Reddit crowd sentimentTracked across the finance subredditsSee the dashboard for live mention dataReddit crowd sentimentTracked across the finance subredditsSee the dashboard for live mention data
Daily Wikipedia lookups for every tracked company — human readers only — each scored against its own 90-day normal, side by side with the same name's Reddit chatter. The rarest read is the divergence: researched, not discussed — lookups spiking while Reddit stays quiet. Attention that precedes narrative.
Today the public suddenly researched: OSCR
Daily · data through 13 Aug 2026 · 2,292 companies mapped
Broke their baseline
OSCR41× a normal day — 7,215 views vs a usual ~174
Each company is matched to its Wikipedia article through Wikidata — a public, openly-licensed identity database — and every row links to the exact article and Wikidata entry we use, so you can audit any match. Every night we fetch the previous day's view count from the Wikimedia Foundation's official Pageviews API, counting human readers only (Wikimedia separates people from bots; we take the human number). A company is flagged when yesterday's views ran more than 3 median-absolute-deviations above its own trailing 90-day median — a robust yardstick one viral day can't distort — and cleared a 500-view floor. Articles get renamed; we follow the Wikidata identifier, not the title, so a rename never breaks a series. What this is not: views aren't buyers, and attention isn't endorsement — this shows what people are researching, nothing more.
Pageview data: Wikimedia Foundation Pageviews API (openly licensed). Article mapping: Wikidata (CC0).