‘Deep tech’ has been in the headlines in Indian news media lately, in the backdrop of Minister Piyush Goyal’s address at the recently concluded Startup Mahakumbh. While we have so far celebrated unicorns which are predominantly consumer tech start-ups that develop apps and algorithms that we have in our devices, deep tech is based on scientific and technological breakthroughs. What exactly makes a deep tech start-up different, and what kind of distinct ecosystem must India cultivate to respond to the minister’s call?

In simple terms, deep tech start-ups are ventures that commercialise products or services based on foundational R&D-based scientific or technological breakthroughs. A key characteristic that differentiates deep tech from consumer tech, is that in addition to market risk, there is the additional technology risk associated with foundational R&D, because of technology uncertainty, longer gestation period to reach the market, and higher capital investment cost.

Let’s look at an example to understand this. One of the earliest indigenously developed and successfully commercialised deep-tech products in India was a low-cost heart valve. This valve was a Class III medical device, a high technology risk product, as it was meant to be implanted into the human body to sustain or support life. This foundational R&D for the valve was done by Sree Chitra Tirunal Institute of Medical Sciences Trivandrum (SCTIMST), a government of India-supported research institute and hospital. For a then non-existent biomedical innovation ecosystem, it took 17 years from the initial seed thought in 1978 to its eventual commercialisation in 1995.

This story of a deep tech innovation ecosystem in India around biomedical devices, documented in a recent Journal of Product Innovation Management article, identified that the development of a working prototype of the heart valve involved experimentation with different material compositions, engaging in unrelated technology partnerships such as space tech, and creating governance mechanisms and guardrails for the technology.

Developing a prototype that passed animal safety trials took the team 10 years and four distinct final versions before human testing could even begin. Then, when moving to human trials, the valve had to be moved from a lab setting to industrial-scale production, and multiple hospitals needed to be willing to participate in the human trials. On both these counts, there was significant apprehension due to high technology risk, due to adverse reputational impact on the industrial partners and hospitals if the technology fails, which in this case will cost a human life.

Thus, when it came to commercialising the technology, SCTIMST had to put their own brand at reputation by calling it ‘Chitra heart valve’. Then, along with an industry partner — who happened to be from an unrelated industry like pressure cooker, as there was no other biomedical innovation ecosystem partner — they had to co-create a technology proving facility, where the heart valve production was translated from lab to industrial scale. Furthermore, a lead scientist had to move from SCTIMST with the technology to the commercialisation partner, in the process of creating a new start-up that then commercialised heart valve.

Multiple lessons

This example offers multiple pertinent lessons for envisioning a deep tech ecosystem. First, need for patient capital, that is willing to take risks while being at ease with the long development cycles. This needs dedicated government capital working alongside philanthropy capital from family foundations targeting deep tech research centres or early-stage ventures with matched co-investment from the industry.

Second, more seamless engagement between academia and industry, wherein there is flexibility in the form of sabbaticals offered to work on commercialising a technology developed in a lab on a mission mode working in partnership with an industry partner. This could also take the form of a university or research institute start-up spin-off with tripartite stakes between the university, industry, and core academic group developing the technology.

Third, create an incentive structure within academia that pushes to go beyond journal publishing and patenting to focus on real-world translation of the research. This will also need reimagining time allocation and workload models in Indian universities, which currently have a higher prioritisation on teaching.

Finally, we could learn from policies of ASEAN countries like Singapore’s SG Tech programme, to develop stage-gated competitive research grants for proof-of-concept and proof-of-value projects. To complement this, drawing on Korea’s example of the Deep Tech Incubator Program for Startup, MeitY Startup Hub could first outline priority deep tech fields of high potential where India has strong expertise such as biomedical, space tech, and clean energy to offer targeted mentoring on both technological and market aspects.

Ultimately, fostering deep tech start-ups isn’t about scaling the existing start-up models, but about mastering a different, more patient, and specialised ecosystem creation that nurtures technological innovations from lab to market.

The writer is Professor, University of Glasgow Adam Smith Business School

Published on April 16, 2025