 2.


Option 1: Two metadata values / columns

  • Column A: Continent

  • Column B: Countries grouped by Continent

Impact:

  • The deeper the hierarchy, the more repetition for users since they will have to pick the top-values many times to select lower values (e.g., Europe and then Netherlands)

  • No relationships between the columns when filtering search results since columns are not connected

  • More metadata on the document: Europe and the Netherlands


Option 2: One metadata value / column

  • Column C: Hierarchical values with Continent as parent and Countries as children

Impact:

  • Reduces the number of clicks to tag documents

  • Customization required to show hierarchy as search result filters

  • The relationships is managed in M365 which means you will find files about the Netherlands when searching for Europe, but the file only get the last tag as metadata: the Netherlands


Feel free to contact us if you need help establish a metadata framework for automatic classification of information in Microsoft 365.

177 views

Recent Posts

See All