While entrepreneurs and investors are well versed in the Lean Startup model for developing tech companies, the relevance of this approach to business building diminishes in the realm of deep tech. The reasons for this are varied and formidable.

Often deep tech opportunities emerge from the laboratories of our leading research institutions. The academic incentive to create new knowledge rarely matches the ambitions of businesses to solve problems. Early stage investors and entrepreneurs can often find themselves trying to identify use cases and applications for a new technology – essentially a solution looking for a problem to solve. This is at odds with the normal startup mode in which the business has been created explicitly to solve an identified problem. Cue head scratching and an extensive deep dive to find a home for such new solutions.

Why it’s hard for deep tech

Having identified a problem to solve, deep tech solutions are typically capital intensive, requiring far more investment to develop through to proof of concept, prototype and product stages than other sectors. A lean software startup, for example, requires no more than a few laptops equipped with the right software and an internet connection. Laptops represent the drop in the ocean in the deep tech world. More often than not laboratory space, expensive infrastructure and equipment along with layers of regulation to navigate are the demands placed on nascent deep tech companies. The cost of approvals alone on new therapeutics and/or medical devices run into the tens of millions. Deep tech companies then encounter risks just getting an idea to work before even thinking about whether or not they can produce it at a competitive price. Non deep tech startups rarely encounter these risks, often focusing exclusively on product, market and growth challenges.

Death Valley

For the fortunate few deep tech startups able to navigate the valley of death between solution and product stages, a swathe of new challenges await. Talent is the most pressing concern, since the skills required to develop products and solutions are niche, rare and in high demand. Early stage deep tech companies encounter stifling bottlenecks when trying to hire the right people, who are often weighing up far more lucrative and secure opportunities in the corporate world. At the same time, academic founders have to balance an array of incentives that can often lead to them not being able to fully commit to a startup. Building a deep tech company from scratch requires all of your time and then some. Balancing this with an active role researching and teaching rarely delivers the desired outcomes.

Towards a solution

All of the above requires investment. This is a challenge given that the pool of investors open to the above risk profile is relatively small in comparison to funds scouring the early stage ecosystem for the next software-based unicorn. The difficulty in agreeing valuations for deep tech companies, with no established customer or revenue base, compounds the challenge of raising investment.

As interest in deep tech continues to build, there is a reasonable case to be made for adapting our collective approach to getting deep tech businesses off the ground. Technology pull is often spoken about as a means of overcoming the solution-looking-for-a-problem conundrum. A lot is written about technology pull, but the forums whereby industry, investors and researchers can rigorously interrogate the vulnerabilities to disruption and better understand R&D pathways, seem few and far between.

Building minimum viable products and iterating growth thereafter works extremely effectively when building software companies. But these should be complimented in deep tech by a process of continuous de-risking whereby the inherent product, idea and people risks are managed and mitigated throughout the process. A pragmatic approach to formulating valuations, that, in the absence of cash flows, moves beyond traditional financial models would help pave the way for an agreed upon approach to valuing and financing deep tech.

Collaboration

From an infrastructure perspective, mechanisms whereby lab space and equipment could be pooled and shared would go a long way to de-risking these sorts of opportunities. For instance, universities currently face the challenge of making lab time available for commercial opportunities given their tax status; tweaking this regulation would be helpful. At the same time, if corporates were able to open their doors for prospective deep tech companies to undertake proof of concept and prototyping that would be another step towards de-risking and uncomplicating deep tech opportunities. A more collaborative approach to talent would make sense whereby greater training, incentives and awareness makes building deep tech companies a more attractive option for the right people.