If finding last year’s content requires a message to the entire team, files may have been saved without a working library ever being built. I judge a library by whether someone can find appropriate material and understand why it is safe to use.
Start with a real search
Imagine a team looking for vertical footage showing a product in use. A filename containing a date and sequence number makes every clip a separate investigation. Recording the product, situation, format and approval status makes retrieval a meaningful selection process.
I would not begin by tagging the entire historical archive. I take a small selection of frequently reused material and collect the words people actually search for. Several labels that mean the same thing become one shared term. The system should remain understandable when its original organiser is away.
What each record needs to answer
A useful record identifies the subject, reader question, source file, published version and conditions of use. Knowing where an image came from matters; knowing whether its permission has expired matters just as much. Unclear permission should not quietly become an approved status.
I distinguish source material, working files and published outputs. Otherwise an editable project can be mistaken for a deliverable, or artwork containing an outdated price can return to the calendar. One authoritative version link is more reliable than several files called final.
Retrieval is not approval
Finding an asset does not make it ready for publication. Is the information current? Is the product still available? Are the permissions for people, music and distribution suitable for this use? I review these questions before planning content repurposing.
Strong historical performance is not a universal endorsement either. A short learning note can be more valuable than a view count: which question did this answer, what remained unclear, and where did the audience ask for more detail? The library then carries editorial reasoning as well as material.
Give maintenance a place in the workflow
Complete the record when content is published, then review time-limited assets at an agreed interval. An inactive status can preserve publication history without inviting accidental reuse. Personal information needs a separate assessment of whether retaining it is necessary at all.
My practical test is simple. Can a new collaborator find a suitable image without help and explain the conditions under which it can be used? If they can, the archive has started to function as shared knowledge rather than a collection of folders.