Hi everyone! When setting up the data model, I connect to a database that contains a lot of columns I don't really need at the moment. I can delete the facts and attributes not needed, but if it turns out I need it later on it seems like I can't just add the attribute/fact back to the dataset? At the same time, I don't want to burden the user with having to see a lot of unnecessary attributes/facts in the analyze page , but I've understood these can be hidden via data permissions? What is best practise for this kind of case? thanks!

Hello Josefin! Are you considering embedding? If so, I would recommend you to explore

excludeObjectsWithTags

URL param that allows you to "hide" certain metadata objects (metrics, facts, attributes) from the catalog in Analytical Designer: https://help.gooddata.com/doc/enterprise/en/expand-your-gooddata-platform/gooddata-in[…]d-gooddata-elements-into-your-applications/embed-a-dashboard.

that’s right, you should be able to add any columns that were deleted back to the LDM by editing the LDM. Just be sure you map it to the correct source column.

Alright, that's great news, thanks!