Both describe assets and power search, but they serve different purposes:
Custom metadata: Structured data stored in predefined fields, such as select lists, dates, numbers, or text. Configured by library owners to enforce governance and faceted search.
Tags: Free-form, open-ended keywords added directly to an asset for broad discoverability.
Common alternative terms include fields, properties, or labels for metadata, and keywords for tags.
Should I use custom metadata or tags?
Choose between custom metadata and tags depending on how structured your asset information needs to be:
Use custom metadata when:
Assets need standardized values across your library, such as Target market, Asset type, or Usage rights.
A field must be required before an asset is published.
You need specific view or edit permissions on certain fields.
You want consistent dropdown options to prevent typos and duplicate filters like US and USA.
You want faceted filter pills in search results.
Use tags when:
You need flexible descriptive keywords that vary by asset, such as visual themes like flat-lay, minimalist, or lifestyle.
You want AI-suggested tags added automatically for common visual objects.
Consistency is secondary to speed and open-ended discovery.
Where do I find custom metadata and tags?
Both custom metadata and tags appear in the asset details sidebar:
Custom metadata: Located under Custom metadata. Properties are configured in library settings.
Tags: Located under Tags, where editors enter keywords or accept AI suggestions.
Important notes about metadata and tags
Moving assets: Custom metadata schemas are set per library independently, while project metadata is set at the brand level. Metadata properties and tags do not automatically transfer when moving assets between libraries or from projects into libraries.
AI auto-tagging limits: AI auto-tagging identifies common objects like laptop or desk. It does not detect brand names, product codes, or text inside images.
Embedded file metadata: Embedded IPTC, EXIF, or XMP keywords import automatically as asset tags. Technical camera and file properties remain read-only asset details and do not map to metadata fields.
Bulk tag removal: Removing a tag from multiple selected assets only deletes that specific tag. All other unique tags on those assets remain intact.
Search indexing: Both custom metadata fields and tags are indexed in search. Metadata powers structured filters, while tags support keyword queries.
