Managing a growing library
After thirty or forty thumbnails, your Thumblore library stops being a record of what you made and becomes a working asset — or it becomes a mess. Collection management is the small amount of upkeep that decides which.
The goal is simple and concrete: at any moment, you should be able to answer "which three layouts work for this channel, and where are they?" in ten seconds.
The structure you are aiming for
| Layer | What it holds | How many |
|---|---|---|
| Working set | The three layouts you actually reuse | 3 |
| Series templates | One base frame per ongoing series | 1 per series |
| Faces | Named avatars — host, guests, alts | As needed |
| Archive | Everything else, searchable by template name | Unbounded |
Only the first two need active management. The archive manages itself as long as you keep naming things properly.
Naming is 90% of it
Thumblore files by the template name you type when generating, and Projects groups by it. That one field is your entire filing system, so treat it as structured data rather than a note to self.
Series — ep Nfor episodic content:Cabin build — ep 7Format — topicfor one-offs:Versus — Canon vs SonyBrand — series — ep Nwhen running several channels:ACME — how-to 12
Because filters live in the URL, a filtered Projects view is a bookmarkable link. Bookmark the series you are actively shipping.
The rule that saves the most time
Never type "test". Every thumbnail is filed forever under whatever you typed, and "test 4" is unfindable in three months. If you are genuinely experimenting, name it SANDBOX — idea so the whole experimental set filters together and can be ignored.
The quarterly review
Twenty minutes, once a quarter. This is the entire practice:
- Pull your numbers. List your last thirty videos by click-through rate.
- Find the top five thumbnails. Open them side by side.
- Name what they share. Face size? Text position? Background darkness? Expression? There is almost always a pattern, and it is usually not the one you expected.
- Promote three to the working set. Download them and keep them as your reference templates.
- Retire what is not working. A layout that has underperformed five times is not unlucky.
- Update the brand kit. Write the finding into your eight decisions so it survives your memory.
Reusing instead of regenerating
A managed collection changes how you make thumbnails. Instead of a blank brief per video, you start from a proven frame:
- Pick the working-set layout whose format matches this video.
- Open it in the editor.
- Swap the face if it is a guest episode.
- Replace the headline in place.
- Change the scene so it does not look like a rerun.
- Download.
Two or three edits instead of a generation plus several corrections — and the result is on-brand by construction rather than by luck.
Managing the face collection
Faces need the same discipline as layouts. Keep the naming prefixed and purposeful — Main, Main-Alt, Guest-Sam, Brand2-Host — and delete faces you have not used in six months, especially if you are on a plan with an avatar cap. Deleting a face never alters thumbnails you already generated with it.
If someone else takes over
The test of a well-managed collection is whether an editor or VA could produce an on-brand thumbnail without asking you anything. Give them: the brand kit's eight lines, the three working-set images, the avatar names, and the naming convention. If that is enough, your collection is managed. If it is not, the gap you just found is exactly what to fix.