What really prevents deep tech breakthroughs from becoming scalable companies?

Deep tech promises to solve some of the world’s hardest problems, from climate resilience to quantum computing and advanced materials. But turning a breakthrough discovery into a scalable company is often paved with “brilliant failures.” It requires crossing what investors and innovators call the “Valley of Death”: the risky stretch between a validated prototype and a product that can attract customers, generate revenue and secure growth capital.

For decades, the dominant narrative has been that these ventures fail because of the uncertainty of scientific research, long development timelines and, above all, scarcity of funding.

However, a deeper look at the ecosystem suggests a different tension: deep Tech doesn’t fail in the lab; it fails at the interface between scientific discovery and market execution. Brilliant discoveries are not translated early enough into market problems, customer needs, business models and execution strategies. 

The “Valley of Death” is the dangerous middle stretch in a deep tech startup’s life, when a technology has moved beyond the lab but is still far from becoming a reliable, scalable product that customers will buy.  

This phase typically falls around Technology Readiness Levels ((TRLs)4 to 7 — part of a nine-step maturity scale used by space agencies and the European Union to assess how advanced a technology is, from early lab idea, TRL 1, to a fully proven technology in real-world conditions, TRL 9. At this stage, teams are trying to turn prototypes and pilots into solutions that work consistently in real-world environments, just as early grant money is fading and before major investors are ready to commit. 

Official European data suggest that while most Horizon Europe projects (Horizon Europe being the EU’s main research and innovation funding program) begin as basic research or early ideas, only around half reach the demonstration stage by the end. This is why European deep tech support programs increasingly focus on the TRL 4–6 phase, where startups must improve not only their technology readiness, but also their business and investor readiness.  

In Europe, the stakes are particularly high: according to the European Deep Tech Report 2025, deep tech startups attracted around €15 billion in investment in 2024, nearly one-third of all European venture capital. 

So, is funding really deep tech’s biggest gap, or is the deeper challenge learning how to make science survive the market? 

We asked founders, investors and deep tech experts to explore this question. 

When Scientific Excellence Becomes a Blind Spot 

If the Valley of Death is often described as a funding problem, the deep tech founders and experts we spoke to describe a much more complex reality. The missing resource is the ability to translate scientific excellence into a market problem that customers, investors and partners can understand. 

Most deep tech founders emerge from world-class research centers with PhDs in hand and disruptive patents in their pockets. 

They possess what Phil de Luna, CTO of Cura, calls "hard science" intelligence, a world that is definitive, data-driven, and quantitative.  

However, commercializing that science requires a different set of skills that are not inherent to a scientific or engineering education.  

As Oliver Kahl, principal of MIG Capital, observes, academics often have a grand idea of what their technology can do, but when they finally meet the market, the market tells a completely different story.  

This creates a "blind spot" where founders focus on the elegance of the solution rather than the urgency of the problem.  

Rasmus Bankwitz, founder of Link Photonics, admits that he and his co-founder were "two tech nerds" with a revolutionary idea for quantum computing but zero experience in building a company or raising funds. Without a bridge to the business world, even the most groundbreaking technology remains trapped in the lab. 

From Changing the World to Choosing One Market 

The primary hurdle at the science–business interface is often a lack of focus. Scientists are trained to explore every variable, but entrepreneurs must learn to prioritize ruthlessly. Rémi Moriceau, CEO of Fermun Photonics, recalls that before joining a structured acceleration program, his team “wanted to change the world” and address every possible client at once. 

This ambition, while noble, can be fatal for early-stage ventures with limited resources. Deep tech acceleration programs can therefore play a critical role: acting as “obstacle removers” and “narrative crafters” for founders who need to turn technological potential into a clear market proposition. 

They help scientists apply the scientific method not only to their experiments, but also to their business models: testing hypotheses, collecting market data and iterating fast. 

By setting strategic business objectives beyond technology validation, such programs help founders shift their mindset from research and development to sales, scale and market adoption. 

How Can Business Schools Help Deep Tech Reach the Market? 

Crossing the Valley of Death requires business knowledge: the ability to identify real customer needs, define a viable business model, understand investors’ expectations, navigate regulatory constraints and decide which market opportunity to pursue first. 

For deep tech founders, this knowledge can be difficult to acquire alone. Scientists and engineers remain essential to pushing the technology forward, but turning a breakthrough into a company also requires people who can connect that technology to customers, investors, regulators and industrial partners. The challenge is not to replace scientific expertise with business expertise. It is to bring both together early enough for the venture to move from prototype to market. 

This is where the idea of a business school as a “School of Solutions” becomes concrete. Its role is to train people who can act as translators and connectors: people able to understand enough of the science to ask the right questions, enough of the market to test real use cases, and enough of the business logic to turn discovery into strategy, execution and impact. 

HEC Paris Dean Éloïc Peyrache has long advocated a multidisciplinary approach to higher education. In his view, business school graduates must increasingly "master multiple fields" - from artificial intelligence and data to climate and geopolitics - to build bridges between experts and turn knowledge into solutions. 

In deep tech, that connector role can determine whether a breakthrough remains a prototype or becomes a company. 

The value also lies in the strength of the ecosystem. Joint programs between engineering and business schools, such as those offered by HEC Paris in collaboration with École Polytechnique de Paris (in entrepreneurship and data science) or initiatives like CDL-Paris, bring together scientists, entrepreneurs, investors, students and mentors around the same challenge: helping science-based startups test assumptions, navigate regulation, attract capital and find their first customers and markets. 

For Aymeric Penven, Director of the HEC Paris Deep Tech Center, this is not simply a pedagogical exercise. It is a strategic necessity: the only way to turn technological breakthroughs into scalable companies capable of addressing the €1.3 trillion market opportunity for climate and societal resilience. In that sense, the business school’s contribution is not to produce engineers. It is to provide the business knowledge, human connectors and ecosystem that help science survive the market.