I-Corps model

Nguyễn Đặng Tuấn Minh

Alongside basic research—which any scientific field must invest in and value—there is another layer of research that focuses on the end user right from the start of a project. This layer requires a very different approach and implementation.

Assoc. Prof. Trương Thu Hương from the Faculty of Communication Engineering, Hanoi University of Science and Technology, together with her students, is researching solutions to integrate extended reality (XR) technology into classrooms.

Market-oriented research does not mean only researching what can be sold immediately. Every scientific field also needs long-term basic research, the applications of which may only become apparent ten or twenty years later. But alongside that, we need another layer: research designed from the outset with the end user in mind. This layer is where products, solutions, and science-technology enterprises are created, and it is also where social trust is built, showing that science does not stand still.

This article proposes a practical approach: (1) understand the role of each actor in the innovation ecosystem; (2) learn from the I-Corps model—a famous U.S. program that “pulls scientists out of the lab”; and (3) follow specific steps in market-oriented research, including risk-mitigation mechanisms that Vietnam can implement immediately.

Start by changing the initial research question

Traditionally, research teams begin with a scientific hypothesis: a new technology, material, or process. The goal of the project is to prove that this technology works, is stable, and is better than existing solutions. This approach is essential for advancing knowledge. But if the goal is to go to market, the initial question should be reframed: “Who actually has this problem out there, how are they currently addressing it, and are they satisfied with the current solutions?”

It sounds simple, but many groups skip this step. We often speculate on users’ needs instead of asking them directly. Market-oriented research requires doing what may feel counterintuitive in academia: leaving the lab, talking to potential users or buyers, and letting them describe the problem in their own words. If they do not see it as a “pressing problem,” it is unlikely to be a good starting point for commercialization—even if it is scientifically interesting.

I-Corps: The bridge forcing scientists to meet the market

In the U.S., the NSF’s I-Corps program was created to answer a straightforward question: why do so many promising technologies in labs not become products or companies? The answer turned out not to be weak technology, but that scientists had never been taught how to listen to the market.

I-Corps does not teach technical skills. I-Corps teaches dialogue.

In this program, a research team (usually a principal scientist, an entrepreneurial lead, and a mentor experienced in commercialization) must:

– Interview dozens or even hundreds of potential customers in just a few weeks.
– Describe their technology in terms understandable to end users rather than in academic language.
– Record honest feedback, including “we don’t need this.”
– Adjust hypotheses: if customers are not interested in Application A but respond positively to Application B, then Application B should be taken seriously.

The funding mechanism is particularly interesting. I-Corps does not say, “Just keep researching and we’ll fund your product later.” Instead, further funding is provided only after the team demonstrates that they have found a real market problem with people willing to pay to solve it. In other words, funding is based on evidence of demand, not just scientific novelty.

Where does this reduce risk?

– For the government: funds are not spent on directions with unclear social or economic outcomes.
– For scientists: they do not need to start a company prematurely just to “prove seriousness.” They are allowed to learn and pivot first.
– For private investors later: projects that have gone through I-Corps enter fundraising with initial market data, not just scientific “belief.”

In short, I-Corps does one fundamental thing: it forces scientists to talk to end users before receiving more funding.

Vietnam can learn from this quickly. There is no need to copy the entire U.S. model, but the idea of “funding after validation” can be tested at the ministry, university, or provincial level. Done correctly, it is a new kind of risk-reduction mechanism: risks don’t disappear, but they become visible early.

Typical example from I-Corps in the U.S.:

A biomedical research team thought their new technology would save hospitals time in diagnosis. When they interviewed hospitals, they discovered something no lab scientist had considered: hospitals were not most concerned about speed—they were worried about legal liability if a diagnosis was wrong. The value the researchers thought mattered (faster results) was not what hospitals were willing to pay for (safer results, legally defensible). This insight alone was enough to force the team to adjust priorities.

This is the first step in market-oriented research: instead of proving that a technology has general value, prove that it solves a specific pain point for a specific group in a specific context. To do this, you must listen in the user’s language, not your own.

Be honest about market size and conditions

Many research ideas sound technically plausible, but when put into real economic and social contexts, they become difficult to implement. A solution for treating agricultural wastewater may be excellent in the lab, but if a small shrimp farmer must invest more than the profit of one harvest, no one will adopt it; or if permits are more complex than the farmer can manage, the solution remains on paper—not because it is bad, but because it does not match current cost or regulatory structures.

Market-oriented research requires sometimes blunt honesty: not every problem is a commercial problem. Some serious social problems may not yet have a viable market model. These are still worth pursuing, but should be recognized as public service tasks, not forced to be “immediately sellable.” Conversely, for problems with real revenue potential (e.g., traceability of export agricultural products, energy loss reduction in factories, remote health sensors in understaffed hospitals), analyzing market size, legal regulations, and current competitors is crucial to avoid dead-ends.

  B-life is a technology developed by Assoc. Prof. Lê Thanh Hà at the University of Engineering and Technology, Vietnam National University, designed to help people who are fully paralyzed and unable to speak communicate with others.

 Before proceeding further, it is crucial to answer: “If this product exists, who will pay, from which budget, and by what mechanism?” Failure to answer these questions is a warning sign that the research direction may need adjustment—not a signal to abandon the project. Adjusting research at this stage is far less costly than doing so after three years of development.

 Bringing Market Analysis Back to the Research Table

 This is often the most uncomfortable step for scientists, because it challenges ingrained thinking habits. In traditional academic culture, “changing direction midstream” is often seen as a sign of immaturity. Market-oriented thinking, however, treats adjusting research based on real-world data as normal—even healthy.

 

In the U.S., I-Corps teams often experience this moment: they start confident, claiming “We are developing solution X for market Y,” and finish with a completely different statement: “Market Y is not interested, but segment Z urgently needs it and is willing to pay; we will pivot solution X toward Z first.” Traditional thinking might label this “off-topic,” but market-oriented thinking sees it as optimizing the commercialization path: finding a real outlet for knowledge.

At this stage, two seemingly “non-scientific” factors become critical: intellectual property (IP) and the value model.

Intellectual Property (IP): IP does not always mean “file a patent as soon as possible.” Rushing to file can lock the team into a particular implementation before understanding the suitable applications. Conversely, sharing too freely with partners without protection can cause conflicts later when the technology gains commercial value. Institutions like UCL and Cambridge work early with scientists to map out feasible scenarios: licensing, spin-off companies, or joint industrial development—clarifying rights, benefit sharing, and responsibilities upfront to build trust.

Value Model: Though it sounds business-oriented, this is crucial for applied science survival: what value does your solution create for users—cost savings, increased output, legal risk reduction, new export markets, diagnostic accuracy, or labor reduction? Different types of value lead to different pricing strategies. Understanding this informs how you communicate your product—not as PR, but as a bridge between scientific language and buyers’ language.

From Prototype to Commercial Product

Once three layers are clear—real demand, real market conditions, and real value—it is time to develop a prototype.

A prototype (a functional, though imperfect version) is not for presentation—it is for testing with potential users. Its goal is not “does it work,” but “will users actually use it?” Scientists, engineers, and business-minded team members must collaborate. Users’ feedback—like “I don’t have time for five steps” from a nurse, or “I’m too busy to input data daily” from a farmer—becomes essential design input.

Moving from prototype to commercial product can create illusions. Many think “if the technology is good, a company will take over and sell it.” Even at top universities like UCLB in the UK, it can take nearly ten years of patient investment before major spin-outs emerge. Universities must maintain technology transfer offices during this period. Commercialization is never simply “give it to a company”; it is a collaborative process.

In Vietnam, a common challenge is “who will accompany us?” Honest answers: universities cannot just push research out; companies cannot demand perfect products immediately; regulators cannot require instant economic efficiency. If all three parties do not work together, products often fail—not due to technology quality, but due to lack of patient collaboration.

A small but meaningful change is inviting scientists to initial industry discussions as experts or co-designers, with legal and institutional support from the university. This allows scientists to step out safely, building natural, durable connections between labs and market.

Market-Oriented Research: Seven Steps as an Operational Loop

Market-oriented research is not simply “ask customers what they want and deliver it.” The seven steps should be understood as an iterative loop—each step is both a technical activity and a risk control point.

Step 1. Identify market needs
Start with “Where is there a real, unsolved problem?” rather than “I have technology X.” Observe, interview, and document in users’ own words, not technical translations. This is the first risk-reduction mechanism: if the problem cannot be described in users’ language, it indicates incomplete understanding.

Step 2. Market and competitive analysis
Assess market size, existing solutions, legal barriers, and international standards. Many projects fail commercially because this step is skipped. This step reduces financial and legal risk.

Step 3. Adjust R&D direction based on market data
Prioritize research branches that address significant demand; defer others. This reduces opportunity risk.

Step 4. Protect Intellectual Property (IP)
Establish ownership, enable transparent negotiations with partners, and lay foundations for licensing, revenue sharing, or spin-offs. This reduces relational risk.

Step 5. Develop prototype / Minimum Viable Product (MVP) and validate
Test with users early. Scientists, engineers, and business-minded team members collaborate. This reduces technology and user adoption risks.

Step 6. Build commercialization strategy
Decide on licensing, spin-offs, or joint development. Develop go-to-market plan, pricing, distribution, and fundraising. This reduces execution risk.

Step 7. Market launch and scale-up
Scientists remain involved in early post-launch stages, ensuring scientific and technical accuracy.

Four Key Risk-Reduction Mechanisms in Mature Systems

1. Perceptual risk for scientists: I-Corps-style programs allow scientists to learn market language without immediate startup pressure.
2. Financial risk for government: Funding after initial demand validation ensures public resources are not wasted.
3. Legal and IP risk: Clear IP rights prevent disputes among scientists, universities, and industry.
4. Social expectation risk: Communicate that commercialization is a multi-year process; early engagement and co-design are socially valuable steps.

References:

• No Science, No Startups: The Innovation Engine We’re Switching Off
• From Lab to Market: The Power of Research Commercialization
• Durolabs: Commercialization
• AMA Marketing News: Market Analysis Guide
• Competitive Intelligence Alliance: Market Research Guide
• Stryber: R&D Commercialization Success Story
• Pragmatic Institute: 7 Keys to Market-Driven Organization
• SBA Market Research & Competitive Analysis
• SmartBug: 5-Step Marketing Research Process
• Drive Research: Market Research Guide

Source: Tia Sáng, Issue 22/2025