Your Documents folder has 1,200 PDFs accumulated over four years. Some you need, most you don't, and you can't tell which is which. The thought of cleaning it makes you do anything else first.
One afternoon. Three piles. Done forever.
Set up three destinations
Create three folders: `/Archive/<year>/` for things to keep, `/Review` for things you're not sure about, `/Trash` for things to delete. Move every PDF into one of these.
Don't agonise. Three buckets, fast decisions. The point of the cleanup is to break the paralysis, not to perfectly classify every file.
Sort the source folder by date and triage chronologically
Oldest first. For each PDF: glance at it (use Flint for a quick view), make the call, move it. Aim for 5 seconds per file. 1,200 files at 5 seconds = 100 minutes.
For any PDF older than 3 years that you don't immediately recognise, it goes to `/Trash` or `/Archive`. Don't `/Review` everything — be decisive.
Rename the keepers as you go
Anything moving to `/Archive` gets renamed: `YYYY-MM-DD_<context>_<doc-type>.pdf`. The triage and rename happen in the same pass so the archive starts clean.
This is the moment of investment. Five seconds of renaming now buys you years of findability.
Empty review, empty trash
After the main triage, work through `/Review` — these are the genuine 'maybe' files. Force a decision: archive or delete. Don't let the review folder become a second clutter folder.
For `/Trash`, wait a week before emptying. Anything that survives a week without you noticing it's gone is safely deletable.
FAQ
What if I can't decide on a file?
Default to archive. Storage is cheap; recreating documents is not. The only files that should be deleted are clear duplicates or genuinely useless garbage (random downloads, OS-generated test files).
How do I handle 200 identically-named scanner files?
Open a sample in Flint, figure out what category they are, then batch-rename in your OS with a meaningful prefix. Treat the whole group as one unit.
Should I do this in one session or spread it out?
One session is better — you build momentum and the decisions stay consistent. Spread out, you'll re-evaluate the same files multiple times.
What about PDFs that link to other PDFs?
Rare in practice. If you have internal cross-references, keep the linked group together (same folder), but otherwise treat each PDF as standalone.
One afternoon, three folders, ruthless decisions. Start with the worst folder in Flint and end the chaos.