Sequoia
and Kleiner Perkins passed on Palantir. Peter Thiel funded it himself instead —
and built a company now trading at roughly 58 times trailing sales.
At a pitch meeting sometime in the
mid-2000s, Sequoia Capital’s Michael Moritz reportedly doodled through the
entire presentation. A Kleiner Perkins executive told the founders their
company was destined to fail. Mainstream Silicon Valley venture capital wanted
no part of Palantir Technologies. Peter Thiel bankrolled roughly $30 million of
the early costs himself. Two decades later, Palantir is an S&P 500 and
Nasdaq-100 component with a market cap around $304 billion — and the story of
why VCs said no is also the story of why the company built a moat almost nobody
else could copy.
Born From
a Fraud Detector, Built for a Market No One Wanted
Palantir’s origin traces back to
PayPal’s internal fraud-detection system, nicknamed “Igor,” which caught the
FBI’s attention after the company had built it to fight financial crime. Thiel,
PayPal’s co-founder, funded a prototype in 2004, built by PayPal engineer
Nathan Gettings and Stanford students Joe Lonsdale and Stephen Cohen. He then
hired Alex Karp — a former Stanford Law School classmate with a PhD in
philosophy from Frankfurt — as CEO, a role Karp still holds today. The idea was
to adapt PayPal’s pattern-detection approach into a platform for the U.S.
intelligence community, at a moment right after 9/11 when most of Silicon
Valley wanted nothing to do with government work. The one institutional
investor that said yes was In-Q-Tel, the CIA’s own venture arm, which put in
about $2 million and helped Palantir reach its first real users inside
government. Thiel named the company after the palantÃri — the seeing-stones of
Tolkien’s Lord of the Rings.
The
Business Model VCs Couldn’t See Coming
What makes Palantir difficult to copy
isn’t its technology alone — it’s two structural barriers stacked on top of
each other. First, security clearance: Palantir’s software is one of a small
number of platforms authorized for Impact Level 5 classified U.S. Department of
Defense workloads, an accreditation that takes years to earn and that most
startups simply can’t get. Second, the Forward Deployed Engineer model —
Palantir staff embedded directly with customers to configure the platform for
their specific workflows. It’s a labor-intensive, expensive way to sell
software, and it looks unscalable on a spreadsheet. But it wires Palantir’s
product into a customer’s actual operations in a way that’s very expensive to
rip out. The lesson, as the dossier puts it, is that high-touch delivery can
beat pure SaaS — worth remembering before assuming “scalable” always beats
“high-touch.”
That same underlying model — an
“ontology” data layer mapping relationships across an organization’s messy data
— has now shipped across four products spanning three technology eras: Gotham
for intelligence work, Foundry for the big-data era, Apollo for continuous
software delivery, and AIP, the generative-AI platform launched in 2023 that is
now the company’s primary growth driver. The wrapper changed three times. The
core didn’t.
From a
34% Loss to a 36% Profit Margin — and a Valuation That Prices In Perfection
The financial turnaround is real and
dramatic: Palantir moved from a 34% net loss margin in 2021 to a 36% net profit
margin in 2025, as revenue scaled from $1.5 billion to $4.5 billion on a mostly
fixed cost base. Commercial revenue, growing far faster than government
revenue, now makes up roughly 45% of the total — a mix shift that changes both
the company’s long-term growth ceiling and its risk profile.
But operating leverage and valuation are
two separate questions, and the second is where caution belongs. Even after a
28% pullback from its December 2025 peak, Palantir trades at roughly 58 times
trailing sales and 142 times trailing earnings — multiples that assume years of
continued 50%-plus growth with no stumbles, against an S&P 500 that trades
around 20 to 25 times earnings. There’s also a gap worth noting between free
cash flow ($2.1 billion in FY2025) and GAAP net income ($1.6 billion): a real
driver is stock-based compensation, added back in the cash flow statement but a
genuine, ongoing cost to shareholders through dilution. Shares outstanding grew
4.7% in FY2025 alone.
The Risk the
Numbers Don’t Capture
Roughly 55% of FY2025 revenue is tied
to government contracts, making a change in administration or contracting
posture a real, non-diversifiable risk. Karp and Thiel are unusually outspoken
public figures, and Palantir’s government work — including ICE-related
contracts, UK NHS data contracts, and a strategic partnership with the Israel
Defense Forces — has drawn sustained protest and scrutiny. Karp, Thiel, and
Cohen hold supervoting shares, meaning public shareholders have limited
influence over decisions that can move the stock. And the 2022 episode is a
live reminder of how fast sentiment can turn: Palantir’s market cap fell from
$36.9 billion to $13.5 billion, a roughly 63% drawdown, when growth and macro
conditions shifted.
Watch the Full Breakdown
Palantir’s path — from a company VCs
wouldn’t fund to one of the most expensively priced software stocks in the
market — is a case study in how contrarian markets can become moats. For the
full financial charts, the valuation math, and how Palantir stacks up against
Snowflake, Databricks, and C3.ai, watch the complete deep-dive video on the
Company Narratives YouTube channel.
Comments
Post a Comment