I keep this in my Scans directory, this is just a backup.
For what it's worth, here's my workflow. (I use VueScan for multi-page scans. It's the only scanning software that works with my Brother MFW-J805DW on Ubuntu, because it provides better drivers for Linux than Brother does. That's an endorsement.)
I created a
\Scans directory in my
\Home directory1 and use this as a temporary landing spot for scans. (This is set as default in VueScan.) I scan multi-page docs and save them there, using 001.pdf, 002.pdf, etc. as filenames. Then it's later very easy to combine them with pdftk:
$ pdftk *.pdf cat output newfile.pdf
If you have to scan the front sides and back sides separately, use
$ pdftk A=001a.pdf B=001b.pdf shuffle A B output 001.pdf
where we assume that 001a.pdf contains the right-hand (i.e., odd-numbered) pages and 001b.pdf contains the left-hand (i.e., even-numbered) pages.
Chances are your document feeder will scan the left-hand pages in reverse order. In that case, use this command:
pdftk A=001a.pdf B=001b.pdf shuffle A Bend-1 output 001.pdf
If I have to break into a project, I will just create a subdirectory for the current project and dump all those scans in there temporarily.
1I'm generally opposed to adding new subdirectories here, but what can you do? I don't want this under
\Documents as this directory contains transient files for the most part.