Data study · Updated 2026-06-20
The state of research-paper overload
Research output keeps growing faster than anyone can read. Using Scollr's own indexed corpus, here is a concrete look at the scale of new research papers — and what a workable response looks like.
The numbers behind the overload
Scollr indexes 836,810 research papers from open scholarly sources, spanning 4,798 research topics. Over the last year, roughly 806,064 papers entered that corpus — an average of about 7,738 new papers every day.
No researcher can read at that rate, and no single field is small enough to escape it. The result is a familiar failure mode: important work is published, but it is never seen by the people who would build on it. Both Nature and the LSE Impact blog have documented information overload as a real, growing barrier to keeping up with the literature.
Overload is not a reading-speed problem; it is a filtering problem. The papers worth your attention exist — they are just buried in the volume.
What actually helps
The workable response is to stop trying to see everything and start following the slice that matters. Track specific topics, authors, and sources; let a personalized feed surface new and relevant papers from those areas; save the ones worth reading; and map related work when you need the full picture. That turns an unreadable firehose into a short daily scan.
Method
Figures are drawn directly from Scollr's indexed corpus on 2026-06-20 (sourced primarily from OpenAlex, enriched with Semantic Scholar). "New papers per day" is the 90-day average of papers by publication date. Counts exclude retracted papers, paratext, and low-quality sources. Numbers shift as new work is ingested.