Last week, defense-tech startup Anduril announced a $5 billion Series H at a $61 billion valuation. That’s more than double what the company was worth less than a year ago, and investors just paid about $28 for every $1 of Anduril’s trailing revenue. For perspective, Lockheed Martin — the largest U.S. defense contractor, with about $75 billion in annual revenue — trades at $1.60.
The explanation to those staggering numbers is hiding in how Washington buys.
What The Multiple Is Telling You
Anduril is a real company doing real revenue: more than $2 billion in 2025, up 110% from the prior year. It’s also raised more than $11 billion in less than a decade — more than most defense companies raise in their first century. Lattice — Anduril’s AI command software — runs counter-drone, missile-defense and battle-management programs for the U.S. Army, the Royal Australian Navy and the Dutch Ministry of Defense. The company is also a contractor on the new ‘golden dome’ space-based missile shield. The round was co-led by Andreessen Horowitz, already a defense-tech bull through its American Dynamism fund, and Thrive Capital, Joshua Kushner’s consumer- and AI-leaning firm. Thrive’s participation is what’s new. When a fund without a defense thesis co-leads a $5 billion defense round, it tells you the category isn’t on the periphery of venture investing anymore.
But none of that — the revenue, the customers, the lead-investor lineup — explains the multiple gap to Lockheed. The only thing that does is a forward bet on something most coverage hasn’t yet priced: the U.S. government has started buying at a clock speed that finally matches venture velocity.
The Procurement-Speed Unlock
In March, the Army awarded Anduril a 10-year, $20 billion enterprise contract. Buried in the announcement was a number worth re-reading: that single contract consolidated more than 120 separate Anduril procurement actions into one acquisition vehicle.
Anduril isn’t alone. The Army has signed 14 enterprise contracts in the last eight months, replacing 118 individual contracts in the process — an 88% reduction in contract volume across its commercial-technology buying. Russell Vought, the White House budget director, called the new $1.5 trillion FY26 defense budget ‘paradigm-shifting’ specifically because it allows multiyear contracts at a scale Congress hadn’t previously authorized.
Programs that used to take seven years to award now move in months. Contracts that used to require fresh paperwork for every new product version now run under one agreement. The clock speed of the buyer just compressed by an order of magnitude.
Anduril was built for this. Small teams, fast iteration, software-first products that can ship in weeks instead of years — exactly the company shape the new contracts are designed to buy. Schimpf, Anduril’s CEO, told Andreessen Horowitz in a recent interview that the F-35 took 25 years to move from initial concept to fielding. Commercial aircraft and autos run on 2–3 year cycles.
Schimpf isn’t pitching, he’s describing operating reality. The $4.3 billion revenue forecast he just put on the record can only materialize if the new procurement rules hold, and the $5 billion the company just raised isn’t going to R&D. It’s going to manufacturing — production capacity built on the assumption that the buyer’s new pace is permanent.
Why Generalist Money Showed Up
The category signal in this round isn’t the dollar amount. It’s the co-leads. American Dynamism — the a16z fund Katherine Boyle built on exactly this thesis four years ago — has been waiting for the volume of evidence. This round is it.
Thrive Capital is the new face at the table. Joshua Kushner’s firm primarily writes consumer, fintech and AI infrastructure checks (OpenAI, Stripe and Instacart) and doesn’t have a dedicated defense vehicle. Its co-lead position says generalist money is now showing up to a category that two years ago was reserved for specialists.
Why now? the buyer-side timeline finally matches the math that lets venture funds underwrite a bet. Venture funds run on ten-year lifecycles. Defense procurement used to take seven years just to reach scaled revenue — which meant a venture investment couldn’t compound inside one fund’s life. Procurement reform compresses that timeline, and helps make the math now work.
That shift reshapes the next ten companies. The standard reason for passing on a dual-use deep-tech pitch was ‘great market, slow buyer.’ The slow-buyer half just stopped being true. Founders building in autonomous maritime, secure compute or biotech for warfighter health won’t hit the funding ceiling they would have a year ago.
What Could Slow The Clock Back Down
The multiple isn’t permanent. The new procurement model can reverse. An administration change, a defense-budget reversal or a major Lattice program stumble all sit in the path.
So does the response of the incumbents. The Big 5 primes — Lockheed, RTX, Northrop, General Dynamics and Boeing — collectively run on more than $250 billion in annual revenue built around the old contract shape. They won’t watch the rules get rewritten without fighting back, and ‘fighting back’ in defense isn’t a press release. It’s permanent lobbying operations on the Hill, factories in every congressional district that matters to an appropriations vote and decades of Pentagon relationships that don’t reset overnight.
There’s also the simpler execution risk. Schimpf’s $4.3 billion 2026 projection is already a near-doubling. A miss of even a quarter or two would matter — at 14 times forward revenue, the cushion is thin.
The interesting part is which direction this risk runs. Five years ago, the bet was that the buyer would eventually speed up — a speculation that required something new to happen. Today, the bet is that the buyer, having sped up, stays at the new pace. That’s a continuity bet, and continuity bets historically price higher. That’s the one investors who came in at $30 billion last year and $61 billion this year are making.
The closest historical analog is NASA’s 2006 commercial cargo program. The procurement-rule change — letting NASA buy launches as a service instead of as a program — was what let SpaceX become a real company, years before any specific launch milestone made the multiple look obvious. The valuation compounded for a decade. The only thing that had to keep happening was the rule.
Anduril at $61 billion is a bet on Washington buying like venture investors buy — fast, repeated and at scale. If the clock holds, the multiple isn’t expensive but if it doesn’t, Lockheed at 1.6 isn’t cheap. The price tag on Anduril is just the first one written in the new units.
