Hey all, So I have set up my gooddata environment with a Production Parent workspace and then children workspaces that correspond with each of my user groups. Some of the data in the Parent workspace is shared to all children while some is specific to only single children spaces. I have implemented filters on these children workspaces that is working great but have run into some situations that are causing confusion for users when using the Analyzer, Attributes, facts and Metrics will appear in the left hand menu that don't have any data in them for that specific child workspace. Is there an easy way to filter and hide these empty fields? Or apply tags on them so they can be group in the menu bar. Haven't found anything in the docs yet regarding this. Thanks

Short answer: there’s no built-in “hide if empty” switch in Analyzer. Items in the Data Catalog are shown/hidden based on visualization context, not based on whether they have data for a given child workspace. You can, however, organize what appears and how it’s grouped: - Organize/Group with tags: You can tag metrics/facts/attributes to group them in the Data Catalog; groups and items are shown alphabetically, with untagged items under “Ungrouped” (this affects grouping, not visibility) ([Review and organize the Data Catalog](https://www.gooddata.com/docs/cloud/create-visualizations/catalog-in-ad/)). - Show/Hide relevance per visualization: Analyzer hides items it deems inapplicable to the current visualization; this is contextual and not tied to data emptiness. Advanced users can enable “Show hidden items,” but that’s for revealing, not hiding more ([Review and organize the Data Catalog](https://www.gooddata.com/docs/cloud/create-visualizations/catalog-in-ad/)). - Limit data via WDF/UDF (not field visibility): Workspace Data Filters limit data per dataset in child workspaces but do not auto-hide unrelated attributes/facts across the model; WDFs must be explicitly mapped to each dataset where they should apply ([Set up WDFs](https://www.gooddata.com/docs/cloud/workspaces/workspace-data-filters/), [Assigning WDFs to specific datasets](https://www.gooddata.com/docs/cloud/workspaces/workspace-data-filters/#assigning-wdfs-to-specific-datasets)). UDFs can propagate along LDM relationships but similarly restrict data, not the presence of fields in the catalog ([Set up UDFs](https://www.gooddata.com/docs/cloud/workspaces/user-data-filters/)). - Hiding model fields per child workspace is not supported: You can’t selectively hide attributes/facts in a child while keeping them in the parent; some teams emulate this by using separate workspaces or extending child models via API, but there’s no simple UI switch to hide fields in child workspaces ([Slack: not possible to hide fields per child workspace](https://gooddataconnect.slack.com/archives/C04S1MSLEAW/p1730891494310599?thread_ts=1730891494.310599&cid=C04S1MSLEAW)). Workarounds you can consider: - Use tags to clearly group “Core” vs. “Tenant-specific” items to reduce confusion in the catalog UI ([Review and organize the Data Catalog](https://www.gooddata.com/docs/cloud/create-visualizations/catalog-in-ad/)). - Ensure WDFs are enabled and mapped on every dataset where they should take effect to prevent “leaking” values into pickers; otherwise, attribute filters can show full value lists ([Assigning WDFs to specific datasets](https://www.gooddata.com/docs/cloud/workspaces/workspace-data-filters/#assigning-wdfs-to-specific-datasets), [Slack clarification](https://gooddataconnect.slack.com/archives/C04S1MSLEAW/p1739376686967059?thread_ts=1739376686.967059&cid=C04S1MSLEAW)). - If the goal is cleaner dropdowns on dashboards, “Filter by Metric” can pre-filter attribute filter value lists to only values that exist in related facts (convenience, not security) ([Slack: Filter by Metric tip](https://gooddataconnect.slack.com/archives/C04S1MSLEAW/p1733176312237209?thread_ts=1733176312.237209&cid=C04S1MSLEAW)). I’m sorry, but the knowledge sources do not mention a feature that automatically hides empty attributes/facts/metrics per workspace in Analyzer. The supported approaches are data restriction (WDF/UDF) and catalog organization via tags.

Hi Grant, Please, give it a try and review the previous response, then let us know if you have any further questions.

Hi Grant, if you are embedding GoodData into your application, you might use the “includeObjectsWithTags” or “excludeObjectsWithTags”. It will allow you to filter the catalog in Analyzer to objects that do have or do not have a particular tag. Unfortunately this is not available directly in the portal, when not embedded, it will show all data model objects. Please note that tags are also used for organizing the catalog to folders. There is currently no UI for adding and removing the tags, but that will be available soon. In the meantime you would need to use the API to define the tags.

Thanks @Michal Hauzírek I am using iFrames to embed the analyzer so I will look into the tags options on that

Great, in that case it seems like an ideal scenario for this. BTW I believe all attributes and facts automatically get a tag based on title of their first dataset name (this is how their grouping in Analyzer works). So you can try to utilize these tags without need to add your own. You can check the tags of the LDM objects in the layout API: /api/v1/layout/workspaces/{workspaceId}/logicalModel