Boutique research can still win

A little-known New York research shop called Citrini Research went from Substack niche to Wall Street conversation piece after one February 2026 paper on artificial intelligence was blamed for helping spark a broad tech selloff. Bloomberg said the report added to fears about job losses and business-model disruption, and Business Insider tied the market reaction to February 22. (bloomberg.com) (businessinsider.com) Then it did something even stranger in April 2026: CNBC reported that Citrini said it sent an unnamed analyst by boat to Oman’s Musandam Peninsula to watch ship traffic in the Strait of Hormuz firsthand. The firm’s field note said traffic had recovered to roughly 15 ships a day, which pushed back on the market story that the waterway was simply shut. (cnbc.com) That is the whole appeal in one picture: while bigger firms scrape the same feeds, a small outfit tries to come back with something nobody else has. Citrini’s own pitch says clients should “never have to ask ‘what’s the trade?’,” and its research is built around a few deep thematic primers each year instead of a constant stream of generic notes. (citrini.com) (citriniresearch.com) The company is not selling itself as a giant bank research department. Its public pages say it began publishing in early 2023, focuses on thematic equity investing and global macro trading, and distributes much of its work through a Substack with more than 211,000 subscribers. (citriniresearch.com) (substack.com) Its core bet is simple: most market returns come from a tiny number of companies, so the edge is spotting the narrative early, before the consensus spreadsheet catches up. Citrini says it builds long and short baskets around “megatrends,” and its site claims a first-year return above 100% for its core portfolio called Citrindex. (citriniresearch.com) (citrini.com) That pitch lands at a moment when large parts of finance research are getting easier to automate. If every junior analyst can ask a language model for an earnings summary, the scarce thing stops being summary and becomes judgment, sourcing, and the willingness to chase a fact all the way to a dock in Oman. (citrini.com) (cnbc.com) Citrini’s February paper worked for the same reason, even for people who hated the conclusion. NPR’s WBUR said the note imagined a June 2028 world where artificial intelligence had crushed white-collar jobs and spending, and Bloomberg said the co-author later stressed it was a scenario rather than a prediction. (nprillinois.org) (youtube.com) A scenario can still move prices if it is concrete enough to force investors to redraw a map in their heads. That is what boutique research can still do in 2026: not out-compute the big firms, but out-angle them with a sharper question, a cleaner framework, or one piece of reporting the rest of the market does not have yet. (bloomberg.com) (cnbc.com) That does not mean every small shop wins. It means the bar has shifted: if routine research becomes cheap, the value moves to original information, strong synthesis, and a view that ends with an actual position instead of a 40-page maybe. Citrini’s rise is a reminder that in markets, being smaller can still help if it makes you faster, stranger, and earlier. (citrini.com) (substack.com)