What sits inside Dealroom's deep tech universe, how many deep tech and life sciences spinouts Europe and the US have produced and how much they have raised, how their follow-on rounds are tracked, and how finely the data can be cut.
Dealroom platform data pulled 29 October 2026, company HQ or founding location. Total raised excludes debt, IPO proceeds and acquisitions. Spinout cuts include companies that have since matured. Live platform figures move daily and will differ slightly from these.
Every company carrying the deep tech technology tag has been classified by Dealroom's Intelligence Unit against a published definition, and the tag is reviewed as technologies mature.
Dealroom defines deep tech as technology based on tangible engineering innovation or scientific advances and discoveries applied for the first time as a product. A company qualifies when it meets two primary criteria and, typically, a third.
The technology takes a long time to reach market-ready maturity because the R&D is novel and complex. The company employs highly educated staff early and often creates new hardware or intellectual property.
Substantial investment is needed to research, develop, test and scale the product or service before it generates revenue.
The technology can create significant positive change in health, environment, security or quality of life.
The segment comes first: quantum hardware, nuclear fusion or semiconductors qualify on their own. For borderline segments, analysts look at patents, spinout status, whether founders hold PhDs or scientific roles, and the business model (manufacturing raises the likelihood, SaaS lowers it, marketplaces almost never qualify). Patents are a supporting signal rather than a requirement: only about a quarter of funded deep tech startups hold one. LLM-assisted screening speeds up review; every final classification is a human decision.
What counted as deep tech a decade ago can be mainstream today. Lithium-ion storage and early machine learning sit differently from solid-state chemistries or foundational models. Classification reflects when the underlying technology was developed and commercialised, so the universe stays a frontier rather than an archive.
Dealroom's taxonomy has 31 top-level industries, 18 mutually exclusive sectors and roughly 2,800 granular tags. Deep tech cuts across them; this is how the Intelligence Unit applies it industry by industry.
| Industry | Deep tech? | Always included | Case by case | Not deep tech |
|---|---|---|---|---|
| Semiconductors & quantum | Mostly yes | All semiconductors and photonics; quantum computing hardware and software, sensing, cryptography, communication | Rarely borderline | Component marketplaces |
| Space | Mostly yes | Upstream: launch, satellites, space operations | Downstream (e.g. earth observation) when IP-intensive | Resellers and services |
| Robotics | Mostly yes | Humanoids, collaborative robots, surgical robotics | Farm and warehouse robotics, advanced sensors, advanced drones | Consumer robots, RPA |
| Energy | Sometimes | Nuclear fusion and fission | Advanced hydrogen, carbon capture, novel battery chemistries, deep geothermal, grid hardware, novel mining tech | Standard solar and wind, energy SaaS, marketplaces |
| Transportation | Sometimes | eVTOL, autonomous aircraft, sustainable aviation | Autonomous vehicle stack, advanced drones, novel EV batteries and motors | EV services, micromobility, apps |
| Health & life sciences | Sometimes | AI drug discovery, brain-computer interfaces, surgical robotics, synthetic biology | Devices and imaging with novel proprietary technology | Pure pharma and biotech, telemedicine, medical SaaS |
| Food & agriculture | Sometimes | Cultivated meat, synthetic biology | Precision fermentation, microbial tech, advanced farm robotics, in-field biosensors | Delivery, restaurant software |
| Chemicals & materials | Sometimes | Judged case by case | Green polymers, e-fuels, advanced additive manufacturing, novel materials | Distribution, commodity production |
| AI | Mostly not | Foundational models | LLM infrastructure, deep scientific AI | AI-powered applications |
| Blockchain, fintech, security | Mostly not | Layer 1 and 2 protocols, zero-knowledge proofs, hardware security | Privacy-enhancing technologies, core blockchain infrastructure | Apps, credit scoring, cybersecurity SaaS |
| Enterprise software, media, marketplaces | Mostly not | Rarely | Foundational software, green cement, synbio materials | Almost everything else |
Within health and life sciences, the sub-tags most relevant to a techbio thesis are AI drug discovery, computational biology and chemistry ("compbio"), synthetic biology, brain-computer interfaces, surgical robotics, and quantum computing for chemistry and pharma. Curated landscapes already exist for AI drug discovery (global and European), computational biology and chemistry in Europe, synthetic biology, brain-computer interfaces and quantum computing.
Dealroom's definition measures the value universities and research centres create, which is the same lens an investor applies.
A spinout is a company that originates from research carried out in a university or research centre. The institution often holds equity, but that is not required: royalty-only or licensing agreements, and cases with no formal agreement but a clear foundational link to the institution's research (including professor-privilege countries such as Sweden), all count.
Alumni-founded companies, companies only incubated or accelerated by a university, and companies that licensed peripheral technology outside their core starting IP are not spinouts. Corporate spinouts are tracked under a separate round type. Closed spinouts remain in the data so trends are not biased towards survivors.
Links come from technology transfer and commercialisation offices and national TTO associations across the UK and Europe; national authorities such as MESR and CNRS in France, Techleap in the Netherlands, MCW in Spain and CNR in Italy; national registers such as the UK's HESA spinout register; institute websites and reports; financial filings where an institute is on the cap table; Dealroom's own review of company websites, press and founder backgrounds; and the WIPO Global Innovation Index partnership, where Dealroom is the data partner on research commercialisation.
Every contribution is reviewed by Dealroom, which keeps the final judgment. A company is tagged only when a reliable source explicitly identifies it as a spinout or as built on the institution's technology.
Deep tech and life sciences spinouts by company count and total VC raised. Europe has produced 70% more spinouts; the US has raised nearly three times the capital, and $189bn of the US total sits in health, most of it biotech and pharma rather than deep tech.
Filters: tag "dt and ls" and a Spinout round, company HQ or founding location in Europe or the US. Bars within each row are scaled to the larger value. Pulled 29 October 2026. Published context from the European Spinouts Report 2025: European deep tech and life sciences spinouts carry a combined enterprise value of $398bn, 84% of all European spinout value and 7x the 2016 level.
Bars show company count; a company can sit in more than one industry. The same cut runs for the US, any country or any founding cohort.
| Europe | Spinouts | Raised |
|---|---|---|
| United Kingdom | 1,780 | $32.6bn |
| France | 1,476 | $14.0bn |
| Germany | 962 | $18.3bn |
| Switzerland | 955 | $12.0bn |
| Netherlands | 606 | $6.4bn |
| United States | Spinouts |
|---|---|
| California | 1,441 |
| Massachusetts | 867 |
| New York | 338 |
| Pennsylvania | 250 |
| Texas | 237 |
Deep tech and life sciences spinouts linked to each institution.
| Europe | Spinouts |
|---|---|
| CNRS | 789 |
| ETH Zurich | 342 |
| EPFL | 305 |
| University of Cambridge | 265 |
| University of Oxford | 225 |
| United States | Spinouts |
|---|---|
| MIT | 359 |
| Stanford University | 333 |
| Harvard University | 195 |
| UC Berkeley | 194 |
| University of Michigan | 127 |
In the European Spinouts Report 2025, the top five institutions by value created were Oxford, Cambridge, ETH Zurich, EPFL and TU Munich.
Spinout counts are solid on both sides of the Atlantic. The asymmetries worth knowing: US dollars are concentrated in biotech and pharma, so a like-for-like deep tech comparison should exclude pure life sciences; and Europe's institutional links are deepest for recent cohorts (3,875 spinouts founded since 2019 against 1,254 in the US), reflecting Dealroom's direct partnerships with European TTOs and national agencies. US institutional links continue to be extended through the WIPO programme, and priority US institutions can be scoped together during onboarding.
A spinout is recorded as a round of type "Spinout" on the company profile, with the university or research institution as the investor, dated to the month of launch. Every later round links back to that origin.
Share of the 8,908 European spinouts that have raised each round type, and the capital raised at that stage. Grants have touched 3,067 of them ($8.7bn), the single most common form of early capital.
Percentages are of all European deep tech and life sciences spinouts in the database, including those founded recently and still pre-seed. The same ladder runs for the US, a country, an institution or a founding-year cohort.
Company level: the full round history of any spinout, with investors per round, from spinout to Series A and beyond.
Aggregate level: funding heat maps filtered to spinouts, by founding cohort, stage, year, country, sector or investor type. Amount or round count per period, exportable.
Investor level: which funds are most active in spinout follow-ons by sector and stage, for co-investor and competitor mapping.
86% of early-stage capital into European spinouts comes from within Europe, while nearly half of late-stage capital comes from outside, mainly the US. US corporates and funds have acquired nearly $24bn of European spinout value since 2019 (European Spinouts Report 2025). Across deep tech as a whole, the Transatlantic Deep Tech Report 2026 puts US funding at $141bn in 2025 against $15.5bn in Europe: roughly 2x at early stage and 16x at late stage, even though Europe closes more rounds.
Undisclosed round amounts are estimated using a documented method and flagged as estimates, so they can be excluded from any benchmark.
Benchmarks nest from the whole universe to a single sub-tag, cohort and city, and the same filters apply across search, heat maps, saved searches and exports.
Every company carries HQ and founding location at country, region and city level. The "founding or HQ" logic means a company founded in Munich that later redomiciled to Delaware is still counted as European, a common reason other datasets undercount European deep tech. Comparisons run to country, state or city level, and the Universities page filters by country and region.
Cuts run by industry, sub-industry, sector tag and technology tag, nested as finely as needed: deep tech, then health, then AI drug discovery, then a founding-year cohort, then a stage. Heat maps can be built on any set of tags, with amount or round count per year, and saved as recurring reports.
dt and ls (or technology deep tech for deep tech only), round type SPINOUT with the institution as investor. Add HQ or founding location, launch-year range, industry or sub-tag, and stage.Example export row (fields available on every company)
| Company | Institution | HQ | Industry / sub-tags | Technology | Founded | Rounds | Latest round | Investors |
|---|---|---|---|---|---|---|---|---|
| ICEYE | Aalto University (Spinout) | Espoo, Finland | Space / earth observation satellites, SAR | Deep tech | 2014 | Spinout, Seed, Series A, B, C, D, growth | Growth equity | Lead and participating investors per round |
Further fields on export: total funding, valuation where disclosed, estimate flags, employee count and growth, patents, founder education and PhD flag, Dealroom Signal, and the institution's location. Other European examples with a Spinout round: BioNTech (Mainz), Oxford Nanopore (Oxford), CureVac (Tübingen).
Very early or stealth companies can be in the database through founder data before they carry a deep tech tag. The recommended sourcing set pairs tagged Deep Tech & Life Sciences spinouts with spinouts not yet tagged and with recent PhD founders from technical disciplines, so nothing is missed at pre-seed.
Saved searches with alerts, exports with customisable fields, pre-built landscapes, heat maps as recurring reports, and a natural-language connector. Onboarding includes building the benchmark set and recurring reports with you.
A live walkthrough is the fastest way to see how these pieces fit a specific thesis. We would propose working through them in this order.