That is a sensible way of turning chaos into order, but let a contrarian voice tell an alternative story. I think filesystem pdf organizing is underrated. I have been curating a tree along years and now I know where to park new pdfs by heart in there. Wrote an arxiv button adder with grease-monkey to save pdfs with title, year and authors as file-name. Windows have a nice abstract property system that can use pdf (bib) metadata by installing a shell extension, that you can use in Explorer, where you can customize displayed fields. Classic Windows search offers a good background indexed database of content and metadata for local files. You just need an "IFilter" component to let it see inside the pdf text layer (suamtraPDF cames with one). It is the kind of functionality one would expect of a commercial product, just for the windows license. Beyond File Explorer, You can query the index from scripting or dot-net. I wrote a flow-launcher plugin to making searching faster than from Explorer search text input. Search supports "saved searches", saving queries in xml. I use a lot two, listing the saved pdfs of the last week and the last month. Windows is weird in having that cute infra built-in, but just having left unfinished the last glue code needed to give some good user experiences managing docs.