Some early Hard Tech VC predictions/thoughts heading into 2027... https://x.com/i/article/2119857360566915072
Hard Tech VC: Predictions for 2027
Hard Tech VC: Predictions for 2027
My bet: more capital enters hard tech next year, yet most founders won’t feel it.
A handful of companies will raise enough to build entire industrial platforms. Everyone else will need a much sharper answer to what the next dollar actually proves.
Few things I’m watching:
LP co-investment becomes a bigger part of the bargain. Think more LPs will want fund exposure and direct ownership in breakout companies. Especially when a factory expansion or production ramp creates room for a meaningful check. Everyone wants more allocation once the winner looks obvious. GPs need to be careful about promising away their best economics to close a fund. If your fund takes the early risk, it needs enough ownership left when that risk pays off. And LPs should ask: am I getting a genuinely attractive opportunity, or financing the piece nobody else wants?
Hard tech funds split between small specialists and massive platforms. Middle gets uncomfortable. Small funds can win through early ownership, narrow expertise, and outcomes that actually move their fund. Massive platforms can finance development, manufacturing, and several follow-on rounds. Stuck between those models is tough Too much capital to make small outcomes matter. Not enough to support an expensive production ramp yourself. Rather see a manager raise the right-sized fund and know exactly where they hand companies off than stretch to some arbitrary institutional threshold. Bigger fund doesn’t fix a confused strategy. Makes the confusion more expensive.
Series A increasingly funds a repeatable business. Founders still need to sell the enormous vision I want the ambition. But I also want to know exactly what this round buys. Second deployment with fewer engineering hours? Qualification completed? Yield improved? Customer acceptance that unlocks payment? “Finish R&D, explore three markets, build a factory, and launch another product” is a lot of unresolved shit bundled into one financing. My bet is strong founders get better at sequencing those risks. Start with a commercial foothold that proves something valuable and transferable A small first market doesn’t bother me nearly as much as a giant market with no credible way into it.
Growth investors stop treating revenue growth as proof the business is getting better. You can double revenue and barely reduce uncertainty. One development contract. A giant installation. Customer-funded engineering. Looks great in a fundraising deck. Still tells me very little about repeat demand, margins, or cash conversion. Think growth diligence gets much more intrusive here. What happens after installation? How much support does the customer need? When do you actually get paid? Does deployment five cost less than deployment two? Some companies growing slower will raise better because each sale strengthens the business. Others will discover they’ve built a very expensive custom engineering shop.
Everyone piles into industrial bottlenecks. Some investors mistake a shortage for a moat. I like electrical equipment, thermal management, precision components, testing, and specialized manufacturing. But “customers can’t get enough of this” needs a follow-up question: what happens when they can? Exceptional margins during a supply shortage don’t automatically survive new capacity. I want to understand what remains when lead times normalize. Better performance? Lower costs? Qualification customers can’t easily replicate? Reliability nobody else can deliver? Plenty of money will get made removing bottlenecks. Plenty will also get invested at valuations that assume the bottleneck lasts forever.
Defense stays hot. The blanket valuation premium gets harder to defend. I’m bullish on real demand, but much less comfortable with prices that assume every procurement milestone goes right lol. Operational need, funded procurement, development contracts, and political enthusiasm are four different things. We keep treating them like they’re interchangeable. They aren’t. I think investors get much more specific about who owns the budget, what’s funded, and how a deployment becomes production revenue. Defense spending can grow while a lot of defense investments disappoint. If you pay for procurement success before it happens, you’d better be fucking right.
“Physical AI” keeps opening doors. Eventually investors ask which business you actually own. Model company? Hardware manufacturer? Systems integrator? Field-service business? You can combine those and build something exceptional. You can also combine them and make weak economics difficult to spot. My bet is fundraising starts splitting along those lines. If you’re building intelligence, show me learning that transfers across deployments. If you’re deploying machines, show me repeatable installation and support. If you’re manufacturing hardware, show me a production advantage. Some founders will raise better once they get precise about where they win. Calling every layer a platform won’t do the work.
Debt becomes more important. Founders discover “we’ll finance the factory later” isn’t a plan. Lenders need something they can underwrite. Repayment visibility. Useful collateral. Enforceable contracts. Predictable cash flow. A warehouse full of custom equipment and an order book with cancellation rights may not get you very far. I think stronger founders start designing for financing eligibility earlier: customer deposits, clear acceptance milestones, reusable equipment, proven production economics. Venture equity should buy down the risks other capital won’t take. Burning it indefinitely on expansion because nobody planned the financing handoff gets expensive fast.
Political tailwinds matter. Getting paid matters more. I expect a messier funding environment under divided government even for sectors with broad strategic support. Proposed budgets aren’t purchase orders. Announced awards aren’t cash in the bank. A friendly administration doesn’t eliminate appropriations risk. The issue doesn’t have to be a policy reversal. Six months of delay can force a brutal financing while everyone still agrees your company matters. I’d pay a premium for multiple routes to demand, durable support across political constituencies, and enough runway to survive Washington’s calendar. “America needs this” is a reason to care. Founders still need to know who pays, how much, and when.
LPs get less impressed by portfolio logos and more interested in what GPs actually own and return. Being in a great company and making a great investment are different accomplishments. Entry price matters. Ownership matters. Follow-on requirements matter. So does waiting another five years for liquidity. I think LP diligence gets sharper about all of it. Secondaries become a real judgment call: return some capital and establish credibility or preserve exposure to a company that could drive the whole fund. There’s no universal answer. But GPs should explain why they sold with the same conviction they explain why they bought. At some point, the mark needs to become money. My posture heading into 2027: big ambition, very specific underwriting. I want founders attacking enormous problems. I also want to know what each round proves, who funds the next phase, and how ownership survives the journey.
A shit ton of capital entering hard tech doesn’t mean every company gets funded or every fund earns a great return.