Build a research database in Notion from the web
Most research databases collapse under their own weight. Not because of what went in: because of what wasn't labelled when it did.
Filed captures the structure along with the content. Every save includes the author, publication, and the full extracted text. The database stays navigable because each entry is complete.
The capture
On any paper, article, or preprint page: press ⌘⇧Y. Filed extracts the title, author, publication name, and the full text, then creates a page in your Articles database. The text is formatted as Notion blocks inside that page.
For pages where the keyboard shortcut doesn't fire cleanly (PDFs, some journal gates), right-click and choose "Save as Article."
Properties captured on each save
- Title (from page headline or title tag)
- Author (extracted from article metadata)
- Publication (journal, site, or preprint server)
- URL (original source)
- Saved date (automatic)
- Full article text (as Notion blocks inside the page)
Organizing as you go
Filed creates the Articles database automatically. You don't design the schema. What you do: add tags, status fields, or project links from Notion after the save. Filed brings the content; how you organize it stays yours.
This isn't a capture problem. It's a structure problem. Filed handles one half; Notion handles the other.
Domain rules for regular sources
If you read from arXiv, PubMed, SSRN, or any specific journal regularly, set a domain rule. Filed detects the domain and routes to Articles automatically. You stop thinking about which workflow to pick.
Domain rules are a Pro feature. The Articles workflow itself is free.
Articles capture is free. Domain rules are Pro.
Add to Chrome · it's freeRelated