How to Commercialise Deep Tech: From Technology to Market
By Dr Chloe Sharp
Deep tech commercialisation rarely starts with a finished product and a clearly defined customer.
A university spinout might begin with a scientific discovery. A biomaterials company might have a platform technology with applications across cosmetics, food, healthcare and industrial manufacturing. A sensing technology could potentially solve problems in agriculture, defence, infrastructure or healthcare.
The challenge is not simply how to sell the technology.
It is working out where the technology can create enough value to become a viable business, and generating the evidence to prove it.
That means technical development and commercial development need to happen alongside one another.
A higher Technology Readiness Level (TRL) tells you something about whether the technology works. It doesn't necessarily tell you whether customers need it, whether the economics work, whether it can navigate regulation or procurement, or whether anyone will pay enough to create a sustainable business.
So how do you commercialise deep tech?
1. Understand what you actually have
The starting point is the technology itself. What is genuinely novel? What does it make possible? What advantage does it provide over existing technologies? What intellectual property exists around it, and what constraints might affect its commercial development?
You also need to consider the possible route to commercialisation. Could the technology become:
a standalone product
a component within somebody else's product
a manufacturing process
intellectual property that is licensed
a joint development opportunity
a platform supporting several future applications
a new spinout business
These choices have very different implications for the market, funding, team and infrastructure required. For research-led innovation, UKRI similarly distinguishes between different routes to commercialisation, including licensing intellectual property and creating a new venture or spinout.
2. Map the possible applications
Deep tech often has the opposite problem to a conventional startup. Instead of a problem then solution, you may initially have a technology then multiple possible problems and applications. A new material, manufacturing technique, biological process or sensing technology may be technically capable of serving several industries.
But being technically applicable doesn't make every market commercially attractive. The job is to map the opportunity space and ask:
Where does the technology create the greatest value?
Which problems are important enough for organisations to act on?
Where does the technology have a meaningful advantage?
Which markets are accessible?
How difficult is adoption?
What regulatory or manufacturing barriers exist?
How much capital will be required?
How long could reaching revenue take?
This allows founders to narrow a large application space towards a smaller number of credible opportunities.
3. Choose a beachhead market
Eventually, a choice has to be made. Trying to commercialise into five industries simultaneously is usually a recipe for spreading scarce resources too thinly and a disaster waiting to happen as your focus is so spread. The aim is to find a beachhead market: somewhere narrow enough to focus the business, but valuable enough to establish a commercially meaningful position.
That decision shouldn't be based on market size alone. A £10 billion market may be less attractive than a £500 million market if accessing it requires ten years of regulatory development, enormous capital investment and replacing deeply embedded infrastructure. Founders need to consider market attractiveness alongside technical fit, competitive advantage, customer urgency, adoption barriers, economics and their ability to reach the market.
This is an initial hypothesis rather than an irreversible decision. The next step is to test it.
4. Understand the commercial ecosystem
Deep tech commercialisation usually involves more than identifying a "customer". So consider a technology being introduced into a manufacturing environment or supply chain or system.
The person benefiting from it might be different from the person buying it. Procurement may need to approve it. Engineers may need to integrate it. Operations teams may need to change their processes. A regulator may have to approve it. A manufacturer or distributor might be required to get it to market.
So customer discovery needs to investigate the wider system: Who uses it? Who benefits? Who buys? Who influences? Who approves? Who integrates? Who can block adoption?
This is one area where the principles behind UKRI's ICURe programme are particularly useful. ICURe applies the scientific method to commercialisation: forming hypotheses about the market, testing them through extensive market engagement and using the evidence to determine what to do next. Current ICURe programmes explicitly encourage teams to investigate customers, competitors, supply chains and regulators rather than looking only at the end user.
For deep tech founders, that ecosystem understanding can be as important as understanding the technology itself.
5. Validate whether there is real demand
Customer conversations are useful, but people saying "that's interesting" isn't commercial validation. The next step is finding out whether customers care enough to commit something meaningful. Depending on the technology and stage, evidence might include:
introducing you to a decision-maker
sharing internal data
providing samples or access to equipment
agreeing to technical requirements
offering access to a production environment
signing a letter of intent
entering a joint development agreement
participating in a feasibility study
running a pilot
paying for a trial
signing a commercial contract
Not all evidence is equal. The more effort, money, organisational risk or reputation someone is prepared to commit, the stronger the evidence tends to become. At the same time, founders need to test the commercial model: Who pays? For what? How much? Is the technology sold, licensed or provided as a service? Are customers buying a component, an outcome or access to intellectual property?
UKRI's current commercialisation guidance takes a similar approach. Before deciding on a route to commercialisation, it recommends understanding the market opportunity and target market while demonstrating both the technical feasibility and financial viability of the intended product, process or service.
6. Prove the whole proposition
This is where technical and commercial development increasingly converge. A laboratory demonstration may prove that the science works, but it doesn't necessarily prove that the business works.
Depending on the technology, founders may now need evidence across several dimensions:
Technical: Does the technology reliably achieve the required performance?
Customer: Does it produce an outcome customers genuinely value?
Integration: Can it work inside the customer's existing environment or workflow?
Manufacturing: Can it be produced reliably and at the required volume?
Economics: Can it be produced and delivered at a cost that supports a viable business model?
Regulation: What approvals, standards or certifications are needed?
Intellectual property: Is there sufficient protection and freedom to operate?
Environmental impact: Does the technology deliver the environmental improvements being claimed?
Tools such as techno-economic assessment (TEA) and lifecycle assessment (LCA) may become increasingly important here, particularly for climate tech, engineering biology, advanced materials and manufacturing technologies. This is why technical readiness and commercial readiness should not be treated as the same thing.
UKRI's Proof of Concept approach explicitly describes this stage as developing both technical and commercial readiness. It combines prototype development with activities such as market validation, intellectual property management and preparation for spinout creation or other routes to market.
7. Build the funding strategy around the evidence
Deep tech businesses frequently need significant investment long before they generate substantial revenue. That might involve a sequence of research funding, translational funding, innovation grants, customer-funded development, strategic investment, equity investment and eventually growth capital.
But funding shouldn't become a separate activity from commercialisation. The more useful question is: What does the business need to prove next, and what funding will enable it to generate that evidence?
A grant might fund a prototype, a customer-funded feasibility project might prove integration, seed investment might fund regulatory development and initial hires, and a Series A might finance manufacturing capacity and entry into a proven market.
The funding strategy should therefore follow the commercialisation strategy rather than replace it. This direction is increasingly visible in UK innovation policy. UKRI's 2026–27 Delivery Plan describes investment aimed at closing early commercialisation gaps and specifically talks about building commercial readiness alongside technical development, while connecting promising deep-tech businesses with specialist private capital as they progress.
8. Enter the market
Once there is credible evidence that the proposition works, attention shifts towards making buying repeatable.
That means developing:
positioning
pricing
sales processes
partnerships
distribution
procurement routes
implementation processes
customer onboarding
pilot-to-paid conversion
The important question changes from: "Can we find someone interested in this?" to: "Do we understand how this gets bought?" For deep tech, the route to market may depend heavily on strategic partnerships, manufacturers, distributors, systems integrators or major industry customers rather than direct selling alone.
9. Scale and revisit the opportunity
Once the first market works, another challenge begins. The company may need to build manufacturing capacity, strengthen its supply chain, implement quality systems, secure new regulatory approvals, develop commercial capability and raise growth capital. And for platform technologies, commercialisation often loops back on itself.
A technology might initially succeed in cosmetics because the route to market is relatively accessible, before expanding into medical devices or pharmaceuticals where the opportunity is larger but development and regulatory requirements are greater.
The first successful market becomes evidence for the next. Commercialisation is rarely a straight line and is messier than you’d expect. But if you’re willing to be adaptable and keep learning, that can help!
How this fits the Sharp Insights Commercialisation Framework
This process we just mapped out is why we developed the Sharp Insights Commercialisation Framework around six stages:
MAP → EXPLORE → VALIDATE → PROVE → ENTER → GROW
It isn't intended to suggest that every company follows six stages neatly from left to right. Instead, it helps founders understand what they need to know next and what evidence they need before making the next major commitment.
MAP: Choose where to focus
For deep tech, this begins with the technology, potential applications, market opportunities, intellectual property, business models, funding position and constraints. A platform technology may have numerous possible markets.
MAP asks: Where might the strongest commercial opportunity lie, and what assumptions are we currently making about it?
This is where application mapping and early beachhead selection begin.
EXPLORE: Understand the opportunity
Once promising applications have been identified, founders need to understand the market properly. That means moving beyond market reports towards customers, buyers, partners, regulators, procurement, competitors and the wider value chain.
This stage is particularly closely aligned with the principles behind ICURe: treating assumptions about the market as hypotheses that need to be investigated through real-world market engagement. The objective is to establish whether there is a valuable problem, who experiences it most strongly and where the best beachhead might be.
VALIDATE: Build commercial evidence
Validation asks a harder question: Will the market actually do something?
Founders test propositions, pricing, adoption assumptions, routes to market and willingness to commit. In deep tech, commercial evidence might not yet mean significant revenue.
A letter of intent, access to a production environment, a development partner or a paid feasibility study may represent substantial commercial progress. The important thing is to distinguish genuine commitment from general interest.
PROVE: Make the whole proposition work
Technical proof matters enormously in deep tech. But PROVE deliberately asks a broader question: Can this work technically, commercially and operationally in the real world?
This may involve prototypes, pilots, trials, technical demonstrations, regulatory work, manufacturing development, TEA, LCA or implementation inside a customer's environment. The objective is not simply to increase TRL but to reduce the important uncertainties standing between the technology and the market.
ENTER: Turn evidence into traction
Once the proposition has survived real-world testing, founders need to turn early evidence into a route to market. That means sharpening positioning, establishing pricing, developing partnerships and channels, understanding procurement and building a sales process.
For many deep-tech companies, one of the most important transitions happens here: pilot → paid customer → repeatable customer
The company is no longer proving only that somebody might adopt the technology. It is learning how the market buys it.
GROW: Build for repeatability
Finally, the company needs to make what works repeatable. That may involve commercial systems, manufacturing capability, supply chains, partnerships, new facilities, additional funding and expansion into further markets.
For platform technologies, GROW may eventually lead back to MAP as the company evaluates its next application or market.
Commercialisation is a process of reducing uncertainty and in turn, risk
Perhaps the biggest mistake deep-tech founders can make is assuming that technical development automatically creates commercial progress.
It doesn't.
A company can reach a high TRL while still having major unanswered questions about the customer, market, economics, regulation, manufacturing or route to adoption. Equally, a technology at an earlier technical stage can have unusually strong commercial evidence if customers, partners and investors are already committing resources to its development.
That is why we treat commercialisation as a series of decisions under uncertainty.
At every stage we ask:
DECIDE → TEST → LEARN → COMMIT
What decision are we making?
What would need to be true for this to work?
What don't we know?
What is the quickest credible way to find out?
What does the evidence tell us?
And are we ready to make the next commitment?
For deep tech, this matters because those commitments can be substantial: another year of R&D, a clinical or regulatory programme, a new manufacturing process, a major grant application or millions of pounds of investment. The purpose of commercialisation isn't to eliminate uncertainty. It is to generate enough credible evidence to know which risks are worth taking next.
And ultimately, that is how a promising technology becomes a product, a market and a sustainable business.
