Executive TL;DR
|
Venture capital built its modern playbook almost entirely around software: capital-light businesses that iterate fast, reach product-market fit within a couple of years, and scale toward an exit within a fund’s typical lifecycle. Deep tech companies those built on genuine scientific or engineering breakthroughs rather than iterative product development are asked to raise capital through that same playbook, from the same funds, on the same timelines. The mismatch this creates is not a minor inconvenience. It is a structural reason capital-intensive, scientifically ambitious companies struggle to get funded even when the underlying technology and market opportunity are sound.
The Numbers Behind the Mismatch
The scale of the gap is stark once quantified directly. According to Bessemer’s XB100 index of leading deep tech companies, the average company requires 9.2 years and $801 million in capital to reach scale compared to a software company, which typically scales on $50-150 million. Deep tech investment writer Leo Polovets frames the underlying cause precisely: hardware companies are capital-intensive upfront, spending heavily to prove technical viability before any revenue exists, while software companies are capital-intensive later, spending on customer acquisition once a product already works. Total capital intensity across a company’s life may be comparable but the timing is inverted, and venture capital’s standard structure was built around the software timing, not the hardware one.
Global deep tech venture funding has grown meaningfully as a result of specialist investor appetite from $18 billion in 2020 to $48 billion in 2025 but that growth has occurred within an investor base that has, if anything, narrowed. Generalist crossover funds like Tiger Global and Coatue, which had expanded into deep tech during the 2021 capital boom, largely retreated from the category after 2021-2022 losses, according to industry analysis from Joltoo’s Deep Tech Funding Playbook. The result: thousands of generalist VCs fund software, while only a small and shrinking fraction have the genuine technical expertise and risk appetite to lead hardware-intensive rounds.
The Structural Root: Fund Life-cycles Built for a Different Asset Class
Traditional venture capital funds operate on 10-year life cycles, with limited partners typically expecting meaningful realized returns within 5-7 years of a fund’s launch a structure built around software’s fast iteration and exit timelines. Deep tech companies routinely require 5-10 years just to reach commercial validation, before the growth phase that would generate the kind of returns VC funds are structured to capture even begins. A fund nearing the end of its investment period has strong structural incentive to avoid a deep tech bet that will not show meaningful progress, let alone a return, until after the fund itself needs to start returning capital to its own limited partners.
This produces a specific and well-documented failure pattern known as the ‘valley of death’: a funding gap between the Technology Readiness Levels roughly TRL 4 through 7 where a technology has moved beyond early lab-scale proof of concept but has not yet achieved full commercial scale validation. Early-stage grants and seed capital can fund the science; growth-stage capital wants to fund an already-validated, scaling product. The years in between, where a promising technology still needs to prove commercial viability at meaningful scale, is precisely where deep tech companies most often run out of runway, because neither category of capital is structurally suited to fund that specific stage.
How the Best-Funded Deep Tech Companies Have Actually Raised
The deep tech companies that have successfully raised at scale illustrate the pattern rather than contradict it: they have generally required either exceptionally patient specialist investors, strategic corporate capital, or unusually large single rounds rather than the standard staged venture progression software companies use. Helion Energy raised $425 million in a single round in January 2025, bringing its total to over $1 billion, racing toward a fusion power commitment to Microsoft by 2028 a multi-year delivery horizon few standard VC funds could underwrite comfortably within their own return timelines. Commonwealth Fusion Systems has raised $863 million including strategic investment from Google and NVIDIA, with Google committing to buy half the output of its planned commercial plant a structure that substitutes a strategic off take commitment for some of the standard financial-return underwriting a fund would otherwise need to perform alone. Anduril, in defense technology, raised a $2.5 billion Series G at a $30.5 billion valuation in mid-2025 and was reportedly raising again at a $60 billion target round sizes that reflect how few investors can write cheques at the scale deep tech increasingly requires.
Why Investors Are Learning to Read Different Signals
A meaningful shift in deep tech investing practice is underway that partially addresses the timeline mismatch without solving it structurally: specialist investors are learning to treat scientific and engineering milestones a successful prototype test, a patent grant, a regulatory pre-approval as deep tech’s equivalent of the product-market-fit signals software investors rely on. New specialist fund structures are also emerging specifically designed for longer hold periods and higher capital calls; Cloud-berry Ventures, for example, launched a EUR 50 million deep tech fund in March 2026 led by a former Google Moonshot X advisor, explicitly built around this longer-horizon thesis rather than a standard 10-year vehicle retrofitted for hardware.
These are meaningful adaptations, but they remain confined to a small and specialised pool of capital. They do not solve the more fundamental mismatch: the vast majority of the world’s long-horizon, patient capital pension funds, sovereign wealth funds, insurance-backed capital still reaches deep tech, when it reaches it at all, indirectly through the same narrow set of specialist venture funds, rather than through structures built to connect that patient capital to deep tech opportunity directly.
What Actually Fixes the Mismatch
Solving this structurally means building capital formation mechanisms around deep tech’s actual risk and cash flow profile, rather than asking deep tech to conform to venture capital’s software-native structure. That means milestone-based capital release tied to scientific and engineering progress a successful pilot, a regulatory approval, a demonstrated yield threshold rather than software-style monthly recurring revenue growth. It means structures capable of accommodating a 9-10 year value creation horizon without requiring every capital provider to commit to that full horizon themselves, through mechanisms that allow some participants to exit earlier while patient capital continues funding the underlying project. And it means distribution reaching patient capital pools pensions, sovereign wealth funds, endowments, insurance-backed capital directly, rather than exclusively through the small number of specialist venture funds currently gate keeping deep tech access on both sides of the transaction.
Blockmaze and Deep Tech Capital Formation
Blockmaze’s infrastructure is built around exactly the structural mismatch this analysis identifies. Rather than forcing deep tech’s multi-year R&D and validation cycles into a fund structure built for software’s faster timelines, the platform supports milestone-based capital release tied to scientific and engineering progress a demonstrated prototype, a regulatory approval, a validated yield threshold rather than growth metrics deep tech companies cannot produce on a software timeline.
Identity-verified allowlists and compliance-enforced transfer restrictions let patient capital pools participate directly in deep tech opportunity without requiring an existing relationship with one of the small number of specialist venture funds currently gatekeeping access, while DAO governance provides the structured oversight a multi-year technical development programme genuinely requires. As deep tech’s capital needs continue to outpace what a narrowing pool of specialist venture funds can underwrite alone, infrastructure built for deep tech’s actual timeline and risk profile not adapted from software’s is what closes the gap.
Frequently Asked Questions
1. How much capital does a typical deep tech company need compared to a software company?
According to Bessemer’s XB100 index, the average top deep tech company requires $801 million and 9.2 years to reach scale, compared to $50-150 million for a typical software company though most deep tech startups will require significantly less than the top-100 average, and many will take even longer.
2. What is the deep tech ‘valley of death’?
The funding gap between early-stage grants and seed capital, which can prove a technology’s scientific viability, and late-stage growth capital, which wants to fund an already-validated, scaling product roughly Technology Readiness Levels 4 through 7, where many promising deep tech companies run out of funding before reaching commercial scale.
3. Why don’t traditional venture capital funds work well for deep tech?
Traditional VC funds operate on 10-year lifecycles with limited partners expecting meaningful returns within 5-7 years a timeline built around software’s fast iteration cycles. Deep tech companies typically require 5-10 years just to reach commercial validation, creating a structural mismatch with fund return timelines.
4. How have the best-funded deep tech companies actually raised capital?
Largely through exceptionally large single rounds, strategic corporate capital, or offtake commitments rather than standard staged venture rounds examples include Helion Energy’s $425 million round backed by a Microsoft power commitment, and Commonwealth Fusion Systems’ funding including Google and NVIDIA with an associated offtake agreement.
5. What would actually fix the capital mismatch for deep tech?
Capital structures built around deep tech’s real risk profile milestone-based capital release tied to scientific and engineering progress, mechanisms allowing some investors to exit before a project’s full multi-year horizon completes, and distribution reaching patient capital pools like pensions and sovereign wealth funds directly rather than only through a small pool of specialist venture funds.
Disclaimer: This article is for informational purposes only and does not constitute financial or legal advice. Readers should conduct their own research and consult with qualified professionals before making any investment or business decisions.