From lab bench to shop shelf: how scientific discoveries become products

Why the pipeline matters

Every so often a headline announces a scientific breakthrough: a new battery material, a gene-editing technique, a quantum sensor. Yet years, sometimes decades, can pass before such discoveries reach a market, if they ever do. Understanding this pipeline matters because it explains why promising research so often stalls, why governments and investors talk about “deep tech” as a distinct category, and why patience is as important as ingenuity in turning science into economic and social benefit.

Deep tech differs from software-driven innovation because it is usually rooted in fundamental science or engineering, physics, chemistry, biology, materials science, rather than clever applications of existing digital tools. That makes the path from idea to product longer, more capital-intensive and riskier, but potentially more transformative when it succeeds.

From discovery to prototype

The pipeline typically begins in a university or research institute, where a scientist observes an effect or develops a method with no immediate commercial purpose in mind. This early stage is usually funded by public research grants, charitable foundations or research councils, because the outcome is too uncertain and too far from market for private investors to back it.

If the discovery looks promising, researchers or a technology transfer office will often file a patent to protect the intellectual property before publishing further. The next step is proving the idea can work outside a controlled laboratory setting, building a basic prototype or demonstrating the effect at a slightly larger scale. This is where many discoveries quietly die: a phenomenon that works beautifully with a single sample under ideal conditions may prove impossible, or prohibitively expensive, to reproduce reliably.

Crossing the valley of death

The gap between a working prototype and a manufacturable, reliable product is so notoriously difficult to bridge that it has earned its own nickname among investors and policymakers: the valley of death. At this stage, the technology needs serious engineering work, testing for safety and durability, and often entirely new manufacturing processes. Costs rise sharply, but the technology is usually still too immature and unproven to attract mainstream venture capital or corporate investment, which tends to favour ventures with a clearer, faster route to revenue.

Specialist funding sources exist to help bridge this gap, including government innovation agencies, patient capital funds willing to accept longer timelines, university spin-out schemes and, in some sectors, defence or industrial partners interested in early access to novel capabilities. Founders at this stage often need to build a spin-out company, recruit engineers alongside scientists, and start engaging early with regulators or standards bodies if their product will need approval, as is the case for medical devices, new drugs or novel materials used in construction or aviation.

Scaling, regulation and adoption

Assuming a working, reliable product emerges, the next challenge is manufacturing it at scale without losing performance or blowing the budget. Many deep-tech ventures that succeed scientifically fail commercially because scaling up a process that worked in a small batch turns out to be far harder, or far more expensive, than anticipated. This is particularly true in areas such as advanced materials, batteries and biotechnology, where chemistry that behaves predictably at laboratory volumes can behave differently at industrial volumes.

Regulation and standards form another significant hurdle. New medicines must pass clinical trials and gain approval from bodies such as the Medicines and Healthcare products Regulatory Agency. New materials or components may need to meet safety standards before they can be used in buildings, vehicles or consumer goods. These processes exist to protect the public, but they add time and cost that must be planned for from an early stage rather than treated as an afterthought.

Finally, even a safe, scalable, well-engineered product must find customers willing to adopt it, which can mean displacing an established incumbent technology, changing supply chains, or persuading conservative industries to take a risk on something unfamiliar. Government procurement, industry consortia and early adopter customers all play a role in helping genuinely useful deep-tech products complete the final stretch from laboratory curiosity to everyday reality.