![caotue management downfall 2x1]()
Summary List PlacementLast spring, Coatue Management, a $25 billion hedge-fund giant, did something unusual: It made an appearance.
In the 20 years since its founding, the secretive, tech-focused investment manager has had a stellar track record under the billionaire Philippe Laffont. But the "tiger cub" usually demurred when it came to discussing business publicly.
In May 2019, though, two Coatue execs spoke for nearly 45 minutes to a crowd of data-science wonks at Domino Data Lab's Rev conference.
The moderator, Matthew Granade, until just recently the chief market intelligence officer at Point72 Asset Management, underscored the rarity of the moment.
"Coatue is one of the most successful hedge funds of the last two decades. But they are very discreet, very quiet, don't talk a whole lot in public," he said before introducing Coatue's cofounder Thomas Laffont, Philippe's brother and the leader of the firm's private-investment strategies, and Alex Izydorczyk, its 26-year-old head of data science.
What compelled Coatue to pull back the curtain?
The firm had announced in an investor letter a few months earlier that it was raising several hundred million dollars to launch its first quant fund, an outgrowth of a data-science group that had been expanding under Izydorczyk. The firm boldly predicted that its team of 30 scientists and engineers would eventually reach 100.
Data science started at the firm in 2014 as more of a casual dalliance, a "bolt-on" to old-school stock picking that was nice to have, as Thomas Laffont put it. But that changed along the way.
Data available for investors has exploded over the past five years, with vendors selling everything from geolocation-based foot-traffic data to ocean-freight data gleaned from satellite images. Companies are also selling more prepackaged data sets.
In 2017, Coatue began spending lavishly to build out its data capabilities, which were bolstering not only its stock bets but the growing portfolio of venture and private-equity investments overseen by Laffont. He called the data efforts "existential."
"If we don't get this bet right, we won't be around in five years," he told the conference.
Coatue, named after Philippe Laffont's favorite beach in Nantucket, started in 1999 with $45 million and survived the dot-com crash despite being a tech-focused firm.
After a successful trial run managing internal capital in 2018, the quant-fund group was ready to work its magic on investor cash north of $350 million — a sizable bet, though dwarfed by the nearly $20 billion run by the firm's traditional hedge-fund strategies. Izydorczyk made other appearances that year to discuss the approach.
But Coatue's quant fund wouldn't last another 15 months.
By June 2020, Coatue had returned what outside capital remained and laid off a substantial chunk of the team. It vowed to continue running the fund's strategies with internal capital.
A short while later, the quant fund was mothballed entirely and ceased trading, sources familiar with the matter said. The remaining staffers were reassigned to support the traditional hedge fund, which has seen stellar returns in 2020.
Business Insider spoke with more than 20 sources — including 10 ex-employees as well as investors, vendors, and hedge-fund recruiters with firsthand knowledge of the firm — about the rise and demise of Coatue's quant fund.
These sources described a grueling environment where leaders micromanaged, hurled insults, reamed out staffers in public, and fired people seemingly on a whim. And they said that environment and a lack of buy-in for the strategy's philosophy at the very top of Coatue hobbled the heavily hyped fund.
All the sources requested anonymity to preserve business relationships and protect against reprisals. Many had signed non-disparagement or nondisclosure agreements.
Coatue declined to comment for this story.
In the hypercompetitive world of hedge funds, stars fade and burn out as often as they rise, typically because their hot streak runs cold and performance falters.
But as state-of-the-art data-science strategies become not just fashionable but table stakes for relevance, traditional funds are trying to adapt and straddle the two worlds. Even if they get the math right and make money, cultural pitfalls lurk for these new-age investors — and fanfare, pricey investments in data scientists and coders, and hundreds of millions in fundraising can obscure them for only so long.
The management style and decisions that minted billions in the age of pure-play stock picking could prove to be a disastrous fit for a quant fund, turning it into a short-lived money pit.
'Bait and switch'
The coronavirus pandemic has presented a huge challenge for quant strategies, with big names like Bridgewater and Renaissance Technologies struggling and other funds shuttering as markets experienced never-before-seen volatility.
Coatue wasn't immune. Sources close to Coatue told Business Insider that the quant fund abandoned certain strategies and pared back exposure in the early days of the pandemic. Waves of redemptions followed.
But the quant fund's struggles started long before the pandemic. It failed to produce impressive returns in 2019, returning just 2% for the year and losing money in the fourth quarter, even as it went on a hiring blitz. Strategies that soared the previous year fell back to earth.
Deeper cultural issues festered.
Data scientists and quants are generally well paid and highly coveted not just in finance but at tech titans like Facebook and Google and Silicon Valley's hottest startups. Cream-of-the-crop recruits in STEM subjects may hear from top firms years before they've even completed their Ph.D.
"There's a crazy talent war going on in this space for these brains," one recruiter who specializes in buy-side quant hiring said.
But Coatue was burning through them. The unit quietly earned a reputation for having one of the most miserable work cultures in all of hedge-fund land.
One industry headhunter, expressing a sentiment shared by several other recruiters Business Insider spoke with, put it bluntly: "The culture there sucks. They've turned over an insane amount of people."
Some employees who joined the data group expecting a lean, forward-thinking startup environment were jolted in their first week by a different reality.
Highly credentialed scientists and coders often found themselves doing grunt work and racing to handle a barrage of daily fire drills, sources told Business Insider. Even after 12- to 14-hour days, employees were on call, even on weekends and vacations. Several described it as a "bait and switch."
"The culture was shocking, even from an industry standard," a former data-science employee said.
Exploding offers
"I should've known what I was getting into then."
"It was the most aggressive offer timeline I've experienced."
"I remember feeling like I couldn't think about it at all, that I had to act on it."
In retrospect, exploding offers should have been a red flag, ex-staffers and industry recruiters told Business Insider.
Recruiters and six ex-employees said the practice was common at the firm. After several rounds of interviews and a final dinner with team leaders, an offer would materialize with as little as 24 hours to decide.
The practice is banned by some universities and discouraged by others because of the pressure it puts students under, as well as the problems it can create for other employers recruiting the same candidates.
For experienced professionals, it's viewed as more of a bad omen.
Many companies impose some time limit on offers. But of 10 veteran recruiters Business Insider asked — some of whom have worked with Coatue — all said such aggressive windows were uncommon for hedge-fund jobs outside of unusual circumstances like a bidding war.
Do you have a tip you want to share? Contact Alex Morrell via encrypted email (amorrell@protonmail.com) or on the encrypted messaging apps Signal, Telegram, or WhatsApp (262-573-1023). Contact Bradley Saacks (bsaacks@businessinsider.com) or on the encrypted messaging apps Signal or WhatsApp (919-816-5537).
Several said they'd never experienced it in their career.
"I can't think of any good reason why you'd do that. This is people's lives, not chess," a buy-side recruiter said.
So why did employees accept?
One former employee put it simply: "Money."
Coatue paid well, former employees and recruiters said, though not substantially more so than other quant funds.
Data scientists often started at a $150,000 base salary, while for quant researchers it was about $175,000, according to H-1B visa salary records.
Discretionary comp varied by role, though many employees didn't last longer than one bonus cycle.
Central Park views, kombucha, and golf simulators
Money wasn't the only draw. Office perks were also an alluring recruitment tool.
Coatue runs shop out of the iconic Solow Building at 9 W. 57th St. in Manhattan. The sloping, 49-floor glass edifice has housed firms like KKR, Apollo Global, Silver Lake, Chase Coleman's Tiger Global, and, more recently, the billionaire Dan Sundheim's red-hot D1 Capital.
Coatue's 25th-floor office has panoramic views of Central Park and the surrounding New York City skyline.
Near the marble-floored entry, in a plexiglass case with in-set lighting, an Apple II computer, an early iPod, a microchip processor, and other vintage Silicon Valley bric-a-brac form a mini museum installation. A bookshelf holds investing favorites including "Moneyball" and "The Big Short."
The rest of the office was designed with more modern accouterments.
![Solow Picasso Sculpture]()
Pantries were stocked with snacks and premium beverages like coconut water and kombucha. A private chef who cooked meals for top brass also whipped up protein bites, smoothies, and other healthy treats for the rank-and-file, who ordered Seamless on the house for lunch and dinner.
The office also provided a full-service gym, complete with two personal trainers. Laffont's private gym was transformed to hold a golf simulator — one source identified it as a TrackMan, the same state-of-the-art variety President Donald Trump installed in the White House, with custom pricing starting at $50,000 — with a golf pro on standby.
But only after joining did employees learn that life in the data-science group, which prized face time and long hours, discouraged taking advantage of these perks.
"If you left before 8 p.m. you felt like an idiot, like you weren't doing your job. That was sort of the bare minimum," one ex-employee said.
Some said they felt obligated to stay past midnight and felt pressure from superiors not to seem as though they were enjoying the benefits.
One ex-employee who joined a major fund with a competitive reputation described it as relaxed compared with Coatue.
"At first I was having stomach issues and anxiety issues," the ex-employee said of joining Coatue. "I had to get used to insane pressure about everything being due in the same day."
Another former employee recalled venting about the toll of the anxiety with another Coatue coworker.
"He said to me, 'When I lay down at night my heart is pounding, and I don't know why.' And I was like, 'Me too!'" they said.
Many pinned the blame on Izydorczyk, the head of data science.
The Wharton wunderkind
Before 2014, Coatue's data-science group didn't exist. That was the year Izydorczyk forged a relationship with Philippe Laffont.
Izydorczyk, a triplet, hails from Winnipeg, Manitoba. Both his parents are professors.
Requests to reach Izydorczyk for comment were routed to Coatue, which declined to comment for this story.
He arrived at the University of Pennsylvania in 2011 to study economics and statistics. He showed flashes of promise even as a freshman, deploying coding skills to assess the relationship between social-media sentiment and the stock market, an online bio says.
He caught the eye of a well-connected Wall Street quant recruiter named Alexey Loganchuk, who ran trading competitions to identify under-the-radar talent.
Loganchuk mentored Izydorczyk and brokered a meeting with Laffont, for whom he has placed several professionals, according to people familiar with the matter. Loganchuk declined to comment for this story.
Laffont ended up creating a summer internship specifically for Izydorczyk in 2014, with the undergrad working closely with the billionaire fund manager, sources said.
Initially, Izydorczyk was web scraping and wrangling data sets to track tech companies Coatue was investing in, helping to model customer traffic, retention patterns, and revenue growth.
He impressed Laffont that summer.
"Alex is really good. He's smart. Great coder, great statistician. He gets it. And he doesn't sleep. He works 24/7," a former Coatue employee said. "He was able to crank out a ton of work that looked incredibly promising."
Laffont hired Izydorczyk full time after he graduated in 2015 and a year later promoted the 23-year-old to head of the nascent data-science group.
Growing pains
While Izydorczyk was talented, smart, and hardworking, the recent college graduate had no previous management or leadership experience.
He quickly hired college connections.
They included Max Scheiber, a computer-science grad who spent a year as a software engineer with Google before joining Coatue, and Justin Bleich, who was a TA for Izydorczyk at Wharton while working on his statistics Ph.D. and who took the role of head of quantitative research.
The group's primary mission was to transform an avalanche of data sets, often raw and unstructured, into value for the broader fund, both the public trading side and the private investing side.
It took shape as three units: the quant team, which worked on building the group's algorithmic models; the quantamental or "special projects" team, which handled requests and projects that supported fundamental analysts; and the data-engineering team, which focused on the broader infrastructure and data pipelining.
But Izydorczyk's management style created friction.
"He's very direct and very impatient. And there's not enough hours in the day. His management style was very brash," a former employee said.
Izydorczyk, and often his top deputies Bleich and Scheiber, would routinely berate or insult employees in public and on Slack, ex-staffers said.
"He would scream at people, call them an idiot to their face, ask what the f--- they were doing," one former employee said.
He had colorful ways of dressing down staffers, including calling them "muppets" or saying they were "literally useless," ex-employees who worked under him said. He described some subordinates as "no better than well-trained monkeys."
He once lashed out at a staffer and angrily informed them that he'd seen moose that could do their job better.
But there was some precedent for this behavior. Several former employees described Philippe Laffont as "a screamer" with a short fuse who was prone to ripping into people.
Another troubling tradition, several former employees said, was Izydorczyk and his inner circle gathering around a table over dinner and deriding coworkers, fundamental analysts, vendors, and even potential new hires.
"They would be openly talking s--- about people," a former employee said.
Another former employee said he was told not to speak to certain people who had fallen out of favor with leadership, characterizing the environment as middle-school-like.
Long hours and fire drills
Former employees also described micromanagement, intense resistance to diverging points of view, and daily "fire drills" over nonurgent matters.
"If you were working with him, every hour he would check with you. There was no breathing room," an ex-employee said.
"He didn't have much management experience. Which is fine if you're willing to listen to people and willing to learn. But he wasn't," the source said.
Izydorczyk often arrived at 8 a.m. and worked past midnight. He sometimes slept on a mattress in the massage room in the gym, former employees said.
He didn't consider what he asked of others unreasonable and looked down on people who didn't meet his exacting standards, several ex-employees said.
Multiple former employees said people were often afraid to leave in the evenings, even if they weren't working, because they didn't want to be the first out the door.
At one point, Izydorczyk instituted a rule requiring the team to alert the data-science Slack channel if they would be unavailable over the weekend.
The firm hemorrhaged talent. Some quit, while others who drew the ire of Izydorczyk or his top lieutenants were fired. But the group continually hired to keep seats filled.
One former employee recalled that during a string of resignations, Coatue vigorously enforced noncompetes, and employees had to come in though they had little to do. They often hung out near the pantry by reception drinking cappuccinos; some jokingly nicknamed the area the "Expat Cafe."
One hedge-fund recruiter said Coatue was good at getting talented data-science staffers in the door but struggled to retain them.
"I get excited when I talk to people who are ex-Coatue, because their quality, in my experience, is usually pretty strong," the recruiter said.
'An enormous amount of time and money'
But long hours paid off where it counted: impressing Philippe.
By early 2017, a year after Izydorczyk took over as data head, Philippe Laffont was devoting substantial resources to the group, which had grown to 20, according to an investor letter Bloomberg reported on.
"Data science has become an integral part of our investment process," Laffont wrote. "We are spending an enormous amount of time and money building out and developing our data science and quant team."
It was no longer simply crunching numbers and providing a growth metric for tech companies that portfolio managers were researching. A data emphasis was starting to touch every facet of Coatue's business.
The group had, for instance, built a system to track "billions of anonymized credit-card transactions," Laffont wrote.
They ingested pitch decks, investor presentations, and other documents and analyzed those too.
For the public arm of the firm, credit-card data could be used to forecast income statements "in real time," Izydorczyk said at the Domino conference.
On the private side, it was about uncovering inflection points and the companies that stood to benefit from them, as well as adding value to the portfolio companies.
The group was building toward being an "operating system" for the firm's entire investment arm, Izydorcyzk said. But it would also get to invest its own money.
At the start of 2018, Laffont cleaved about $30 million in internal capital from the main fund to seed a quant-focused fund run by the data-science group, several former employees said.
One former analyst at Coatue theorized that the move was in part a marketing ploy — rivals like Viking Global and Tiger Global had built teams of their own, and data science had become an industry buzzword.
Another former data-science employee echoed that sentiment and noted that investors like funds-of-funds or family offices with allocation limits for tech-centered fundamental strategies could apportion additional funds toward a quant-focused effort.
The quant strategy
The data-science group wasted little time making a mark with the internal capital. While 2018 was a bloodbath for hedge funds — the industry had its worst year in a decade, losing 3.42% — Coatue's budding quant operation produced double-digit returns, former employees said, easily outpacing its flagship fund.
The group's secret sauce was acquiring alternative-data sets and finding clever ways to model corporate performance.
"A lot of what they were doing was basic econometric forecasting," an ex-employee said. "The lifeblood was really data and data pipelining."
Its breadwinner, a dollar-neutral approach that exploited credit-card data, could be distilled to this: Go long some 50 stocks forecast to beat consensus, and short 50-odd stocks forecast to miss.
How to predict the winners and losers?
Coatue bought raw, unstructured credit-card data from vendors like Yodlee, Earnest, and Facteus. It cleaned it up, tagged transactions to individual companies, and forecast quarterly performance with two factors in mind: How fast a key performance indicator — usually revenue — was accelerating, and the delta to sell-side analyst consensus.
The strategy worked best with small-cap stocks like the restaurant chain El Pollo Loco or the video-game retailer GameStop. Larger stocks like Chipotle worked if the business model was straightforward and the forecast metric was central to the company's story.
It didn't always work. In one example from the fund's early days, the method nailed a significant revenue beat for GameStop, but the stock fell because net margins decreased, one source explained.
With so many companies in play, sophisticated knowledge of a specific firm isn't realistic.
But it backtested the strategy, and averaging those predicted long and short bets would make money over the long run, a former employee said.
The group's other big strategy focused on consumer-packaged-goods companies, or manufacturers of products found on the shelves of supermarkets and convenience stores, like Pepsi, Nestle, Kraft Heinz, or Unilever. The firm placed bets on several dozen similar stocks, informed by point-of-sale data from vendors including IRI and Consumer Edge.
But there was no magical artificial-intelligence solution for tagging transactions to parent companies, meaning processing the data involved a lot of grunt work.
It paid off. The group's 2018 performance generated considerable internal hype, winning buy-in from Philippe and other managing partners, sources said.
In early 2019, it opened up to outside investors, raising more than $300 million.
The quant fund's momentum carried into 2019. But midway through the year, it hit choppy waters.
The primary credit-card strategy had limitations, including capacity constraints; there's only so much capital you can plow into a $100 million-market-cap stock. And with larger corporations, business models often get messier and difficult to project, and the ability to predict a salient metric solely with credit-card data shrinks.
Moreover, one source said, larger funds caught on to the general strategy, taking away some of the edge. Credit-card data is one of the most widely used subsectors of alternative data in the investment-management industry.
By late summer 2019, with the strategies faltering, the fund had significantly dialed back risk, former employees said.
"Then it started to turn into a bit of a panic mode," one said.
The downfall
During the Domino panel in 2019, Thomas Laffont, an avid baseball follower, cited an example from Major League Baseball to underscore the importance of leadership in the data-obsessed corporate landscape. He touted the Houston Astros and their "Moneyball" approach.
As documented in the book and the hit film, the rise of data analytics in the sport prompted a rivalry between traditional talent scouts and metrics enthusiasts.
Laffont pointed to a news item about how the Astros, not long after winning the World Series in 2017, essentially fired their last scouts. The "data-science nerds," as he put it, had won — for the league-leading Astros, there weren't "even scouts left at the table," he said.
He said that "leadership from the top is really important." Bridging communication gaps and smoothing over tensions between old-school employees and data wonks is "one of the most difficult issues" for a firm in transition, he said.
Laffont had no way of knowing then that the Astros would be caught up in an embarrassing sign-stealing scandal, revealed in part by former team scouts, that would claim leaders and result in a $5 million fine.
Nor could he have envisioned the downfall of Coatue's quant fund just around the corner.
The quant fund struggled in the second half of 2019. The point-of-sale strategy dragged, while the credit-card play wound up roughly flat, sources said. But it didn't suffer substantial losses in 2020.
Sources who worked in the firm's data-science group told Business Insider they were caught off guard when Philippe Laffont pulled the plug.
When the firm was hiring in 2019 to build out the data-science group and quant fund, prospective employees believed the effort had complete buy-in from senior management.
Philippe Laffont's track record as an investor is stellar, but he has no formal training in data science or quantitative trading.
He frequently second-guessed the fund, challenging strategies, signals, and positions, insiders said. Strategies were scaled up and down based on his wishes rather than a quantitative rationale. He'd focus on individual losing positions and bristle at the explanation that the strategy could net out positive overall.
That was a "loser mentality" for a big-bet investor like Laffont, an ex-employee said.
"He wasn't looking at it from a quant perspective," another ex-employee said. "He wanted to make it more and more like a hedge fund."
But as the CEO and owner of the fund, Laffont had the final say. It was his reputation and fortune on the line.
Part of the rationale of the buildout was to add classic quant strategies, such as factor-based trading, to supplement existing strategies — a priority at the outset of 2020. But when some of these strategies, which had small capital allocations, struggled early in the pandemic, they were suddenly mothballed.
Instead, the group concentrated its efforts on the flagship credit-card and point-of-sale data plays — and some were tasked with tracking the impact and spread of the coronavirus.
Former employees said that they were kept in the dark and that the significant shifts were poorly communicated and only after final decisions had been made. The same script played out over decisions to move the fund to cash positions and when investors pulled money.
Some former employees said they were alarmed at attempts by leadership to hide investor redemptions, including a substantial one in April.
By about midyear, the fund was down a couple of percentage points, though its primary strategies were roughly flat. Meanwhile, Coatue was up roughly 20% through June.
Then in June, roughly half of the quant team was abruptly laid off. The remaining members were informed that their role would pivot to supporting the broader hedge fund's momentum, but a precise rationale wasn't offered.
Coatue's flagship fund has continued its scorching run. Through November, it was up more than 52%, according to sources familiar with the matter and to other news reports.
With data analytics now so intertwined with the firm's operations, the data-science team's work contributed to the broader success — and the group, still led by Izydorczyk, is unlikely to disappear anytime soon.
But for Izydorczyk and the quant fund, which some ex-employees likened to his "child," it was an abrupt reversal.
Will it ever be revived? Laffont hinted in a June town hall that it could be, though he wouldn't favor the classic quant strategies the firm abandoned early in the year.
Some suggested he was never interested in a cutting-edge systematic trading fund but wanted to find a way to streamline and automate the fundamental-style investment process that made him rich — a robo-advisor of sorts, but with Laffont's investing DNA as the central algorithm.
More broadly, Coatue's quant-fund experiment exposed the type of culture clash that can hobble a wholesale melding of data science and traditional hedge-fund approaches, as well as the perils of anointing incredible young talent without regard for management ability.
Whether in professional baseball, hedge funds, or any other competitive corner of the new data frontier, even the most promising data strategies can be scuttled by corrosive culture and the whims and decisions of senior leaders.
SEE ALSO: Julian Robertson's Tiger Management is at the center of a quarter-trillion-dollar web linking billionaires, the Pharma Bro, and a 'Big Short' main character
SEE ALSO: The rise of Dan Sundheim: How a Wharton whiz kid became the LeBron James of investing, launched one of the hottest hedge funds on earth, and minted a billion-dollar fortune in the process
SEE ALSO: A rising star from billionaire Philippe Laffont's Coatue is starting his own hedge fund and has at least $300 million lined up from investors
Join the conversation about this story »
NOW WATCH: What makes 'Parasite' so shocking is the twist that happens in a 10-minute sequence
Pantries were stocked with snacks and premium beverages like coconut water and kombucha. A private chef who cooked meals for top brass also whipped up protein bites, smoothies, and other healthy treats for the rank-and-file, who ordered Seamless on the house for lunch and dinner.
The interest in putting down residential and commercial roots is occurring even as widespread vaccinations hover on the horizon, offering a glimmer of hope for normalcy in New York.
360 Rosemary promises a futuristic "hands-free" environment, with motion sensors and facial recognition technology for a touchless entry, for example, and bathroom faucets that operate with sensors.
Private-equity firm Comvest Partners (also in Chicago), investment firm Norwest Equity Partners (also in Minneapolis), and a coworking space operated by IWG's Spaces brand (headquartered in Switzerland) are already lined up as tenants for the project, which is slated for a grand opening in 2021.
